探究全民英檢中高級寫作測驗高分與低分作文之主題發展與連貫性

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. 探究全民英檢中高級寫作測驗高分與低分作文之主題發展與連貫性. Thematic Progression and Cohesion in Higher-rated and Lower-rated Essays in High-intermediate GEPT Writing Test. 指導教授：張珮青. 博士. Advisor: Dr. Pei-Chin Chang 研究生：吳俊. 穎. 中華民國一百零五年二月 February 2016.

(2) 摘要在寫作上使用主位推進的策略可有效地加強文章的連貫性，而分析文章的主位推進亦能清晰呈現文意的流動與訊息結構。然而，就此領域而言，分析台灣學生作文中主位推進的文獻相對較少，文獻上探討高分與低分文章中主位推進之差異的研究亦十分有限。因此，本研究旨在探討台灣高分與低分的英文作文中，主位推進法質與量方面的差異。研究方法為文章分析法，資料取自財團法人語言訓練測驗中心，以量化與質性並重的方式，分析 106 篇全民英檢中高級測驗中的論說文，並探討其主位推進的狀態與語意的銜接性。本研究結果如下：第一，不同的主位推進法就頻率而言，並未在高分與低分作文中有明顯差異；超過 40%以上的文句皆未透過主位推進法連接，而頻率最高的主位推進法為重複推進法，其次為線性推進法。第二，就主位推進的質而言，低分作文中的主位推進法常無法獲得有效運用。分數較低的考生常重複同一種主位推進法，其使用的銜接方式亦較高分考生單調。此外，低分考生經常中斷文句之間的連結，以致語意無法連貫。最後，本研究文章分析的結果，希望能解釋出台灣學生在使用主位推進法上的共通的習慣與困難。英語教師也可以藉此了解到學生面臨的困境，並以此為基礎設計相關課程，協助學生增進作文上的銜接性。關鍵字：主位推進法、主題結構、論說文、銜接性、全民英檢. i.

(3) ABSTRACT Deploying thematic progressions (TP) has been considered to be a valuable technique to achieve textual cohesion in writing, and TPs across clauses have often been assessed to reveal information flow over a longer stretch of text. However, research which empirically investigated TP strategies employed by Taiwanese EFL learners is scant, and little literature has been published on the discrepancy of TP between higher-level and lower-level writers in Taiwan. The purpose of this study, therefore, is to explore TP in higher-rated and lower-rated expository writings written by Taiwanese EFL learners. The corpus consisted of 106 expository essays from the Guided Writing section of high-intermediate level General English Proficiency Test (GEPT). Halliday and Matthiessen’s (2013) framework of Theme system and Danse’ (1974) categorization on TP were adopted to both quantitatively and qualitatively analyze the TPs in the essays. The proportion of TP types, marked Themes, and how TP contribute to cohesion qualitatively were investigated. The results of the study showed that the proportion of TP types were similar between higher-rated and low-rated essays. Breaks, links that did not fall into any TP types, constituted more than 40% of total links, followed by constant TPs and linear TP. The large number of breaks prohibited the ideas from developing in lower-rated essays. The qualitative results also revealed that lower-level writers employed repetitive patterns of constant TPs in their text, and they often used existential “There” to start a clause. Theme with unclear reference and empty Rheme were also distinctive features of their essays. On the other hand, Marked Themes appeared more than unmarked Themes in both groups of essays, but the use of marked Themes was limited mostly to conjunctions or conjunctive adjuncts, which indicated that the writers failed to effectively use a comprehensive range of marked Theme. ii.

(4) To conclude, this study has importance in informing writing instructors of common problems in TP used in essays with unsatisfactory information structure, as well as in proposing a more scientific and theory-informed approach to instruct learners in effective deployment of TP. Keywords: Thematic Progression (TP), thematic structure, expository writing, cohesion, GEPT. iii.

(5) ACKNOWLEDGEMENTS This thesis would have remained a dream had it not been for the guidance and support of many people who helped me through this great challenge. I wish to thank, first and foremost, my advisor, Dr. Pei-Chin Chang, who had not only taught me the essence of academic writing but enlightened me through the tough process of logical reasoning and argumentation. It is with immense gratitude that I acknowledge her continuous support and help over the two-year span of my thesis writing process. Her advice on my arguments and structure have always enabled me to gain a better insight into this field and to come up with a better way to illustrate my ideas. This thesis would not have been possible without her expertise and selfless guidance. Secondly, it gives me great pleasure in acknowledging the support and help of my committee members, Dr. Yuh-Show Cheng and Dr. Hsueh-ying Yu. Their generous and invaluable comments substantially improved the logic of my argumentation and helped my drafts take better shape. I owe my gratitude to The Language Training & Testing Center (LTTC) of Taiwan for providing me precious data of expository writings from a highintermediate General English Proficiency Test (GEPT). They have assisted me in obtaining all the data I required to complete my research. I wish to thank National Taiwan Normal University for providing me rich resources and the Department of English in specific for giving graduate students a clean and quiet space to conduct their research. I am also indebted to the professors and classmates in NTNU TESOL program for offering me all the help I need throughout the process. Finally, I must express my very profound gratitude to my parents, close friends, iv.

(6) and my colleagues who accompanied me during the exhausting process. Their encouragement and comfort from time to time helped me survive the hard times over these years.. v.

(7) TABLE OF CONTENTS CHINESE ABSTRACT ......................................................................................................i ENGLISH ABSTRACT .................................................................................................... ii ACKNOWLEDGEMENTS ..............................................................................................iv LIST OF TABLES .............................................................................................................ix LIST OF FIGURES ............................................................................................................x. CHAPTER 1 INTRODUCTION .......................................................................................1 Background and Motivation ......................................................................................1 Purpose of the study ...................................................................................................1 Significance of the Study ...........................................................................................3 CHAPTER 2 LITERATURE REVIEW ............................................................................5 Systemic Functional Linguistics (SFL) ....................................................................5 Metafunction ......................................................................................................6 Register ...............................................................................................................7 Genre...................................................................................................................8 Theme and Rheme .....................................................................................................9 Definition of Theme and Rheme.....................................................................10 The Realization of Theme ............................................................................... 11 Thematic Progression (TP)..............................................................................15 Empirical Research on TP .......................................................................................20 TP in Writings Written by Non-native Writers...............................................21 TP and Genre ....................................................................................................23 TP in Higher-rated and Lower-rated Essays ..................................................25 Marked Theme in TP .......................................................................................26 vi.

(8) Instruction in TP...............................................................................................29 Research Questions ..................................................................................................30 CHAPTER 3 METHODOLOGY ....................................................................................32 The Corpus ...............................................................................................................32 The Writing Task ......................................................................................................34 Approaches to Analysis ...........................................................................................35 Quantitative Analysis of the Proportion of TP Types ....................................38 Qualitative Analysis of the Deployment of TP ..............................................41 CHAPTER 4 : RESULTS AND DISCUSSION .............................................................42 Proportion of TP Types ............................................................................................42 Overall Proportion of TP Types ......................................................................42 Proportion of TP Types in Level 3 and Level 4 Essays .................................45 The Variety of TP Types in Level 3 and Level 4 Essays ...............................49 Marked Theme in level 3 and level 4 essays ..........................................................53 The Analysis of TP from a Qualitative Perspective ...............................................55 Repetitive pattern of constant TP ....................................................................56 The use of Existential “There” ........................................................................59 Theme with an unclear reference ....................................................................61 The Empty Rheme ...........................................................................................64 The use of marked Theme ...............................................................................65 Breaks in the Essays ........................................................................................69 CHAPTER 5 CONCLUSION .........................................................................................73 Summary of Major Findings ...................................................................................73 Pedagogical Implications .........................................................................................74 Limitations of this Study .........................................................................................76 vii.

