Luận văn thạc sĩ Ngôn ngữ Anh: Genre analysis of PhD thesis literature reviews written by Vietnamese in English

Therefore, this thesis attempts to explore what characterises the genre of the thesisliterature review chapter written in English by Vietnamese PhD students tohighlight the issues in lig

INTRODUCTION .- Gà HH HH HH ng nưệt 1

Aim and objectives of the SUỈy .- 5 112211121113 1119 1119 11191 1 ng ng 6

The overarching aim of the thesis is to explore what characterises the genre of the thesis literature review chapter written in English by Vietnamese PhD students to highlight the issues in light of general academic conventions and help improve the quality of literature review writing among PhD students The overarching aim can be specified in the following objectives:

1 Exploring the generic structure of the thesis literature review chapters written in English by Vietnamese PhD students.

2 Exploring the typical linguistic resources employed in the literature review chapters.

The following research questions are formulated to achieve the aim and objectives mentioned above:

1 What is the generic structure of the thesis LR chapters written in English by Vietnamese PhD students?

2 What typical lexicogrammatical features are employed to realize the generic structure of LR chapters written in English by Vietnamese PhD students, and how are they employed?

In addressing these questions, the study will identify the macro-structure, or framework, for novice researchers to understand the generic structure of PhD literature reviews and help facilitate the thesis writing process In doing this, move analysis is conducted to explore whether the literature reviews comply with the rules of good writing with a clear organizational structure and what kind of and how moves and strategies are presented Besides, the study will examine the linguistic realizations of moves and strategies in order to provide guidance on how to use proper linguistic resources to fulfill communicative purposes or functions specified for a thesis literature review It is hoped that the linking of move analysis and linguistic analysis can enhance our understanding of how a thesis literature review is constructed both at the macro and micro levels.

Scope of the S{UY - SH HH HH HH HH Hhh 6

Given that this study is concerned with genre analysis of 30 LR chapters written in English by PhD students of English language from a university in

Vietnam, its scope will be confined to (1) investigating their general macro-structure and (ii) investigating typical language features such as reporting verbs and modal verbs to operationalize the communicative purposes of generic structural elements.The purpose is to explore how these 30 chapters are organized in terms of genre and what outstanding language features are employed to fully operationalize the functions of LR chapters.

Significance of the Study 1 ố ao

The study attempts to integrate move analysis and linguistic analysis to analyse genre features of LR chapters in doctoral theses written in English by Vietnamese PhD students in the field of the humanities and social sciences.

The move analysis proposes a representative model of the overall schematic structure of LR chapters in doctoral theses written in English by Vietnamese students in the field of the humanities and social sciences Thus, the study will reference future rhetorical edits and comparisons across academic disciplines The application of linguistic analysis in this study identifies lexicogrammatical features used to realize different rhetorical moves Integrating the two approaches will potentially enable us to understand the concept of rhetorical moves and their linguistic realizations more completely.

It is expected that this study’s findings will enable Vietnamese PhD writers to more successfully meet the requirements of academic discourse so that they can(i) understand more deeply the generic structure of a literature review chapter in a doctoral thesis and (1) write this LR chapter correctly and appropriately in terms of generic structure, communicative functions, and linguistic realisations Finally, this study aims to help these researchers feel more at ease participating in activities in a professional setting conducted in English.

Structure of the tẽ€S1S eseccscssecseeseessecssecscesseessecssesseeseesseesseseesseesseeseees 7

This thesis consists of six chapters The first chapter is the introduction to the study, which briefly presents the rationale of the study, research aims and objectives, scope, significance of the study, and structure of the thesis.

Chapter 2 reviews relevant literature for this study, including the definition of the genre, genre analysis, the role of literature review in a PhD thesis, a review of the Creating a Research Space (CARS) model, and previous studies on genre analysis of different parts of PhD theses, in order to establish the research gaps that this study tries to fill.

Chapter 3 provides in detail the research methodology of the study, which consists of the selection of research philosophy, research approaches for the present study, the description of data collection, the data analysis procedures, as well as the criteria for ensuring that the study was conducted in reliable and ethical conditions. The chapter ends with detailed descriptions of coding and quantifying a search word in thesis literature reviews.

Chapter 4 presents results and discussion of the results concerning the generic structure of thesis literature reviews.

Chapter 5 presents results and discussion concerning lexicogrammatical features realizing different communicative purposes, or rhetorical features in thesis literature reviews written in English by Vietnamese students.

Chapter 6 is the conclusion of the study It provides key findings, examines the pedagogical implications, points out the study's limitations, makes suggestions for further analysis, and makes concluding remarks about the research.

LITERATURE REVIEẨW SH HH HH ngu 9

Doctoral thesis literature review Chapter - s5 S11 + sgk 22

2.3.1 What is a doctoral thesis literature review chapter?

There are several reasons why we need to review the literature when writing a thesis or dissertation According to Paltridge and Starfield (2007), a very important purpose of a literature review is to ‘contextualize’ the student’s research and to describe and synthesize the major studies related to the topic of the research (p 99) However, the functions and format of the literature review vary significantly depending on the academic level Hart (1998) points out that for bachelor projects, the literature review is only descriptive, topic-focused, and mostly indicative of the main current sources on the topic; for master dissertations, the literature review is analytical and summative, covering methodological issues, research techniques, and topics; and for PhD theses, the literature review should be analytical synthesis, covering all known literature on the problem, including, in other languages, critical evaluation of previous work on the problem It means that the doctoral literature review should be made up of a depth and breadth of discussion on relevant traditions and ways in which they relate to the problem Ridley (2000) defines

‘literature review' as the “part of the thesis where the extensive references related to the discussed research and theory have influenced the choice of the research topic and the methodology you are choosing to adopt” (Ridley, 2000, p 3) It can also be used to identify problems in research and to illustrate that there is a gap in previous research that needs to be filled (Ridley, 2000, p 3) The literature review, therefore, serves as the driving force and jumping-off point for the present research investigation Ridley (2000) also advises that the process of writing a literature

22 review is an ongoing activity that begins when you pick up the first book or article related to your research and continues until the day you finish the final draft.

There have been quite a number of definitions of LR from different scholars, showing the varying emphasis that different study guides place on its role and purpose Below are just some representative definitions:

A literature review seeks to describe, summarise, evaluate, clarify and/or integrate the content of primary reports.

LR is a written summary of journal articles, books and other documents that describes the past and current state of information, organizes the literature into topics and documents need for a proposed study.

A literature review can be defined as the selection of available documents both published and unpublished on the topic, which contain information, ideas, data and evidence written from a particular standpoint to fulfill certain aims or express certain views on the nature of the topic and how it is to be investigated, and the effective evaluation of these documents in relation to the research being proposed.

(Hart, 1998, p 13) Typically, the literature review forms an important chapter in the thesis, where its purpose is to provide the background to and justification for the research undertaken.

(Bruce, 1994, p 218) The purpose of the literature review is to locate the research project, to form its context or background, and to provide insights into previous work.

In a literature review a writer extracts and synthesises the main points, issues, findings and research methods which emerge from a critical review of the readings.

The literature review should be a coherent argument that leads to the description of a proposed study.