(9) Suggestions for Future Research .............................................................................77 REFERENCE ...................................................................................................................79. viii.

(10) LIST OF TABLES Table 1: Mood Type and Unmarked Theme Selection (adapted from Halliday & Matthiessen, 2013, p. 104) ................................................................................13 Table 2: Overview of the Corpus .....................................................................................34 Table 3: Procedure in Analyzing TP (adapted from Crompton, 2004, p. 235) .............40 Table 4: Proportion of TP Types in all essays ................................................................43 Table 5: The Difference Between Higher-rated and Lower-rated Essays.....................46 Table 6: The Variety of TP Types in Each Essay ............................................................50 Table 7: Proportion of Breaks in all essays ....................................................................52 Table 8: The Proportion of Marked Themes and Unmarked Themes in Higher-rated and Lower-rated Essays ....................................................................................54 Table 9: A long sequence of constant TP in lower-rated essays ....................................56 Table 10: The use of constant TP in higher-rated essays...............................................58 Table 11: The overuse of “There” to start a clause in lower-rated essays ...................60 Table 12: The overuse of “There” to start a clause in lower-rated essays revised......61 Table 13: Theme with an unclear reference in lower-rated essays ...............................62 Table 14: Theme with an unclear reference in higher-rated essays ..............................63 Table 15: An example of empty Rheme in lower-rated essays (b) .................................65 Table 16: An example of overusing textual Themes as marked Themes .......................67 Table 17: A Solution to overusing textual Themes as marked Themes .........................68 Table 18: An excessive number of breaks in Lower-rated Essays .................................69 Table 19 An excessive number of breaks in Higher-rated Essays .................................71. ix.

(11) LIST OF FIGURES Figure 1: Genre as an additional stratum of analysis beyond tenor, field and mode (Martin & Rose, 2007, p. 17)............................................................................9 Figure 2: TP with a Constant Theme (Danes, 1974, p. 118) ..........................................16 Figure 3: Simple linear TP (Danes, 1974, p.118) ...........................................................17 Figure 4: TP with a Derived Theme (Danes, 1974)........................................................17 Figure 5: Exposition of Split Rheme (Danes, 1974) ......................................................18 Figure 6: Rhematic Progression.......................................................................................19 Figure 7:Proportion of TP Types in all essays ................................................................44 Figure 8: The Difference of the Proportion of TP Types Between Higher-rated and Lower-rated Essays .........................................................................................47 Figure 9: The Difference of the Number of Writers Using Each TP Types Between Higher-rated and Lower-rated Essays ............................................................48 Figure 10: The Proportion of Marked Themes and Unmarked Themes in Higher-rated and Lower-rated Essays ..................................................................................55. x.

(12) CHAPTER 1 INTRODUCTION Background and Motivation In recent years, an increasing number of studies have explored the English writing ability of Taiwanese EFL learners and revealed numerous challenges that Taiwanese students have been confronted with (Chao, 2002; Chen, 2002; Liou, 2008; Shen, 2004; Tsai, 2012; Yang, 2003). These challenges may partially result from the scarcity of writing training in school environment and the lack of motivation to write in English. Furthermore, writing is difficult for Taiwanese learners because several essential skills are required in the process of writing, one of which is organizing information structure (Chen, 2002; Shen, 2004; Yang, 2003). Yang (2003) indicated that Taiwanese high school students frequently fail to organize clauses in a composition; Chao (2002) and Shen (2004) also suggested that high school students fail to attend to connection between clauses, which results in unsatisfactory information structure. When learners compose English writings, maintaining the connection at clause level and arranging old and new information are often difficult to them (Mellos, 2011; Moore, 2006; Wang, 2007; Witte & Faigley, 1981). The inability to organize information might result in inferior cohesion and prohibit readers from understanding the text. Thus, the above reasons reveal the need to explore the area of cohesion in Taiwanese EFL students’ writing of English. Purpose of the study The purpose of this study is to investigate how Taiwanese learners achieve cohesion in their writings. It compared the difference of cohesion in higher-rated and lower-rated writings in order to deepen our understanding on the features of cohesion in higher level writings and display the weakness of cohesion in lower-level writings written by Taiwanese EFL learners. Cohesion is defined as the relation of meanings 1.

(13) that exists in a text (Halliday & Hasan, 1976), and it “occurs where the interpretation of some element in the discourse is dependent on that of another” (Halliday & Hasan, 1976, p. 4). Cohesion concerns the quality of text since developing information cohesively increases the readability of the text. A passage lacking cohesion cannot become a text since the relation of meanings fails to be established. Thus, the cohesion of students’ composition is an area worth exploring and improving in that cohesion deals with the information structure at global level, making it a crucial element of a successful writing. In addition, the present study particularly explored how cohesion was achieved in expository essays. In Chinese EFL context, the skill to compose expository essays has been a crucial skill when students advance their education or apply for job positions (Xu, 2000; Yang, Ramírez & Harman, 2011). This skill is also assessed in major English examinations such as college entrance exams and public English proficiency tests in Taiwan. For genre theorists, teaching explicitly the specific features of a genre may benefit English learners in ESL or EFL context. Therefore, expository essays written by Taiwanese learners were studied quantitatively and qualitatively, and how cohesion was constructed in successful expository writings was revealed. In sum, this study seeks to propose some explicit suggestions to guide the teaching and evaluation of cohesion in expository essays. Those suggestions might hopefully offer a more comprehensive and scientific way to treat Taiwanese English learners’ cohesion, and the teaching and learning of cohesion may also become more consistent. The information can become the reference for students to consider when organizing their ideas in writings; English writing instructors can also refer to it when evaluating students’ information structure or designing writing classes. 2.

(14) Significance of the Study First of all, only few researchers (Chao, 2002; Shen, 2004; Tsai, 2012) have tapped into the information structure in Taiwanese EFL learners’ essays, so a comprehensive understanding of Taiwanese English learners’ cohesion in various genres is still critically lacking, and literature comparing students of higher and lower level is little. It would thus be worth investigating the features of cohesion in expository or argumentative essays in Taiwanese EFL context (Tsai, 2012). If general patterns of how information cohesively developed in expository writing can be proposed, the patterns can help English instructors deepen their understanding in Taiwanese writers’ issues in cohesion. Secondly, in Taiwanese EFL context, the training of writing starts after students enter senior high school, where teachers train students to compose short narrative essays and expository essays. However, as in many EFL contexts, the teaching and learning of writing are limited to local features such as grammar points or sentence structure. The global feature in learners’ essays such as cohesion is often unaddressed, and students have little knowledge of how to develop their ideas cohesively. Thus, if explicit suggestions and common errors of cohesion in Taiwanese writers’ essays are proposed, teachers can depend on these information to design an effective and efficient writing class on cohesion. Thirdly, teachers often lack a tool to assess cohesion in students’ texts (Belmonte & McCabe, 1998). While few studies have investigated the problems of cohesion in ESL or EFL learners’ writing, suggestions for evaluating students’ cohesion in writing are still critically lacking. Numerous researchers investigated the cohesion of students’ essays (e.g., Ebrahimi & Ebrahimi, 2012a; Green, Christopher, & Mei, 2000; Hawes & Thomas,1997; Hewings, 2004; Li, 2009; McCabe, 1999; Wang, 2007), but 3.

(15) few have proposed more comprehensive and explicit suggestions to treat cohesion in expository writings. Therefore, the results of the study may benefit teachers or examiners that they can identify and treat students’ cohesion problems based on these suggestions. In sum, the findings of the study are expected to contribute to learning and teaching of cohesion in expository writings and also the evaluation of cohesion.. 4.

(16) CHAPTER 2 LITERATURE REVIEW Relevant literature and research questions are presented in this chapter, and it is structured as follows. The first section of this chapter deals with the theoretical framework of the present study, including the theory of Systemic Functional Linguistics (SFL) and SFL concepts such as metafunction, register, and genre. In the second section, details and discussion on Theme and Rheme, Theme types, and thematic progression (TP) are provided. The third section presents empirical research on TP, which explores TP from different approaches and in various disciplines; it also contains studies on instruction in TP. Research questions of this study are then proposed in the last section of this chapter. Systemic Functional Linguistics (SFL) The aim of this study is the cohesion in EFL learner’ writings, namely, the relation of meanings in their texts. The focus of this study is, therefore, on the textual meaning, which is concerning how learners organize their information in their texts. The concept of textual meaning originated from Systemic Functional Linguistics (SFL), which was developed by Michael Halliday and his colleagues from the Prague School. Systemic Functional Linguistics (SFL) is a functional linguistic approach that targets how language is used by people and how it is structured for use (Eggins, 2004). It is named systemic since “compared with other theories it foregrounds the organization of language as options for meaning” (Martin & Rose, 2007, p. 21). In this light, relationships of choice between elements of a language is the key. It is termed functional because “it interprets the design of language with respect to ways people use it to live” (Martin & Rose, 2007, p.24). To scholars of SFL, therefore, language is closely related to our social environment where people communicate in a 5.

(17) language to make sense of their experience and enact their social relationship (Halliday & Hasan, 1976; Halliday & Matthiessen, 2013). That is, speakers of a language make choices of language (what you say) so as to reach communication in various social contexts and for particular social purposes. Metafunction To realize the goal of communication, SFL scholars proposed three metafunction components that exist in a language system, including the ideational component, interpersonal component, and textual component: 1.. Ideational component: According to Halliday and Hasan (1976), the first component, ideational component, deals with the content of the language, construing human experience with lexicogrammar. It includes “what’s going on, who’s doing what to whom, where, when, why and how and the logical relation of one going-on to another” (Martin & Rose, 2007, p. 24). It can be further categorized into experiential ideational metafunction, which deals with experience and culture, and logical ideational metafunction, which presents logical relations indirectly resulted from experience.. 2.. Interpersonal component: The second component is interpersonal component, which is concerned with the attitudes, motives, and judgments of speakers, and how they encode the relationships in certain situations (Halliday & Hasan, 1976). Namely, when people communicate in a language, they have to act out their social relationship with one another, and negotiating social relations concerns “how people are interacting, including the feelings they try and share” (Martin & Rose, 2007, p. 24).. 3.. Textual component: The third component is the focus of this study, the textual component, which indicates how language is structured as a piece of speech or 6.

(18) writing. Textual component “build[s] up sequences of discourse, organizing the discursive ﬂow and creating cohesion and continuity” (Halliday & Matthiessen, 2013, p. 31). It is related to “information flow: the ways in which ideational and interpersonal meanings are distributed in waves of semiosis, including interconnections among waves and between language and attendant modalities (action, image, music etc.)” (Martin & Rose, 2007, p. 24). Register Aside from the metafunctions of language, Halliday and Hasan (1976) considered that social context was also the factor that determined which linguistics features to appear. They termed the social context the “register” of language, constituted by “the linguistic features which are typically associated with a configuration of situational features - with particular values of the field, mode and tenor” (Halliday & Hasan, 1976, p. 22). The “field” of discourse concerns the social action which is happening, what the participants are acting out; the “tenor” of discourse refers to the participants who are engaged in the action; the “mode” of discourse means the nature of language, which indicates “the channel taken by the language - spoken or written, extempore or prepared - and its genre, or rhetorical mode, as narrative, didactic, persuasive, 'phatic communion' and so on” (Halliday & Hasan, 1976, p. 22). As mentioned above, the textual metafunction combines the resources of language, and mode concerns the situation in which it is realized. However, the context of language involves not only social context and situation, but also the culture factors which influenced how language is actually used, that is, the “genre” of the language.. 7.

(19) Genre As mentioned at the beginning of this section, SFL experts paid attention to the context or situation in which language is being used, and “genre” was the term they used to describe the context of culture. Genre helps carry out the cultural purpose of texts, and it is realized through the structure or the pattern of text (Eggins, 2004). Genre denotes how people reach their culturally proper goals through language. Therefore, for a text to achieve its purpose in a culture, the choice of words or phrases at textual level must be carefully selected so as to abide by the conventions of its genre. If the generic identity of a text cannot be attributed to a certain genre by readers, the text might become problematic (Eggins, 2004). From SFL approach, language consists of three metafunction components (ideational, interpersonal, and textual), and linguistics features are influenced by situational factors of register (field, tenor, and mode). Both register and genre should be considered in order to understand how language is used, while genre is a broader and more abstract concept that determines the feature of language in a specific context of culture. Martin and Rose (2007), therefore, placed genre at a higher stratum which functions independently from register. They believed that field, tenor and mode were resources which could be generalized across genre. The following Figure 1 illustrates the relationship and hierarchy between metafunction of language, register, and genre:. 8.

(20) Figure 1. Genre as an additional stratum of analysis beyond tenor, field and mode (Martin & Rose, 2007, p. 17) SLF linguists targeted the function of language through a system of choices, and they analyzed language in terms of three metafunction components. The textual metafunction component concerns how language is structured as a piece writing, and it was explored in the present study. How information is organized in the writings of Taiwanese EFL learners was investigated. To realize this research purpose, the cohesive development of the language in these writings was analyzed and patterned through an analytical framework developed by SFL theorists, the system of Theme and Rheme. This system is introduced and explained in the following section. Theme and Rheme The notion of Theme and Rheme was proposed by Prague School linguistics. To understand the notion of Theme and Rheme, the unit of meaning in a language system should be introduced beforehand. In SFL, the basic unit of language in use is “text,” which is a semantic unit of meaning instead of grammar, and the Theme system explains the thematic structure of each clause. Halliday and Matthiessen (2013) defined “text” with the following terms: The term ‘text’ refers to any instance of language, in any medium, that makes sense to someone who knows the language; we can characterize text as 9.