(Rudestam & Newton, 2007, p 63) From these definitions and quotes, we can see that the literature review is an examination of a body of related literature to connect the significant ideas, documents, and arguments presented in the literature to the research question under investigation It also serves many different purposes and entails a wide variety of activities Initially, the literature review helps researchers to formulate the research questions; at the same time, they begin to identify the relevant theories and related research studies to their own and the methodology that they might adopt for their research Later, the literature assists researchers in the analysis and interpretation of their data It is therefore not surprising that it may seem a formidable task- particularly at the beginning However, a thoroughly researched and well-written review can be both an exciting and rewarding experience (Ridley, 2000, p 4).

2.3.2 The functions of a thesis literature review

When considering the content to include in a literature review, it is important to reflect on the purposes or functions that you wish to address The aim of the LR is to provide an in-depth account of the background literature relevant to the context that your study is situated in and, in doing so, to provide an ‘argument’, 'case', or justification for the study (Bitchener, 2010) In order to do this, it is essential to consider the functions of a literature review A range of functions identified in the literature (Cresswell, 2003; Hart, 1998, 2001; Bitchener, 2010; Ridley, 2000, 2012; Kwan, 2006; Paltridge & Starfield, 2007) were presented and discussed, including the following five primary functions:

- toreview of previous research, right up to their date of examination;

- to describe and synthesize the major studies related to the topic of the research;

- to clarify relevant terms used in the research;

- to identify a gap (or a need or a problem or a shortcoming) in the literature.

- to provide a rationale explaining why it is important and significant to fill the gap;

- toinform the design and methodology of the proposed study.

It can be seen from the list that the functions defined form a clear and logical outline for a literature review However, self-reported comments made by graduate students and thesis supervisors in several studies (Bitchener & Basturkmen, 2006; Dong, 1998) reveal a common understanding of the issues that students typically encounter when writing the various parts of a thesis in English, including the literature review It reveals that both LI and L2 students have difficulties understanding and meeting the discourse requirements of the various sections (Bitchener & Turner, 2011, p 125) Regarding structuring the literature review, supervisors often highlight that their students have difficulty developing their ideas and arguments coherently because statements are not appropriately sequenced and because transitions between statements and topics are often not provided (Dong, 1998; Thompson, 1999) Bitchener and Turner (2011) explain that students have these difficulties because of their frequently limited knowledge of the functions of the genre or part-genre under consideration (p 125) Therefore, understanding the functions of LR texts will help researchers determine the type of content that is relevant to the review and guide them in how to most effectively organize it so that it reveals the ‘argument’ underpinning their study.

2.3.3 Standard convention of a literature review chapter

We have discussed the functions of the literature review chapter(s), and now it is time to consider how to organize it most effectively and what content should be included.

Although the structure of the review will vary from thesis to thesis, it will always consist of an introduction, a body, and a conclusion (Kwan, 2006; Paltridge

& Starfield, 2007; Swales & Freak, 2009; Bitchener, 2010; Ridley, 2012; Neuman, 2014) The introductory part of a literature review should start by making generalization, highlight its importance by suggesting it is interesting, problematic, and relevant, establish that a review of the literature is valuable in understanding important aspects of the research area, and end by clarifying the scope and overall organization of the review (Swales & Freak, 2009; Ridley, 2012) The review of a PhD thesis is often very long, so summaries interspersed throughout the chapter are helpful because they justify what you have argued so far and how this connects with

25 what follows, which is named a ‘transition statement' so that the flow of information is smooth (Ridley, 2012; Swales & Freak, 2009) This structure is confirmed by Neuman (2014) when he states that writing a literature review should be applied to

“all rules of good writing,” which is made up of a clear organizational structure with an introduction, conclusion, and transitions between sections (p 142) Bitchener’s (2010) study is also in line with this view when he states that the literature review is organised with “a series of themes or topic units” in the body, a summary of key claims presented in the various themes or topic units in the conclusion, and an overview of the main themes or topics that will be presented in the introduction (pp. 61-95) Besides, the themes or topic units are made up of “a series of moves and sub-moves” (Bitchener, 2010, p 66); each move is used to realize a specific overall communicative purpose through a variety of linguistic strategies (Swales, 1990; Kwan, 2006).

The conclusion of the literature review usually starts with a summary of the key claims presented in the various thematic or topic units of the body of the review In doing so, the author often draws some conclusions about the claims by

“evaluating or weighing up” the significance of the claims in light of the theoretical and research literature that has been shown (Bitchener, 2010, p 91).

One of the frequently asked questions for researchers is: how long should the literature review be? Bitchener (2010) answers this question by saying that the literature review of a doctoral thesis may lengthen up to three chapters and may occupy one-third of the thesis, excluding references and appendices (p 105).

In terms of content, this finds the answer to the question of what needs to be included in a review of literature According to Paltridge and Starfield (2007), a literature review should include: the key issues which underlie the research project; the major findings on the research topic, by whom and when; the main points of view and controversies that surround the issue being investigated; a critical evaluation of these views, indicating strengths and weaknesses of previous studies on the topic; general conclusions about the state of the art at the time of writing, including what

26 research still needs to be done; that is, the gap that remains in the research that the study will aim to fill (Paltridge & Starfield, 2007, p 101)

A review of the Creating a Research Space (CARS) modelL

2.4.1 The original Creating a Research Space (CARS) model

The Creating a Research Space [C.A.R.S.] model was first developed by John Swales in 1981 based upon his analysis of 40 introductions in research articles from various science disciplines At this time, he observed that introductions in research articles tend to be realized in a routinized 4-part structure (see Figure 2.2). This analysis model has wide methodological, empirical and pedagogical influence among the research article genre analysts (Ahmad, 1997).

Establish the field a) by asserting centrality

OR b) by stating current knowledge Summarizing previous research Preparing for present research a) by indicating a gap in previous research

OR b) by raising a question about previous research Introducing the present research a) by stating the purpose

Figure 2.2: The 4-move model of research article introductions from Swales (1981)

Although the ‘4-move’ model had some influence, certain defects had become increasingly apparent Several analysts have found difficulties in separating Move | (establishing field) from Move 2 (summarizing previous research) Swales (1990) realizes that the difficulties of separating Move 1 from Move 2 are due to

‘the increasing practice of spreading references throughout the introduction’ (p. 140) Swales further explains that his four-move model was drawn from analyzing a corpus of shorter research article introductions, while in longer research article introduction the use of references tends to recycle In other words, the four-move model of research article introduction analysis is only applicable for short introductions where the rhetorical work of establishing field of research is clearly separable from that reviewing the items from the previous research.

Establish a territory Step 1: Claiming centrality AND/OR Step 2: Making topic generalizations AND/OR Step 3: Reviewing items of previous research Establish a niche

Step 1A: Counter claiming OR Step 1B: Indicating a gap OR Step 1C: Question-raising OR

Occupying a niche Step 1A: Outlining purposes OR Step 1B: Announcing present research Step 2: Announcing principle findings Step 3: Indicating RA structure

Figure 2.3: A CARS model for article introductions from Swales (1990)

(Swales, 1990, p 141) Responding the comments on the defects of his earlier work, Swales (1990) reformulates the structure and create three-move scheme, known as Creating a Research Space (CARS) model as shown in Figure 2.3 Instead of separating Move 1 and Move 2 in his earlier model, in the new CARS model, Swales combines them into one move and calls it Move 1 (establishing a territory) According to Swales (1990), his CARS model is more capable of capturing important communicative purposes of research article introductions as he suggests “to establish the significance of the research, to situate the research in terms of its significance, and to show how this niche will be occupied and defended in the wider ecosystem” (p 140).