(21) language functioning in context. … Language is, in the first instance, a resource for making meaning; so text is a process of making meaning in context. (Halliday & Matthiessen, 2013, p. 3) A text can be easily identified as a unified whole by people using a language. That is, when people encounter a written or spoken passage of language, it is not hard for them to distinguish the text from a random combination of irrelevant clauses (Halliday & Hasan, 1976). The basic element of a text is “clause,” which comprises a text. To SFL theorists, the meaning of a clause is the main focus of analysis, and they proposed three lines of meaning in a clause based on the three metafunctions mentioned above (Halliday & Matthiessen, 2013). The ideational metafunction is manifested by the “Actor” in the clause, which is the participant of the process construing human experience; the interpersonal metafunction is realized by the “Subject,” which shows the exchange of meaning between interlocutors (Halliday & Matthiessen, 2013). As for the textual component, a related system addresses the textual meaning of a text, the system of Theme and Rheme. This system is introduced in the following subsection. Definition of Theme and Rheme Theme is the focus of the clause as a message. Halliday and Matthiessen (2013) defined it as the following: The Theme functions in the structure of the clause as a message. A clause has meaning as a message, a quantum of information; the Theme is the point of departure for the message. It is the element the speaker selects for ‘grounding’ what he is going on to say. (Halliday & Matthiessen, 2013, p. 83). 10.

(22) The function of Theme is to suggest what the clause is going to be about, and Theme also “locates and orients the clause within its context” (Halliday & Matthiessen, 2013, p. 89). Therefore, speakers select a Theme for each clause so as to express his or her organization of the message (Halliday & Hasan, 1976). The part of message selected as Theme becomes prominent to the listener, so the listener can process the message from that starting point (Halliday & Matthiessen, 2013). Theme is usually placed at the clause initial position, and the rest of the clause is termed “Rheme.” According to Halliday and Matthiessen (2013), Rheme is the part of clause where Theme is developed. Each clause is composed of a Theme and a Rheme, and Theme should always proceed Rheme in the order. In this way, the Theme can be the highlight of clause as the starting point of message, which makes it more prominently noticed by listeners. As people tend to depart the message from familiar information to unfamiliar information, Rheme is usually made up of “new” information (Eggins, 2004; Martin & Rose, 2007). This structure of clause as message, which is comprised of Theme and Rheme, is defined as “thematic structure” (Halliday & Matthiessen, 2013). Since whatever is selected as Theme may be prominent, the choice of Theme may greatly influences how the text is processed by its listener or reader. In addition, the analysis of the way Theme and Rheme are patterned in a clause can reveal the “information flow” of text (Eggins, 2004, Martin & Rose, 2007). How Theme is realized in a clause is explained in the following subsection. The Realization of Theme As mentioned above, the choice of Theme is significant in that it contributes to the organization of text. Therefore, linguistic elements which can manifest and function as Theme were presented in this subsection, and the boundary between 11.

(23) Theme and Rheme was defined. In order to identify this boundary, the realization of different Theme types should be understood beforehand. In Halliday and Matthiessen’s (2013) model, Theme is considered as the first group whichever can manifest experiential structure in a clause. In declaratives, the most common category of Theme is a participant, which is presented by a nominal group (e.g., Teenagers) and functions as the Subject in this clause (Martin & Rose, 2003). It is the first constituent of the clause, and any elements that follow this constituent is the Rheme of the clause. Theme can be differentiated by its “markedness.” When something is “marked,” it is usually atypical compared to the norm and is noticed prominently by its listener or reader. In normal cases of declarative clauses, marked Theme is defined as the Theme which is not realized by the grammatical Subject of the clause but something such as an adjunct (e.g., comment adjunct, conjunctive phrase); it might be noticed prominently by listeners or readers (Halliday & Matthiessen, 2013; Martin & Rose, 2003). In clauses where the Theme conflates with the grammatical Subject of a clause, the Theme of the clause is categorized as “unmarked Theme” (Halliday & Matthiessen, 2013). For example, in Table 1, the unmarked Theme is Teenagers:. 12.

(24) Table 1 Mood Type and Unmarked Theme Selection (adapted from Halliday & Matthiessen, 2013, p. 104) Mood of clause. Typical (‘unmarked’) Theme. declarative. nominal group functioning as Subject e.g., Teenagers love to participate in talent shows.. interrogative: yes/no. first word (finite operator) of verbal group plus nominal group functioning as Subject e.g., Should teenagers spend time practicing talents?. interrogative: wh-. nominal group, adverbial group or prepositional phrase functioning as interrogative (Wh-) element e.g., How long do students spend in their clubs?. imperative: ‘you’. verbal group functioning as Predicator, plus preceding don’t if negative e.g., Don’t smoke on the campus.. imperative: ‘you and. let’s plus preceding don’t if negative. me’. e.g., Let’s go to the mall.. Note. The unmarked Theme in each clause is underlined. The markedness of Theme depends on the mood of the clause, namely, Theme element functions differently in declaratives, interrogatives, and imperatives (Eggins, 2004; Halliday & Matthiessen, 2013). In interrogative clauses, the speaker asks a question in order to find the answer. The typical Theme, therefore, should be the “what one wants to know.” In yes/no interrogatives, the unmarked type of Theme is the finite operator of verbal group and the Subject, as shown in Table 1, the unmarked Theme of the yes/no interrogative is “Should teenagers.” As for wh- interrogatives, 13.

(25) the unmarked type of Theme is realized the Wh- element, as shown in Table 1, the unmarked Theme of the wh- interrogative is “How long” (Halliday & Matthiessen, 2013). In imperative clauses, the speaker usually orders someone to do something, while the Subject is usually absent in the clause. Thus, the Predicator, a verbal group, is the unmarked Theme. In negative imperative clauses, “don’t” may also be in the unmarked Theme. As shown in Table 1, the unmarked Theme of the negative imperative is “Don’t smoke.” Another type of imperative clause includes “you and me” and beings with “Let’s,” and in this case, the unmarked Theme is the first element “Let’s” as shown in Table 1 (Halliday & Matthiessen, 2013). Marked Theme carries ideational meanings and functions differently compared with unmarked counterparts. Some marked Themes manifest circumstantial elements, including adverbial or prepositional phrases, and some may be the participants that do not function as the Subject (Martin & Rose, 2003). Marked Theme may be utilized to establish a new phases in a discourse by setting a new time or changing participants while scaffolding discontinuity (Martin & Rose, 2003). In the above, the notion and the realization of Theme are presented. The system of Theme was adopted in this study because it is responsible for the textual meaning of clause, making the organization of meaning more comprehensible to the reader (AlSharah, 2012). It is a useful framework for analyzing the textual meaning in a clause since these functional components significantly contribute to the cohesive development of the text (e.g., Belmonte & McCabe, 1998; Mellos, 2011; Eggins, 2014; Jing, 2014; Ebrahimi & Ebrahimi, 2012c; Ebrahimi & Khedri, 2012; Wang, 2007). Eggins (2004) even claimed that Thematic choices contribute to the internal cohesion of text, and that “skillful use of Thematic selection results in a text which 14.

(26) appears to 'hang together and make sense'” (Eggins, 2004, p, 321). She suggested that what we choose as the Theme and how the Topical Theme are realized may largely depend on the mode of the language, and using them right reflect the speakers’ or writers’ ability to organize information suitable for the mode. For marked Theme, she argued that successful use of marked Theme can improve the connection of clauses. For example, a previous information can be nominalized as marked Theme in order to be the departure of the new information. Above all, the way thematic elements succeed to one another, namely, the way Themes and Rhemes are patterneed across clauses, contributes to cohesion (Eggins, 2004). Successful thematic choice will lead to efficient thematic development, which contributes to the cohesion of the text (e.g., Downing, 2001; Li, 2009; Mellos, 2011; Wang, 2007), and the patterns of how thematic elements develop are illustrated next. Thematic Progression (TP) The principles concerning the choice of Theme and Rheme have received great attention since appropriate choice of Themes may improve the flow of information (Eggins, 2004). The underlying patterns of how Themes and Rheme are selected in text were explored by Danes (1974), and he proposed the concept of “thematic progression,” which was considered to be the “skeleton of plot.” Thematic progression (TP) in a discourse determines information organization of a text, exposes the information flow, and contributes to textual cohesion (Eggins, 2004). Therefore, TP is usually regarded as an explicit framework of text analysis to reveal the thematic selection and progression of a text. This study adopted Danes’ (1976) framework of TP, in which he categorized TP into four major types, which are (1) Simple linear TP (or TP with linear thematization of Rhemes), (2) TP with a constant Theme, (3) TP with a derived Theme (or a 15.

(27) hyperTheme), and (4) Exposition of split Rheme. Each type of TP is described and demonstrated with a figure and an example text below: TP with a constant Theme. In TP with constant Theme, the item in the Theme of the first clause is also selected as the Theme of the subsequent clause. Figure 2 presents the pattern of TP with a constant Theme.. Figure 2. TP with a Constant Theme (Danes, 1974, p. 118) According to Eggins (2004), by repeating the Theme from the previous clause, the writer may effectively achieve lexical cohesion and maintain the clear focus of the text (Eggins, 2004). Nonetheless, using constant Theme as the departure of each clause prohibits the text from developing, so this cannot actually lead to progression of the text. That is, the new information which is introduced in Rheme cannot be further developed or elaborated on. In the following example, the Theme of the first clause (Young people) becomes the Theme of the second and third clause: Young people tend to add variety in their life and they refuse to be stuck in monotonous daily life. Besides, young people like to be attentive and get recognition from their peers as was as the elderly. In fact, they indeed benefit from it. Simple linear TP (or TP with linear thematization of Rhemes). As proposed by Danes (1974), simple linear TP is the most elementary pattern of TPs, in which the item from the Rheme in the first clause is reiterated as the Theme of the next. Figure 3 illustrates how Themes are arranged in simple linear TP:. 16.

(28) Figure 3. Simple linear TP (Danes, 1974, p.118) In this pattern, a new Theme is deployed in the following clause as opposed to the case of constant TP, and this development can be considered cohesive because the Theme is promoted from the Rheme of the previous clause (Eggins, 2004). This type of TP also leads to a cumulative development of a text in that it develops new information in the succeeding Theme cohesively. In the following example, the Theme of the second clause (They) is the Rheme of the first clause (young people): Talent competition is very popular among young people in Taiwan. They seek for achievements, fame, or they simply need audience. TP with a derived Theme. In TP with a derived Theme, “the particular Themes in subsequent clauses are derived from a hyperTheme or from the same overriding Theme” (Belmonte & McCabe, 1998, p. 18). Figure 4 showed the type of progression:. Figure 4. TP with a Derived Theme (Danes, 1974) HyperTheme can be realized as the topic clause of a paragraph or a text section (Danes, 1974), and the following clauses are controlled or predicted by this clause. HyperTheme is a higher level of clause which is broader or more general, and the subsequent clauses are developed under the clause. Therefore, the Themes in subsequent clauses should not be seen as new items but relevant notions cohesively derived from the hyperTheme. In the following example, the Themes of the second (Some of them) and third (Some) clause are derived from the Theme (Young people) of 17.

(29) the first clause: Young people show great interest in participating in such contest. Some of them may join the competition to win money. Some may see the competitions as the gate to become a super star. Exposition of split Rheme. The exposition of split Rheme should not be seen as a new type but a combination of linear and constant TP, which frequently occurs in texts (Danes, 1974). In this kind of TP, the Rheme of the first clause is split into two or more ideas; these ideas are then developed in the Themes of subsequent clauses. Figure 5 showed the type of progression:. Figure 5. Exposition of Split Rheme (Danes, 1974) According to Eggins (2004), this type of TP is also called “the multiple-Rheme pattern” since the Rheme of the first clause consists of more than one piece of information. He suggested that this pattern is commonly used in longer expository texts, which provides the underlying organizing principle. In the following example, the Rheme of the first clause (their schoolwork and immaturity) is elaborated respectively in the following clauses: In my opinion, young people should participate in talent competitions due to their schoolwork and immaturity. First, spending too much time preparing for the competitions deprive them of the time for schoolwork. Furthermore, since they are immature, they might not be aware of the consequence after they abandon their schoolwork. 18.

(30) Rhematic progression. The focus of TP is on how Theme links to a previous text, whereas some researchers argued that writers very often repeat the elements of the Rheme of one clause in the Rheme of its preceding clause, and this type of TP is named “Rhematic progression” (Chao, 2002; Crompton, 2004; Hawes & Thomas, 1996, 1997). Figure 6 showed the type of progression:. Figure 6. Rhematic Progression Rhematic progression is not mentioned by Danes (1974), while sometimes it can be a more dominant type compared with basic TP types (Crompton, 2004). Chao (2004) also pointed out that in expository essays written by Taiwanese learners, Rhematic Progression appeared more frequently than linear and constant TP. Thus, in order to reveal the information structure of essays, Rhematic Progression is adopted as the fifth type of TP in the present study. The following is an example of Rhematic progression, in which the element in the Rheme of the first clasue, children’s talents, appeared again in the Rheme of the preceding clause: Since studying is not the only way to make a fortune any longer, parents start to devote their money to foster children’ talents. Thus, even schools pay attention to the development of children’s talents. Apart from the TP types above, Danes (1974) also mentioned a case as the modification of TP, which he called “TP with an omitted link.” In this case, one utterance of TP is deleted since the information it carries is claimed to be obvious and implied by the context. This kind of “TP with a thematic jump” (Danes, 1974) or “gap” (Hawes & Thomas, 1996) was not included in this study because the definition 19.

(31) failed to specify how wide this gap between two thematic elements should be. Additionally, there is no consensus on how this type of TP can contribute to the cohesion of text. Therefore, in this study, whenever there is a gap between two thematic links, a specific and concrete rule is adopted: “Themes for which the reader has to go back more than two clauses to find a previous mention of a concept were not counted as part of a thematic progression chain” (McCabe, 1999, p. 176). As for the Themes that did not exceed the limit of two clauses, they were still counted as a part of TP (Crompton, 2004; Ebrahimi & Ebrahimi, 2012a; Jalilifar, 2010; McCabe, 1999). In summary, the patterns of the selection of Themes across succeeding clauses contribute greatly to textual cohesion and the patterned have been explicitly stated by SFL theorists. After the theoretical foundation of SFL, the system of Theme and Rheme, and TP were introduced, empirical research on TP were reviewed in the following subsection. Empirical Research on TP In this section, empirical research on thematic progression (TP) is presented. TP has been investigated as researchers and instructors gradually shifted the focus from traditional grammar to discourse features on a text. Five subcategories of research are reviewed in this section. The first category involves studies which investigated the pattern of TP used by non-native writers; the second section concerns the relation between TP and genre; the third section reveals and compares the features of TP in higher-rated and lower-rated essays; the fourth part targets how writers utilize marked Themes in their essays. All the four categories of research mentioned were done through text analysis approach, while they did not explore the aspects of teaching and learning knowledge of TP. The fifth category, therefore, aims at reviewing studies on whether instruction based on TP improved students’ use of TP in their writing. 20.