The model indicates how scholars support and promote their contribution to the field by first identifying the field of enquiry and summarising previous research (Move 1: Establishing a research territory), then identifying a gap in the existing work that the author’s study aims to address (Move 2: Establishing a niche), and

38 finally stating the purpose of the author’s research and summarising how they will fill this gap (Move 3: Occupying the niche) It means that the research article introduction is structured as the ‘rhetorical movement’, and it is divided into

‘moves’ based on their function in the section (Swales, 1990, p 140).

Within each of these moves, Swales identifies a range of possible “steps” that research article writers should take, such as “claiming centrality” and

“reviewing items of previous research” in Move 1 and “counter-claiming” or

“indicating a gap” in Move 2 Once the moves and steps are separated, Swales specifically analyses the typical text-patterning and lexicogrammatical features within different steps For example, in analyzing Step 3 (reviewing items of previous research) within Move | (establishing a territory), Swales looks at patterns of citation, noting patterns in which research article authors either name the researcher being cited in their citing sentence or reference the researcher in parenthesis at the end of the sentence or in endnotes Moving from text-patterning to lexicogrammatical features, Swales then identifies the frequency of "reporting verbs” (such as “show,” “establish,” “claim,” etc.) that research article authors use

“to introduce previous researchers and their findings” (Swales, 1990, p 150).

The CARS model has been widely studied since its first publication in 1990, and the wealth of research on it has not only validated the three-move structure, but also revealed its recursive nature (Bunton, 2002) and its varied realizations in research writing across contexts (Nwogu, 1997; Samraj, 2002; Lewin et al, 2001; Bunton, 2002; Kwan, 2006).

Bunton (2002) applies Swales’ CARS model to examine the move structure of introductions in PhD theses that were drawn from a variety of disciplines and written by both native and non-native speakers of English at a university in Hong Kong His introduction corpus in principle displays the three-move structure, but also exhibits a range of new move elements (see the italicized parts of Table 2.2) that led him to postulate the following revised model for thesis introductions, which forms a reference for comparison in this study.

Table 2.2: Bunton’s modified CARS model for thesis introductions

Making topic generalizations and giving background information

Move 3: Announcing the Present Research

Purposes, aims, or objectives Chapter structure

Work carried out (Eg, Sj) Research questions/Hypotheses

Materials or Subjects Defining terms

Findings or Results Parameters of research

Product of research (Eg)/Model proposed (So)

Significance/Justification Application of product (Eg)

“A=Arts, So=Social Sciences, Eg=Engineering, Si=Science, M=Medicine

(Bunton, 2002, p 74) Lewin, Fine, and Young (2001) examine the introductions of research articles taken from social sciences journals by using this CARS model Their study shows that five of the steps (i.e., Steps 1.1 and 1.2 of Move 1; Steps 1.A and 1.B of Move 2; and Step 1.A of Move 3) that are posited in the original CARS model occur with relatively high frequency, whereas the others appear much less regularly (for example, Step 1.3 of Move 1) in their corpus This uneven frequency distribution has led the authors to consider the obligatory-optional distinction to differentiate the two groups of steps The work of Lewin et al is one of the few CARS studies that address the obligatory-optional nature of the elements in the model, an issue that is also taken up by Bhatia (2004), who proposes the use of the

40 term strategy in place of steps to denote elements that do not appear regularly or in sequential order (for example, the four steps in Move 2) In the analysis in this study, the two types of move elements are differentiated accordingly, with the obligatory and sequential constituents being referred to as steps and the non- obligatory and non-sequential constituents being referred to as strategies.

Lewin et al also developed a scheme of semantic features to describe the various obligatory steps The scheme is a valuable reference for coding, as studies of the model have so far provided very little information on how the analysts went about identifying the steps and their boundaries Presumably, the coding criteria that are employed in these studies are grounded in Swales (1990) earlier characterizing account, which is functionally and semantically motivated However, some of the studies of CARS do not seem to consistently follow the criteria that Swales provides As Lewin at al observe, some coding seems to take sentences, independent clauses or other lexicogrammatical items as the unit for move-step characterization.

Despite the variations that have been recorded in the above-mentioned works, the structure has been confirmed to be fairly stable at the move level in research articles that are published in English-speaking communities.

Of the numerous studies on the CARS model, one work particularly pertinent to my analysis is the work conducted by Kwan (2006), who applies Bunton’s (2002) modified CARS model to identify the rhetorical structure of the body texts of LR chapters of 20 PhD theses produced by native English speaking students of applied linguistics from various universities in the UK, Australia, Hong Kong, and Singapore Kwan (2006) argues that in thesis manuals and writing programs, the notions of ‘introduction’ and ‘literature review’ (LR) are often used interchangeably to refer to the beginning chapters of a thesis (p 32) He also suggests that LRs and introductions in research writing may belong to the same genre because some functions of the LR have also been posited for the introductions to research articles and post-graduate theses (Kwan, 2006, p 32).

The LR chapter of a PhD thesis is a complicated and very long text, so Kwan (2006) divided each chapter into thematic units based on the sectional boundaries that were indicated by such meta-signals as section breaks, section headings, and their respective numbering systems A move analysis was then conducted on each of these thematic units It was found that moves can be recursive as is suggested by the naming of Move 1(Establishing one part of the territory of one’s own research), which allows for the possibility that the writer will establish one part after another in different theme-oriented sections of the chapter (see Table 2.3).

Table 2.3: The modified CARS model for the thematic units in LR chapters

Move 1 Establishing one part of the territory of one’s own research by:

Strategy A: surveying the non-research-related phenomena or knowledge claims

Strategy C: surveying the research-related phenomena

Move 2 Creating a research niche(in response to Move 1) by:

Strategy C: asserting confirmative claims about knowledge or research practices surveyed Strategy D: asserting the relevancy of the surveyed claims to one's own research

Strategy E: abstracting or synthesizing knowledge claims to establish a theoretical position or a theoretical framework Move 3 Occupying the research niche by announcing:

Strategy A: research aims, focuses, research questions or hypotheses

Strategy B: theoretical positions/theoretical frameworks

Strategy D: interpretations of terminology used in the thesis

Previous studies in literature review sections of doctoral theses - 45 1 Studies focusing on PhD thesis generic SfTUCfUTC€ - 5+5 < + << ss+ecs+ 45 2 Studies focusing on literature review teXts ceceecceseesteeeeeeeeseeeeneeeeeeneees 47 3 Previous studies from Vietnamese authors eee eeeeseeseeeeeeneessecseceseeeneenes 50

2.5.1 Studies focusing on PhD thesis generic structure

Bunton’s (1998) study is one of the earlier works focusing on different sections of a PhD thesis such as Abstract, Introduction, and Conclusion The research was carried with the corpus of 13 PhD and 8 M.Phil theses from six faculties (Science, Engineering, Medicine, Arts, Education, and Social Sciences) selected from the University of Hong Kong (HKU) Together they constitute a corpus of over 4,300 pages of text and 0.86 million words They represent about 13% of theses submitted in English by native speakers of Chinese at HKU over a period of 20 months.