(32) TP in Writings Written by Non-native Writers A line of research (e.g., Green, Christopher, & Mei, 2000; Hawes & Thomas, 1997; Hu, 2008) investigated the features of English learners’ TP by conducting contrastive analysis on learners’ and native speakers’ writings. For example, Belmonte and McCabe (1998) conducted a contrastive analysis in Spanish EFL context. In the first part of their research, 40 professional texts were analyzed to find out how professional writers construct their TP structure, and the results showed that these writers employed a large number of cross references between a Theme and a previous Rheme. That is, the information in Theme was connected to previous information that had been mentioned. In the second part, 25 students’ compositions on two topics were analyzed in order to find the difference of their TP features as opposed to those in professional writings. They revealed that students overused constant TPs in their text while few linear and split progressions were used. This feature was a common problem in learners’ text as shown in numerous studies (Arunsirot, 2013; Chao (2002); Crompton, 2004; Hawes & Thomas,1997; Khedri & Ebrahimi, 2012; Li, 2009; Mellos, 2011; Rakhman, 2013; Soleymanzadeh & Gholami, 2014; Wang, 2007). In additions, some students also used Themes which confused readers, for example, they used “first” in the first clause without any corresponding “second” or “third” in the following clauses. Moreover, some students’ essays showed excessively long intervals within a single progression and an excessive number of brand new Themes, which might resulted from an inferior plan of development. The overuse of “There” to begin a clause was also a problem because a Theme connected to a previous Rheme might provide a smoother progression (Belmonte & McCabe, 1998). Following their research, a few researchers (e.g., Arunsirot, 2013; Green, Christopher, & Mei, 2000; Hawes & Thomas, 2012; Herriman, 2011; Li, 2009; Wang, 21.

(33) 2007) have disclosed problems of students’ Theme-Rheme structures and TP. Wang (2007) analyzed the essay of a sophomore politics major from South-Eastern University in China. The results showed that the text contained few cross-referential TPs and the student overused constant TPs. Similar to Belmonte and McCabe’s (1998) finding, the student was inclined to begin the clause with “There” too often, which might not contribute to the development to Theme. The problem of brand new Theme in learners’ essays was also tackled by numerous researchers (Arunsirot, 2013; Belmonte & McCabe, 1998; Hawes & Thomas, 2012; Li, 2009; Wang, 2007). For example, Hawes and Thomas (2012) investigated 16 essays written by nonnative pre-MA students who majored in media and journalism, and their essays were compared to a native corpora. They pointed out that students might position new information in a brand new Theme instead of Rheme, and some students placed too many elements in Theme, which made the topic vague. Li (2009) also suggested that introducing new things in Themes showed that the writer could not develop the idea mentioned previously. This may force readers to make conceptual connection, and sometimes readers may also be confused and have difficulties following the text. A few researchers found that learners were prone to insert too many materials between one clause and a subsequent thematization (Belmonte & McCabe, 1998; Green, Christopher, & Mei, 2000; Herriman, 2011; Li, 2009). Li (2009) explored 45 expository essays by third year English majors in China, and she revealed that Chinese students sometimes intervened too many elements between a concept in a Rheme and the subsequent Theme. Readers, therefore, had to go through a long length of text to find a previous mention, and the development to of the text became ineffective. 22.

(34) Empty Rheme or empty Theme was another common problem in learners’ essays (Arunsirot, 2013; Belmonte & McCabe, 1998; Wang, 2007). Arunsirot (2013) investigated 114 essays written by second year undergraduate students in Thailand EFL context and analyzed their TP and problems. The result showed that Thai English learners were influenced by their mother tongue and they frequently omitted the subject of the clause. This grammatical problems in turn prohibited the text from developing in that The Theme did not exist in the clause, and Arunsirot (2013) named this phenomenon “the empty Theme.” On the other hand, Belmonte and McCabe (1998) indicated that in learners’ essays, the Rheme of a clause might sometimes be empty of any information, which meant that learners fail to place useful information to develop the Theme but only used vague or irrelevant terms. If the Rheme could not complement information for the Theme, the text could not be further developed. In sum, previous studies generally pointed out that EFL or ESL learners or novice writers had difficulties managing the TP in their writings (e.g., Arunsirot, 2013; Belmonte & McCabe, 1998; Hawes & Thomas, 2012; Li, 2009; Wang, 2007; Yang). The writings written by English learners often manifest problems such as overusing constant TP, selecting confusing Themes, using empty Themes, intervening too many materials between Theme and subsequent thematization, and placing brand new element in Theme. However, the general problems of Taiwanese learners’ writing were scarcely explored, which indicated a niche for empirical investigation into the TP deployed in Taiwanese learners’ essays. TP and Genre Based on SFL definitions, linguistic features vary across different genres (Eggins, 2004; Halliday & Hasan, 1976), which suggests that the deployment of TP is also sensitive to genre. In Fries’ (1995) study, he hypothesized that “different patterns 23.

(35) of thematic progression correlate with different genres, i.e. patterns of thematic progression do not occur randomly but are sensitive to genre” (Fries, 1995, p.319). TP in three kinds of texts, obituaries, narratives, and an expository text, were investigated and compared to testify his hypothesis. The results indicated that obituaries manifested more Theme iteration (constant TP), and the narratives for children showed a tendency to employ Theme iteration. For expository texts, there was a clear tendency for linear TP to be used since the Theme of the second clause usually elaborated the Rheme of the first clause. The findings supported his hypothesis that genre affected TP. After Fries (1995) proposed this difference, Wang (2007) also argued that constant TPs were better employed in narrations to make the text static while crossreferential TPs (Linear TPs) were more necessary for argumentative and academic texts to create a dynamic effect. According to Crompton (2004), argumentative text with higher quality manifested fewer constant TPs than lower quality counterparts. However, contrary to Fries’ (1995) assumption, some studies (Jalilifar, 2010; Mellos, 2004; Rakhman, 2013; Soleymanzadeh & Gholami, 2014) still found that EFL and ESL learners tended to use more constant TPs in expository genre. For example, Soleymanzadeh and Gholami (2014) conducted a study on 13 undergraduate English majors in Iranian EFL context and found that students employed more constant TPs than linear TPs in argumentative essays; Rakhman (2013) carried out a study on nine expositions by secondary school students in Indonesian EFL context and found that more constant TPs were used than linear TPs in argumentative essays Hence, this study also accessed how TP manifest in expository essays so as to shed light on whether the patterns conform to the convention of expository genre.. 24.

(36) TP in Higher-rated and Lower-rated Essays Learners’ proficiency level has always been considered a factor that influences the manifest of TP in essays (Chao, 2002; Hawes & Thomas, 1997; Mellos, 2011). Chao (2002) investigated the TP in in argumentative essays written by Taiwanese senior high school EFL students. The participants were required to express their opinions toward on-line education, and all essays were rated by two GEPT raters to be categorized into high-rated and low-rated group. Results showed that both high-rated and low-rated essays utilized mostly simple linear and constant TP, while high-rated essays manifested simple linear TP slightly more. Similarly, Mellos (2011) argued that in the eight undergraduate ESL essays he analyzed, those with higher scores employed complicated unmarked Themes and various TP to weave different segments of discourse, while the low scoring papers overused simple nominal groups and constant TPs. Hawes and Thomas’ (1997) investigated TP in 100 essays written by – higher-level and lower-level Malaysian EFL students. Their findings revealed that students with higher proficiency outperformed their lower proficiency counterparts in using more marked Themes to create variability, using less repetitive patterns, and committing fewer errors in Themes. More researchers (Crompton, 2004; Hawes & Thomas,1997; Ebrahimi & Ebrahimi, 2012a) also suggested that constant TP appeared most frequently in essays written by lower level students, while advanced learners employed many derived progressions in their essays. For example, Ebrahimi and Ebrahimi (2012a) analyzed 180 compositions written by 60 EFL students from Abadan Islamic Azad University in Iran. The students were categorized into sophomore, junior, and senior groups based on their academic experience, and they were assigned to write on three pictorial stories. The results of Chi-square showed significance in difference between the three 25.

(37) groups in linear TP pattern, constant TPs, and split Rheme. Ebrahimi and Ebrahimi (2012a) argued that linear TP was used more significantly by senior students to guide readers through a more cohesive flow their texts; constant TPs were avoided by seniors for fear that overusing these progressions turned their writings into a repetitive and redundant list. Split Rheme were used more significantly by seniors to create a specific and apparent layout of information structure. A big proportion of clauses were not consistent with any TP types, which were named “miscellaneous progression pattern,” and no significant difference were found across three groups on their use of this pattern. Crompton (2004) investigated 40 argumentative essays by ﬁrst year undergraduates in Brunei and compared them with a native professional writer. His results also supported that higher-level learners employ more linear TP in their essays compared to lower-level counterparts. Marked Theme in TP In normal cases of declarative clauses, marked Theme is defined as the Theme which is not realized by the grammatical Subject of the clause. Marked Theme is usually noticed prominently by its listener or reader since it is atypical compared to the norm. In clauses where the Theme conflates with the grammatical Subject of a clause, the Theme of the clause is categorized as “unmarked Theme” (Halliday & Matthiessen, 2013). They may also be used to establish a new phases in a discourse by setting a new time or changing participants while scaffolding discontinuity (Martin & Rose, 2003). Thus, the role of Marked Theme is significant in that it influences the deployment of TP. Green, Christopher, and Mei (2000) conducted a study on academic texts written by Hong Kong freshmen to analyze how marked Theme was used. The data from 26.

(38) Chinese corpus was compared with that of an American corpus. Results showed that Chinese learners placed an excessive number of connectors as marked Theme. These learners were prone to place new information at marked Theme; they also employed an unconventional large number of marked Theme which interrupted the information structure and cohesion of their essay. The Chinese EFL students were inclined to overuse Themes such as “but,” “and,” and “however” and place too many temporal and special adjuncts at clause initial position (Hu, 2008). Similar result was found in Yang, Ramírez, and Harman’s study (2011), in which they compared Chinese students’ essays with native speakers’ essays. Result showed that the Chinese students used fewer TP than native counterparts and overused conjunctive adverbs, which affected textual cohesion since these marked Themes were not drawn from the Rheme in a previous clause. However, when marked Themes are appropriately used by higher level learners, those marked Themes increased textual cohesion; contrarily, lowerlevel students often limited their Theme to Subject in order to avoid grammar errors (Ebrahimi & Ebrahimi, 2012b). Previous researcher often agreed on the tendency that marked Theme was a notable feature of high level students’ essays (Hawes & Thomas, 1997). Hewing (2004) investigated the native undergraduates’ geography writing to develop a description of the specific disciplinary writing and compared the Themes used by students with different academic experience. Results revealed that first-year students composed their writings with simple TP and employed a prominent number of unmarked Themes. She attributed the results to their lack of knowledge in academic writing and their discipline. In contrast, the essays written by third-year students manifested a greater quality of Theme choice, which realized interpersonal functions and provided logical connection. These students utilized more marked Themes to 27.

(39) establish interpersonal links and created relevance between clauses successfully. North (2005) also investigated the Themes used by native undergraduates of different disciplines, in this case, art majors and science majors. The findings suggested that marked (circumstantial) Themes were usually used to be associated with discontinuity by specifying changes. Art students’ essays were rated higher than science students, and one of the reasons was their better command over the use of orienting (marked) Themes. Art students used more attribution (marked) Themes to construct complex TP and compose a cohesive text. From the above discussions on TP from a text analysis approach, these studies of TP generally revealed that students tended to construct more cohesive texts if they have satisfactory command over TP strategy, and their texts may be poor in cohesion if TP is not successfully deployed; some ESL and EFL writers limited the use of marked Themes to conjunctive adverbs or failed to effectively use a comprehensive range of marked Theme. Some researchers (e.g., Ebrahimi & Ebrahimi, 2012a; Rakhman, 2013) argued that ESL/EFL students focused nearly completely on word and clause levels while ignoring textual cohesion, so teachers and students’ lack of knowledge and experience in TP indicates the need to provided explicit guidance on deploying TP. In light of these concerns, the present study will investigate the TP in Taiwanese writers essays to recommend promising ways to help them improve their cohesion. However, apart from studying TPs, the existing TP types cannot account for all the links in a text; that is to say, some links in a text do not fall into any TP category. Crompton (2004) tried testify whether TP typology accounted for all the TP in all texts, but the hypothesis that all clauses fit into a TP type was not supported. In another study by Hawes and Thomas (1997), clauses that did not fall into any TP 28.

(40) types, which they termed “breaks,” constituted around 30% in students’ essays. Similar results were also found by Ebrahimi and Ebrahimi (2012a) that the clauses that did not fit any TP type in college students’ essays, which they called “Miscellaneous Pattern”, accounted for around 20% of all clauses. These studies concurrently suggested that a text consists of a large proportion of “breaks,” however, so few researchers explored the distinctive existence of breaks in EFL learners’ essay. As a consequence, this phenomenon was examined and discussed in the present study. Instruction in TP Another field of TP research which is receiving increasing attention is the instruction of TP in EFL and ESL context. A few studies (Albufalasa, 2013; Krisnawati, 2013; Liu & Liu, 2013; Yang, 2008) have suggested the benefit of instruction on TPs in improving students’ information structure. Krisnawati (2013) conducted a study on 25 Indonesia EFL university students who were taught explicitly the notion of Theme-Rheme structure in a semester-long writing course. She found that the introduction of TP guided students to construct their ideas better. Students used their knowledge of TP when drafting and revising their writings, and high level students improved their cohesion more significantly. Similarly, Liu and Liu (2013) carried out an experiment in Chinese college English course to examine the validity to apply Theme-Rheme theory in teaching writing. A total of 50 students took a 20-hour course to receive instruction on ThemeRheme structure and TP. In the end, it was found that teaching TP to students significantly improved their coherence in writing. Shen (2004) explored the effects of teaching topical development to 43 Taiwanese EFL senior high school students. Three stages of treatment (Pre-treatment writing, Treatment writing, and Post-treatment writing) were carried out, and students were instructed on topical development and 29.