All 21 theses were first analysed for overall thesis structure as well as for the generic structure and lexicogrammatical accuracy of their Abstract sections The 13 PhD theses were further analysed for the generic structure of their Introduction and Conclusion chapters The introductory chapters of the 13 theses were analysed for moves and steps by using the three identified in Swales' CARS model The result show that most Introductions had moves identifiable as Establishing a Territory (T), Establishing a Niche (N), and Occupying the Niche (O) like CARS model, though more steps making up the moves have been found in PhD thesis Jntroductions than those reported in the research article and Masters dissertation Introductions.

This Bunton’s modified CARS model was applied for further analyzing additional 32 PhD introductions four years later in 2002 The combined corpus of

45 Introductions is a representative sample from the disciplines with the highest output of PhD theses at the University of Hong Kong in the mid-1990s As a result, a model for PhD thesis Introductions proposed by Bunton (2002) does make use of Swales’ (1990) CARS three-move structure with its focus on Establishing of a

45 niche (see Figure 2.2) This link between the first and third moves is particularly important for the PhD thesis as a genre because the primary communicative purpose of the doctoral thesis is to demonstrate to its examiners that the writer/researcher has made an original contribution to knowledge in the chosen field.

According to Bunton (2002), the model is as important for supervisors as it may be for students Supervisors will be more able than their students to see what variations are conventional in their particular fields The model can then work as a

“pedagogic framework” around which they build the knowledge they already have, intuitively if not explicitly expressed, about the way research is presented in their disciplines (pp 74-75).

Nearly at the same time with Bunton’s (1998, 2002) studies on the generic structure of different parts of PhD theses, Thompson (2001) also conducted research into macrostructure and linguistic features of PhD theses such as the uses of citations and modal verbs The aim of the study is to achieve a richer understanding of the nature of the texts that are produced as PhD theses, to support the quality improvement for non-native speaker doctoral students who are preparing to write a thesis.

The corpus of sixteen PhD texts, half of which represent work in applied science (Agricultural Botany) while the other half belongs to the social science domain (Agricultural and Food Economics) was selected from the University of Reading in the UK The theses in the corpus were all written by native English speaker students; the author, therefore promotes the idea that the native speaker thesis is the norm and is the model to be emulated The study also approaches genre analysis the guidelines of which are suggested by Bhatia (1993), combining the use of interviews with expert informants with the analysis of data drawn from a corpus.

In addition, the study also looks at the different uses of citations and the range of functions that the core modal auxiliaries (may, might, can, could, will, would, should, must) play in the texts The data in this study is processed with the aid of WordSmith Tools by Scott (1996) for concordance searches As a result, the analyses of the citation and modal verb data indicate that PhD theses in these two

46 departments vary substantially, in their use of the research literature, and in the ways that writers position themselves in relation to their texts, their findings and to their audience.

2.5.2 Studies focusing on literature review texts

The Bunton’s (2002) modified CARS model for thesis introductions was used by Kwan in 2006 to identify the rhetorical structure of the exclusive LR chapters of

20 PhD theses produced by native English-speaking students of applied linguistics from various universities in the UK, Australia, Hong Kong, and Singapore Kwan (2006) argues that in thesis manuals and writing programs, the notions of

‘introduction’ and ‘literature review’ (LR) are often used interchangeably to refer to the beginning chapters of a thesis (p 32) He suggests that LRs and introductions in research writing may belong to the same genre because some functions of the LR have also been posited for the introductions to research articles and post-graduate theses (Kwan, 2006, p 32) Her findings confirm that that each move has a local purpose and contributes to the overall rhetorical purpose of the text In this study, the function of each text segment was first examined according to its local purpose, such as to highlight major research studies, then according to its contribution to the goal of justifying the writer’s research and the identification of predominant semantic features of the target text segment Although most of the steps in Bunton’s revised CARS model are present in the move structures of Kwan (2006), a new group is identified that relates the writer’s affirmative stance towards the knowledge or research practices that are surveyed in Move 1 (strength-claiming, relevancy- claiming, and the synthesizing of the theoretical framework) (see Table 2.3).

None of the elements occurs in 100% of their respective move, nor do the elements appear in a predictable sequential pattern Thus, they are qualified as strategies only in Kwan’s study She explains that the strategies in each move are arranged in ascending order of frequency and that the arrangement does not suggest any sequential pattern.

Another study about the LR chapters of PhD theses was carried out by Thompson again in 2009 with 24 British theses, six each from four different applied

47 disciplines, those of Agricultural Botany, Agricultural Economics, Food Science and Technology, and Psychology The aim of the study is to explore the nature of literature reviews by examining the patterns in which some of the most frequent nouns in the corpus appear, to establish what is distinctive about the literature review as a (sub)genre and to identify some of the strategies that are used in these reviews The approach taken is a corpus analytic one, starting with information about the frequency and range of nouns in the corpus, and then investigating how and for what purposes some of the most frequent nouns (evidence, problem, and model) are used in the texts, to identify and exemplify characteristics of PhD literature reviews His findings also show that a literature review in a PhD thesis typically follows the Kwan’s (2006) modified CARS model for the thematic units in

LR chapters: ‘Establishing one part of the territory of one’s own research’,

‘Creating a research niche’ and the optional ‘Occupying the research niche’ (Thompson, 2009, p 65).

Another special study conducted for the structure of complete MA theses can be named Chen &Kuo’s (2012) study which also includes the framework for LR chapters (see Table 2.4) This study takes a genre-analytical approach to examine the macrostructure of 20 master’s theses in applied linguistics using the computer software AntConc for data analysis The coding scheme for Literature review texts was based on the study by Kwan (2006) who conducted a very detailed analysis of the Literature Review chapter in doctoral theses in applied linguistics as reviewed earlier In this study the authors found, similar to Kwan (2006), that almost all theses have an Introductory move at the beginning of the Literature Review chapter, a concluding move, and between them are theme cycles The study also reveals that two steps, surveying the non-research-related phenomena and surveying the research-related phenomena, enjoy the highest frequencies and a range of 100% in this chapter in MA theses.

Table 2.4: Chen and Kuo’s (2012) revised framework for the LR chapter territory of one's own research BY

Introduction e Indicating organization of the review chapter(s) and justifying the themes (areas) to be reviewed

Move 1: e Surveying the non-research-related phenomena or

Establishing knowledge claims one part of the e Claiming centrality

Surveying the research-related phenomena

Counter-claiming (weaknesses and problems) Gap-indicating (paucity or scarcity)

Asserting confirmative claims about knowledge or research practices surveyed

Asserting the relevancy of the surveyed claims to one's own research

Abstracting or synthesizing knowledge claims to establish a theoretical position or a theoretical framework

Concluding a part of literature review and/or indicating transition to review of a different area

Indicating research aims, focuses, research questions or Move 3: hypoth

0 ng the ypotheses ccupyin ơ ws :

PY & e Indicating theoretical positions theoretical frameworks research niche by , l

e Indicating research design/ processes announcing ; ; | ; e Interpreting terminology used in the thesis

Conclusion e Providing a summary of the review of the themes and relating the review to the present study

(Chen and Kuo, 2012, pp 48-49) This study identified a new step in the move of creating a research need, which is not reported in Kwan (2006), that is, the step of concluding a part of Literature Review and/or indicating transition to reviewing a different area. Pedagogically, the study provides a more complete model of moves and steps for thesis writing in applied linguistics It is empirical and practical, helping students understand the appropriate schematic structure of theses corresponding to disciplinary communicative purposes In addition, the move/step sequence patterns identified could be further turned into cognitive writing strategies to raise students’ awareness of major and optional moves/steps and a range of options for effectively organizing information in each chapter in the theses However, it may

49 be not appropriate to apply to write PhD thesis because of the difference in length, in level of study leading to the more depth and breadth of discussion expected in the review of the literature.