(41) required to revise on their writings. The result of the post-test showed that the students better arrange their ideas in the text that most students still employed parallel progression (constant TP), with few students using sequential progression (linear TPs). However, in the post-test, more students used “extended parallel progression” in their essay, which meant that they went back to the original topic after a few clauses. This was considered a progress since students did not lose track of the discourse flow. In sum, the literature reviewed in the previous sections showed that EFL and ESL students were not familiar with how to use TP to weave their ideas cohesively throughout their text. Teachers may also lack knowledge of how to apply the teaching of TP in writing classes. This lack of knowledge and experience in TP urges the need to provide explicit principles for guidance. If some suggestions for instructing students on the effective use of TP in writings can be developed, writing instructors can also integrate the knowledge into lesson planning and treat students’ problems in a more efficient way. Research Questions The purpose of this study is to explore the cohesion and TP in higher-rated and lower-rated expository essays from GEPT writing test written by Taiwanese EFL learners. Three research questions guided this study: 1.. What are the proportion of various TP types in the expository essays from GEPT writing test? 1.1 What are the overall proportion of TP types? 1.2 What are the proportion of TP types in level 3 and level 4 essays? 1.3 What are the variety of TP employed in level 3 and level 4 essays?. 2.. How frequently are marked Themes employed in level 3 and level 4 essays?. 30.

(42) 3.. What are the features of the learners’ essays in terms of the deployment of TP and marked Themes?. 31.

(43) CHAPTER 3 METHODOLOGY The purpose of this research is to deepen our understanding of information flow in Taiwanese English learners’ compositions. Intact samples of compositions written for the General English Proficiency Test (GEPT) were obtained from The Language Training & Testing Center (LTTC) of Taiwan. Both quantitative and qualitative methods were conducted to investigate the proportion of different TP types in higherrated and lower-rated essays and the quality of TP. The Corpus The corpus of the present study was comprised of essays from the Guided Writing section of high-intermediate level General English Proficiency Test (GEPT), which was designed by The Language Training and Testing Center (LTTC) in Taiwan. GEPT came into existence in 1997, sponsored and supported by our Ministry of Education. It has five levels of tests, including (1) basic, (2) intermediate, (3) high intermediate, (4) high, and (5) superior level. In each level, all four skills are examined in the test. The results of the test can serve as convincing reference of examinees’ English proficiency, and it is widely accepted by schools, universities, and government agencies. The data of GEPT writing tests is worth exploring since the test takers are not limited to students but a wider range of people, such as businesspersons, engineers, researchers, customs officers, tour guides and so on. In this light, the general problems of Taiwanese learners of different backgrounds can be targeted. Unlike classroom environments or self-made tests, GEPT writing test is administered in a strictly controlled environment, and each tests is designed through a rigorous progress. The writings from the tests, therefore, are more reliable and can be more representative of students’ actual performance. GEPT is also a leveled test, which means that the examinees who take the same level of test might be more homogeneous in English proficiency. This feature can in turn narrows 32.

(44) the gap between all test takers at one certain level. In the present study, the corpus was made up of essays obtained from the Guided Writing section of high-intermediate level. It was claimed by LTTC that examinees who passed this test may have a generally effective command of English and could deal with a wide range of topics; however, this study only included the expository essays from the test, so the genres of other types were not addressed. The essays from the Guided Writing section of high-intermediate GEPT were rated holistically, and the rating scheme ranged from 1 (poor) to 5 (excellent). According to LTTC, the essays rated above level 3 generally answer to the rubrics and were written with proper content, fluent structure, and fine command over grammar and vocabulary; those below level 2 could not conform to the requirements of the rubric and the content could barely be comprehensible to readers; level 1 essays particularly failed to reach the requirement of the rubrics and manifested a large amount of errors in organization and grammar. Therefore, level 1 was excluded since those essays were informed to be too short in length and manifested too many basic or local errors, making it difficult to study cohesion at discourse level. Based on the rating scheme, the researcher thus only applied for essays whose score ranged from 2 to 5, and 240 writings authorized by LTTC were provided. After receiving the data, a pilot test was conducted on a 20 essays, 10 of which were from level 2 and the other 10 from level 4. The results revealed that level 2 essays were not proper for the analysis of this study because many of them did not reach the length requirement of the rubrics (150 words), and the writers’ command over the language was not effective. Some of the writings were off-topic, and their contents were mostly incomprehensible. Therefore, the 39 essays from level 2 were excluded.. 33.

(45) To compare between higher-rated and lower-rated essays, 50 essays from level 3 were selected as the lower-rated group, and another 50 essays from level 4 composed the higher-rated group to exemplify difference, if present. The criteria for selecting these 100 essays was whether they conform to the word length requirement of the rubrics (at least 150 words); namely, essays with inadequate number of words were excluded. For the second research question, the essays were discussed from a qualitative perspective, so all 106 essays from level 3 (50 essays), level 4 (50 essays), and level 5 (6 essays) were included for text analysis. The total word count of higher-rated essays was 12924 words, and the total word count of lower-rated was 12352 words, which indicated that two groups of essays were similar and comparable. The total number of clauses in higher-rated essays was 839, and that of lower- rated essays is 813, which is slightly fewer than the higherrated essays. Table 2 is the overview of the corpus: Table 2 Overview of the Corpus Lower-rated. Higher-rated. Total. Essays. Essays. Level 3. Level 4. Level 5. Total. The Number of Essays. 50. 50. 6. 106. The Number of Words. 12352. 12924. 1361. 26637. The Number of Clauses. 813. 839. 73. 1736. Note. Higher-rated essays included both level 4 and level 5, but only those from level 4 were quantitatively compared with lower-level essays. The Writing Task In the test, each essay was written on the same topic. Examinees would read the guidelines first, which included the topic and two main requirements. They were 34.

(46) required to compose an expository essay on their view points toward “talent show,” and their essays should involve the two aspects: (1) specific reasons why young people participate in the shows and (2) possible consequences for attending the shows. The required length of the writing is 150-180 words; the genre of this writing was expository in nature that students were expected to elaborate their viewpoints on a topic with their arguments. Approaches to Analysis In this section, the procedure to analyze TP in the corpus was explained. The framework for identifying Theme, Rheme, and the markedness of Theme was presented first; then, the way of identifying TP was explained. Before quantitative analysis, each essay was broken down into “clauses,” which was the basic unit of meaning as defined in Halliday and Husan’s (1976) model. In the next step, the Theme and Rheme of each clause were specified based on Halliday and Matthiessen’s (2013) approach. They suggested that each clause can be divided into a Theme and a Rheme, and the former always precedes the latter. In declarative clauses, for example, Theme was identified as the element(s) appearing in front of the main verb, which may be the Subject and sometimes adjuncts or conjunctive phrases. However, although Theme might sometimes be manifested by an adverbial group or a prepositional phrase, some researchers argued that the Subject of the clause should be included in Theme when conducting Theme-Rheme analysis (e.g., Green, Christopher, & Mei, 2000; Lin, 2011). This study also adopted this approach, so the boundary of Theme and Rheme is identified as the following: (1) In declaratives, any element before the main verb of the main clause is the Theme; the remainder of the clause including the main verb is the Rheme (Green, Christopher, & Mei, 2000). 35.