2.5.3 Previous studies from Vietnamese authors

RESEARCH METHODOLOGY .00 ccecccecceeeeceeeeeeteeeneeees 57

Research design cecccescccesscesseceseecscecsaecesnecsacecsaeeeseceeeceaeessaecesaeceeeseasessaeeee 57

The process of research involves emerging questions and procedures, data typically collected in the participant’s setting, data analysis inductively building from particular to general themes, and the researcher making interpretations of the meaning of the data They are most often based on either positivist or constructivist tradition (Cresswell, 2012, 2013; Creswell & Clark, 2018) The positivist research philosophy underpins quantitative research, while the constructivist approach is associated with qualitative research.

A positivist paradigm operates with the belief that material and social reality can be understood objectively; hence, these realities are quantifiable and measurable (Cohen, Manion, & Morrison, 2013; Paltridge & Phakiti, 2015) Quantitative researchers generally propose hypotheses, and use objective measurements and numerical analysis to determine relationships and to explain causes of change. Creswell (2014) defined a quantitative study as one that is “based on testing a theory composed of variables, measured with numbers, and analyzed with statistical procedures in order to determine whether the predictive generalizations of the theory hold true” (p 2) Informed by positivism, quantitative methodology is characterised by a deductive approach to research, and involves the measurement of variables or the testing of relationships between variables in order to reveal patterns, correlations or causal relationships (Creswell, 2012; Walter, 2010) A quantitative way of understanding views the world based on "categorical data, featuring the comparison of frequencies and measurements across subjects and categories" (Maxwell & Loomis, 2003, p 249).

At the other end of the research spectrum, constructivism argues that reality is multiple and typically socially co-constructed, being dependent on who is involved, what is being studied and the context in which a study is conducted (Denzin & Lincoln, 2008; Silverman, 2016) The Constructivist paradigm assumes that human behavior is bound to the context in which it occurs, and cannot be reduced to generalized laws that apply to all situations as is the case for findings in the physical sciences Meaning is constructed inductively and holistically to understand human experience in a context specific setting (Merriam & Tisdell, 2016; Patton, 2015) Qualitative researchers code and categorize data for the purpose of identifying themes that can be used to build a theory grounded in what they observe and hear Ary et al (2006) stated, “The ultimate goal of (qualitative) inquiry is to portray the complex pattern of what is being studied in sufficient depth and detail so that one who has not experienced it can understand it” (p 476) A qualitative approach views the world “in terms of textual data, featuring the treatment of focal entities as singular wholes in context” (Maxwell & Loomis, 2003, p 249).

Often the distinction between qualitative research and quantitative research is framed in terms of using words (qualitative) rather than numbers (quantitative) (Creswell & Creswell, 2018) However, qualitative and quantitative approaches should not be viewed as “rigid, distinct categories, polar opposites, or dichotomies” (Creswell & Creswell, 2018, p 41) Instead, they represent different ends on a continuum (Creswell, 2015) A study tends to be more qualitative than quantitative or vice versa Mixed methods research resides in the middle of this continuum because it incorporates elements of both qualitative and quantitative approaches to draw their strengths Therefore, mixed-methods research is described as the collection of both qualitative (open-ended) and quantitative (closed-ended) data in response to research questions (Cresswell & Cresswell, 2018) Traditionally, quantitative research is sometimes portrayed as being unimaginative while qualitative research is often criticized for lacking generalisation, being too reliant on the subjective interpretations of researchers, and being incapable of replication by

58 subsequent researchers (Brewer & Hunter, 2006) According to Cresswell & Cresswell (2018), the reasons for choosing mixed methods research because of

“exploiting” the strength of both qualitative and quantitative research and

“minimizing” the limitations of both approaches (p 297).

Creswell and Clark (2018) state that in a single mixed research study, both qualitative and quantitative strands of data are collected and analyzed separately, and integrated — either sequentially or concurrently — to address the research question Initially, Creswell (2009) explained that the researchers need to consider the “timing” of their qualitative and quantitative data collection whether it will be phased (sequential) or be collected at the same time (concurrently) (p 206). Therefore, timing is a very important factor for researchers to shape the procedures of a mixed methods study.

Research methods eee cesseseesscesecssecseeesecesecsseeseeeseessesseeeseceseeesessaesseeeseseaeeags 59

The main purpose of this study is to examine what characterize(s) the genre of the thesis literature review written in English by Vietnamese graduates in the field of social science and humanities To achieve this aim, this research was guided by two research questions:

1 What is the generic structure of the thesis LR chapters written in English by Vietnamese PhD students?

2 What typical lexico-grammatical features are employed to realize the generic structure of LR chapters written in English by Vietnamese PhD students, and how are they employed?

In order to answer these two questions, I decided to choose a mixed methods approach based on the flexible and effective qualities that it offered It is the integration or blending of both qualitative and quantitative data collection methods to highlight its value First, qualitative data was collected and analyzed, and quantitative data collection and analysis are carried out to assist in the interpretation of qualitative findings Creswell and Clark (2011) say asa result of this design, three stages of analysis are carried out in turn namely the primary qualitative phase, the second quantitative phase, and the integration phase which integrates the two

59 analyses of data and extends the initial qualitative exploratory findings Accordingly, in this study, the mixed methods research employs the exploratory sequential approach to explore and understand the generic structure and the language features of the thesis literature review written in English by Vietnamese graduates because the data are collected in phases (sequential), which enables the qualitative (exploratory) findings from the first phase to help develop a model with the aid of quantitative data in the second phase to assist in the interpretation of qualitative findings (Creswell, 2009; Morgan, 1998).

The major advantages of the exploratory sequential mixed methods design were to increase confidence in findings, providing more evidence while minimizing possible shortcomings of using a single approach (Creswell & Clark, 2011) The design also helped to generalize the qualitative data However, this method is time- consuming to collect data from two different phases separately and the data collection of quantitative data has to depend on the result of the qualitative data.

Qualitative data collection and analysis

In the first phase of this mixed methods design study, the data was collected from 30 accepted theses that were submitted between 2008 and 2019 at a university in Vietnam The primary qualitative data consisted of textual analysis of thesis literature review chapters (N0) The qualitative data was then coded based on the communicative purposes and rhetorical features of the texts to identify moves and strategies Data from the qualitative phase were used to develop the rhetorical model specifically designed for the thesis literature review The model is made up of moves and strategies to express different functions or communicative purposes of a thesis literature review.

Quantitative data collection and analysis

In order to create a reliable model in the first phase, the quantitative strand is supplementary to work out the distribution, frequency counts, length, sequence, or order of moves and strategies The quantitative strand is also used to list out some search words which are used to realize the moves and strategies identified in the first phase In this process, the researcher first used computer software combined

60 with manual counts to locate certain given words to present the percentages, frequent counts, distribution, density, and location of the search words.

Mixed methods data analysis The findings were then presented through the blending of the two strands of data that define mixed methods research In this study, the use of a sequential design is implemented as the results from the first phase of the research will be used to build the second stage of the research design To address the research questions, integrating the qualitative data from phase one of the study with the quantitative data from phase two of the study is necessary to develop the best findings for the study.

Data Collection 0n .ồ.ồ.ồ Ố

The LR texts analyzed in this study were drawn from 30 doctoral theses in the discipline of Social Science and Humanities such as English Linguistics, Applied Linguistics, Translation, Cross-culture study, and English Language Teaching Methodology written by Vietnamese graduates in a university in Viet Nam The choice of the five disciplinary areas of the theses is aimed at representing scientific discourse written in English by Vietnamese graduates in the social sciences In addition, I am more familiar with these five disciplinary areas than with other disciplines from my previous education and working experiences, and knowledge of the thesis genre in these disciplines is very useful for my teaching in the Faculty of Foreign Language Studies.

The theses were completed within 12 years by Vietnamese PhD students All of them were located in the library catalogs of a university in Vietnam Initially, thirty-two theses were collected in the form of hard copies from the library and in the form of soft copies from the thesis repositories of the university In fact, only 30 theses include separate literature review chapters and two of which do not contain separate ones Thus, it should be noted that not all PhD theses contain a recognizable literature review.

As mentioned in the literature review section, only the LR chapters of the theses that occur between the introductory and methodology chapters were chosen for analysis Besides, the LR part needs to be indicated explicitly in the table of

61 contents, the preview of the structure as usually provided towards the end of the introductory chapter, the head, and the opening paragraphs of the potential literature review chapter.

To create the corpus, the LR chapters of these theses were copied and pasted onto separate files For analytical purposes, each LR is coded as a text Accordingly,

LR chapter 1 is coded as Text 1 (T1); LR chapter 2 is coded as Text 2 (T2), and so on to T30 in the order of the publication year for the ease of reference and the anonymity of thesis writers A detailed description of each LR chapter is provided in Appendix 3 Some background information on 30 LR chapters analysed in this study is presented in Table 3.1 These chapters, as can be seen, vary in length considerably, from 15 pages to 72 pages, but this also needs to be considered with the length of the whole thesis, as the length of theses represented here varies from

137 pages to 201 pages On average, the LR chapter occupies 24.2 % of the whole thesis This ratio is quite close to the conventional length of a thesis literature review in the whole thesis suggested by Bitchener (2010) The final column indicates what percentage the literature review chapter accounts for in a whole thesis, and this ranges from 8 percent to 41.9 percent The resulting corpus of 30 LR chapters comprised over 1278 pages of text (an average of 40 pages each) printed in font size 13 or 14 with 1.5-line spacing The total word count is approximately 400,000.

Table 3.1: Background information of the 30 PhD theses in this study

Text Discipline and area of study publication | (pages) whole thesis

TI | Applied Linguistics/ Discourse analysis 2008 36 18.5

T19 | Teacher education and Syllabus design 2017 39 26.5

There are three reasons for the selected number of LR chapters in this corpus First, it is the total accessibility to the unique available resources in the libraries The second reason comes from the corpus size suggested by Ghadessy,

Henry, and Roseberry (2001) and Conor and Upton (2004) for small specialized corpora In particular, Conor and Upton (2004) state that a small specialized corpus of “up to 250,000 words” (p 19) can be effective for studying the schematic structure and the linguistics features of the LR chapters of PhD theses by Vietnamese students in this study Finally, a review of the literature reveals that in studying the move-step structures of a single chapter or a couple of chapters of MA theses and PhD dissertations, previous scholars studied an average number of five to ten chapters per discipline (Bunton, 2002; Dudley-Evans, 1986; Samraj, 2008). Besides, the corpus of a relatively large number of texts for genre analysis is important to achieve an accurate picture of the rhetorical features of a particular genre (Biber, et al., 2009) Compared with the previous studies of move-step structures in the literature, the analysis of 30 LR chapters in this present study is manageable and big enough to provide a comprehensive picture of how non-native English students in the ELT and

EL discourse community in Vietnam compose this chapter of their theses.

The illustrating examples in this study are extracted from 30 LR chapters in the corpus All the examples are italicized, indented, and typed in a smaller font size of 12 pt Ideally, every example should be the whole text; but in practice, this ideal is unattainable So, to exemplify, I often scour short extracts or passages from complete texts, which are understandable even apart from their contexts and contain several examples in point.

The presentation of an illustrative example is organised as follows:

- Each example is numbered in Arabic numerals sequentially which are enclosed in parentheses ( ) The first numeral indicates the number of chapter; it is separated from the numeral(s) that indicate(s) the number of example by a dot; for example, (4.18) means that it is Example number 18 in chapter 4.

- This is followed by the origin of the example including the coding number of the

LR text which is enclosed in square brackets [ ] A detailed description with the coding number of each LR chapter is provided in Appendix 3 Below is an instance of how an example is presented:

Despite the careful process of delivering the interviews, the interview data of this study were not triangulated with the data from real classroom observations.

The symbol (3.2) [T21] means that it is the example number 2 in chapter 3, and it is extracted from the LR chapter coded T21.

For lexicogrammatical analysis, computer software is used to run certain search words, so the text must be stored in the form of a soft copy, then it is converted and saved as a plain text (.txt) document which is a type of digital file that is free of computer tags, special formatting, and code and can be typically recognized by lexical analysis.

Data analysis procedure - Gv HH gi, 65

The genre analysis processes followed a step-by-step procedure Just after the research objectives and research questions were identified, the exploratory sequential mixed-methods research design was chosen to combine both qualitative and quantitative data collection and analysis in a single study in a sequence of phases (Creswell & Clark, 2018).

In the first phase, 30 literature review chapters were selected and divided into themes for move analysis according to the discussed topics based on the different meta-discourse and their boundaries, which include the section headings in the table of contents, the LR introductory text, and the formatting features The texts then were coded and categorized based on the communicative purposes and rhetorical features guided by Kwan’s (2006) modified CARS model to identify the common generic structure for a thesis literature review Finally, the common rhetorical pattern for a thematic unit in the LR chapter was developed This analysis is to find the answer to the first research question.

In the second phase, first, the rhetorical moves and strategies identified in the first phase were quantified to develop a more reliable model Then, these moves were copied into separate files Once the corpora of moves were made, they were stripped of all graphs, tables, diagrams, etc., and all the texts associated with them, then they were converted to and saved as plain text files for running in the software WordSmith Version 7.0 (Scott, 2016), the Concord Tool of which locates all references to any given word or phrase within the corpus, showing them in standard concordance lines with the search word centered and a variable amount of context at either side.

Reporting verbs are used to report or refer to another writer's work It means that reporting verbs are needed to connect the in-text citation to the information which you are citing Therefore, to quantify reporting verbs, it is necessary to locate canonical citations in the texts with WordSmith’s Concord Tool first To identify the citations, the search words in this corpus are any words containing a year in brackets, a page number in brackets, third-person pronouns (he/she/they), or a proper name that can signal citations That is, the search word is expressed as a search string 19??)/20??)/???)/he/she/they, in which 19??) is for a pre-21st century corpus, 20??) for a 21*-century corpus, ???) for one, two, or three-character string including page number, for example, 158) and p.158), that ends with a closing bracket to catch other forms, and he/she/they string for non-citation types in the chapter, showing them in standard concordance lines with the search word centered and a variable amount of context at either side(see Figure 3.2).

After the citations are identified, they are further analysed for reporting verbs and their tenses and voices In addition to reporting verbs, the modal verbs were also identified by concordance searches conducted on the corpus, using WordSmith Tools.

The sequence of the research process is presented and summarised in the flow chart in Figure 3.1 The analysis procedures go from top to bottom following the directions indicated by the lines and the arrows.

Ttentify thematic units in each literature review

Code and categorize data according to Kwan’s

Create corpora of each move based on the model identified and convert them into plain text (txt) files

Run the search words in each move corpus with the software WordSmith

Analyse the use of the search words in the corpus

Discuss findings to inform research questions

Examples Of COCInE - - +1 111g HH HH 68

In this section, I will report the procedures for coding moves and strategies I will explain how the combined semantic-functional approaches are applied in the text coding I also perform a pilot analysis on two LR texts (those by Writer T6 and Writer T21) before carrying out the whole corpus The following procedures are used during the coding process.

3.5.1 Gaining a macro view of the theses and the literature review (LR) texts

As discussed in Chapter 2, one primary rhetorical purpose of the LR chapter is to justify the writer’s own study or to make this part of the thesis a foundation for what the writer did in his/her research work As such, before I attempted an analysis of each of the two LR chapters, I tried to gain a macro view of the writer’s research and the thesis by studying its title, abstract, table of content, introduction chapter, some major parts of the methodology, discussion and conclusion chapters Having gained an overall picture of the writer’s work, I read the LR text at least twice to familiarize myself with its propositional contents and how the different parts of the

LR texts are connected thematically and logically This survey of the LR text was much needed for coding at a later stage which required my understanding of the logical development of the text The macro view of the writers’ work gained at this stage provides an important perspective for me to appreciate the integrity of each of the two texts in the process of analyzing them (Bhatia, 2004).

Once the overview of the LR chapters identified, they are examined for further details It is observed that both LR chapters are organized with a separate introduction, a number of thematic units, a conclusion, and a transition I will cite from Writer T21’s and T6’s LR chapters to illustrate this internal structure These

LR chapters open with a brief introduction section, which spans a total of four to six sentences (see Examples 3.1 below), summarize the chapters with a slightly longer text which is two paragraphs long (see Examples 3.2 below) and close with a paragraph which indicates the transition text and direct the reader to the next chapter (see Examples 3.3 below).

This chapter reviews a_number of relevant concepts and studies to the present study Section 1.1 begins with how ESP is defined and the history of ESP development in both global and Vietnamese contexts with typical characteristics and related research trends Then, a number of major foci in ESP including needs analysis, ESP practitioners, distinguishing ESP and EGP teaching methodology and some current ESP teaching approaches are presented in Section 1.2 This is followed by Section 1.3 and Section l4 in which the concept of _teachers’ perceptions, teachers’ practices and influential factors are discussed The next two sections critically review previous studies on the relationship between teachers’ perceptions and practices and with the focus on related teacher cognition research in ESP field The final section summarizes this chapter and highlights the research gap which this study aims to fulfill.

This Chapter is to review the theoretical bases relevant to the present study: genre analysis and the framework of systemic-functional grammar As a way of start, we will take a look at the _notions of genre, and genre analysis made as they are understood by scholars such as Jamieson and Campbell (1978), Miller (1994), Martin (1985, 1990, 1993, 2001), Preston (1986), Swales (1990), Bhatia (1993), Halliday (1993) and many others Then we will discuss the most relevant concepts of the systemic-functional grammar, which provide the theoretical framework for our analyses in the Chapters that follow.

This chapter has presented some theoretical backgrounds related to the study. From the descriptions about how the concept of ESP is defined and its development throughout different stages Besides, an in-depth examination of a number of major foci_in ESP clearly demonstrates the importance of needs analysis in ESP courses, the demands _and_requirements for ESP practitioners as well as the existence _of different ESP teaching approaches , the_review_of such issues provided important backgrounds for collecting and analyzing the data.

Apart from that, a synthesis and critical review of studies regarding the relationship of teachers’ perceptions and practices with the focus on ESP field has pointed the limitations of the previous studies

This Chapter set out to give an overview of genre and genre analysis and the theory of systemic-functional grammar We have outlined the notions of genre, the concept of genre, approaches to_genre_and_ genre definition The relationships between genre and language study have been exploited including the two major approaches to genre that are the formal and the systemic functional

The next chapter will be the presentation of research methodology to the study, in which choices of research approach, research sites, participants, the analytical framework and instrumentation are made with detailed demonstration of the whole data collection and analysis procedures so as to answer the above research questions.

Now that we have finished Chapter I, we turn to Chapter 2 where we will examine the generic structure and the linguistic features of English geological textbooks.

Running between the introductory and concluding sections of the two LR texts are presentations of different various themes! I segmented these presentations into respective thematic units for move analysis according to the discussed topics I speculated that the two writers might have treated each of the thematic sections as a mini literature review of its own This speculation was also informed by the findings in Crooke’s (1986) study in which he observed that each recurring move structure tends to bring about a slightly different theme of the discussion that is related to the writer's research.

I also employed the different meta-discourse to identify the thematic units and their boundaries, which include the section headings in the table of contents, the

LR introductory text, and the formatting features I illustrated these types of meta- discourse again by referring to the first LR chapter written by Writer T21 To identify the thematic units in the chapter, I first studied the introductory text

1t refers to the phenomenon or participants which form the major topic of discussion in a thematic unit.

70 provided at the beginning of the LR chapter (see Examples 2.1 cited above) where Writer T21 announces the six major themes she will discuss in the chapter: development of English for Specific Purposes (ESP), major foci in ESP, teachers’ perceptions, teachers’ practices, studies on language teachers’ perceptions and practices, and studies on ESP teachers’ perceptions and practices (see the underlined parts in Examples 2.1) I then divided the chapter accordingly into the six thematic units of development of English for Specific Purposes (ESP), major foci in ESP, teachers’ perceptions, teachers’ practices, studies on language teachers’ perceptions and practices, and studies on ESP teachers’ perceptions and practices I looked for the headings in the table of contents, and I found the following headings:

1.1 Development of English for Specific Purposes (ESP) 1.1.1 Defining ESP

1.1.2 ESP development as a global trend 1.1.3 ESP development in Vietnamese context.

1.2 Major foci in ESP 1.2.1 Needs Analysis 1.2.2 ESP practitioners 1.2.3 Distinguishing ESP and EGP teaching methodology 1.2.4 ESP teaching approaches

1.3 Teachers’ perceptions 1.3.1 Defining teachers’ perceptions 1.3.2 Factors affecting teachers’ perceptions 1.4 Teachers’ practices

1.4.1 Defining teachers’ practices 1.4.2 Factors affecting teachers’ practices 1.5 Studies on language teachers’ perceptions and practices 1.5.1 Relationship between teachers’ perceptions and practices 1.5.2 Some theoretical frameworks for exploring teachers’ perceptions and practices

1.6 Studies on ESP teachers’ perceptions and practices

Apparently, Writer T21 employs section-numbering systems to indicate different levels of sectioning (e.g., 1.1, 1.1.1, 1.1.2 ), which serve as meta- discourse to which I made use of for identifying thematic unit boundaries.

In addition, in T6’s LR chapter, the writer uses not only section-numbering systems but also the formatting features to show the relation among the headings For example, the bold-faced headings represent the major thematic units of the chapter; the different levels of sectioning (e.g., 1.1.1, 1.1.2 ), suggest that they are subsumed under the major heading of the bold-faced headings.

1.1 Genre and genre analysis 1.1.1 The notion of genre

1.1.2 Genre in folklore studies 1.1.3 Genre in linguistics

1.1.4 Genre in rhetoric 1.2 Approaches to genre 1.3 Genre and language study 1.4 Systemic- functional approach to language study 1.4.1 The metafunctions

1.4.2 The Transitivity system 1.4.3 The Mood system

(Writer T6) 3.5.4 Move-coding and strategy-coding of thematic units

Examples of quantifying a search word in an LR chapter ô‹ 80

In order to identify the reporting verbs in LR chapter T21, the citations must be located first To do so, the T21’s LR chapter was stripped of all graphs, tables, diagrams, etc., and all the texts associated with them, then they were converted to

80 and saved as plain text (.txt) files for running in the software WordSmith Version 7.0 (Scott, 2016), the Concord Tool of which locates all references to the search words containing years, page numbers, third-person pronouns, which is expressed as search string 19??)/20??)/???)/he/she/they It means it is a search for any five- character string beginning with 19??) for a pre-21st century corpus, 20??) for a 21*#- century corpus, ???) for one, two, or three-character string including page number, for example, 158) and p.158), that ends with a closing bracket to catch other forms, and he/she/they string for non-citation types in the chapter, showing them in standard concordance lines with the search word centered and a variable amount of context at either side (see Figure 3.2). ad concord MM

File Edit View Compute Settings Windows Help

2 the Early Years (1962-1981), the Recent Past (1981 -1990), the Modern Era (1990-2011) and the Future 1922)

3 that ESP has undergone namely the Early Years (1962-1981), the Recent Past (1981 -1990), the 1977)

4 ), the Recent Past (1981 -1990), the Modern Era (1990-2011) and the Future (2011 plus) The first 2077)

5 of resources.” (p 94) That is why in Borgs (1997, 2003) framework regarding elements and processes 2077)

6 the factors affecting teacher cognition, Borg (1997, 2003) used a framework illustrating “schematic 2022)

7 instructional behaviors in classrooms (Borg, 2003, 2009) However, research of teacher cognition shows 2022)

8 they from the early 1980s The trend of teaching ESP, in fact, was „imported= to Vietnam by a number of teachers and lecturers at some English-major universities after they were sent to Great Britain, Australia and New

Zealand to study for diploma and masterOs ELT courses As a result, some courses mainly geared towards

9 or materials writers (Long, 2005; Huhta et al., 2013) 2F 761 1.2.2 ESP practitioners The concept of 2077)

10 theteachers themselves accept them (Bailey et al., 1996), it forms the basis of teacher beliefs about their 1822)

11 this period was seen with the devotion of Tarone et al (1981) who brought about two influential approaches in 1922)

12 Dinis da Costa, Araujo, Calvo and Albergaria-Almeida (2015) categorized teachers practices into two main 2022)

13 under a cognitive-ecological perspective, Alzaanin (2014) established the Cognitive-ecological Model 202?)

14 individual Building up the CEM framework, Alzaanin (2014) explained that the term ,cognitiveD was used 2022?)

15 among contextual components Therefore, Alzaanin (2014) suggested that the framework of CEM could 2077)

16 (e.g Khuong & Chi, 2008; Dung, 2011; Dung & Anh, 2013); 1A 73 generally, in most ESP courses, needs 202?)

17 universities and colleges in the country (Dung & Anh, 2010) According to this position, although students 202?)

18 the most prominent parts of EFL teaching (Anthony, 1997) Considered as “an exciting fast-developing 1822) 19 (1988) claims ESP as in contrast with GE, Anthony (1997) observes the vague line between GE and ESP 1892?)

20 & Waters, 1987; Strevens, 1988; Anthony, 1997) While Strevens (1988) claims ESP as in 1977)

21 of the fundamental principles of the subject are; iii) an awareness of how much they probably already 222)

2 do not need to be an expert in a specialist area, they do need to have some awareness and afeelfora they

23 whether the teachers in such context behaved as they said 2A 60 With regards to the corner-stone they

24 efficient learning (p 18) Considering such aspects, they concluded that “ESP is an approach to language they

25 Malvetti (2012) mentioned above, Wu and Badger (2009) combined smoothly the semi-structured 2022)

26 explore ESP teachers perceptions, Wu and Badger (2009), Wu and Hung (2011), Huong and Malvetti (2012 2022)

27 (Richards &Lockhart, 1994; Burns, 1996; Bailey, 1996) However, Borg (2003) critically claimed about 1977)

Figure 3.2: A sample of concordance display with WordSmith’s Concord Tool showing citations in the center in T21.

The search words were then quantified in the WordSmith Tools program, and the results are saved in the Concordance list file in the form of Excel spreadsheets In order to make sure that no citations are omitted, all the proper names of the researchers available in the LR chapter are entered the software one more time to find if there are any further discussions for those researchers. Moreover, each concordance line is also scrolled up and down to scan not only the centre sentences with the search words but also as many surrounding sentences as possible, so that the writer cannot miss any nearby non-citation types.

Using this method, citations were located However, to categorize the citations exactly, manual review and count were still needed to confirm that the citations were counted properly and to eliminate overlaps when one citation has more than one author cited (Example3.4) or statements resembling citations (Example 3.5) illustrated below:

With regards to the corner-stone concept of ESP that is needs analysis, studies of Donesch-Jezo (2011) and Ali_and Salih (2013) displayed interesting results on teachers’ perceptions towards such concepts and its impact on ESP courses and ESP material writings.

The trend of teaching ESP, in fact, was imported to Vietnam by a number of teachers and lecturers at some English-major universities after they were sent to Great Britain, Australia and New Zealand to study for diploma and master ELT courses.

In determining every frequency of citations in the chapter, this study follows the method used by Samraj (2013) This is done by assigning one citation to one rhetorical function or assigning multiple citations which appear together (Example 3.6), or there can be more than one citation location in a sentence (Example 3.7) as shown below:

This point of view, in fact, drew attention from some scholars and researchers (Richards &Lockhart, 1994; Burns, 1996; Bailey, 1996).

This reality was evident in the reviews of many authors on teacher cognition studies such as Clark and Peterson (1986), Carter (1990), Calderhead (1996), Verloop, Van Driel and Meijer (2001) and Borg (2003).

When all the citations are identified with the WordSmith concord tool, they are divided into integral and non-integral citations basing on Swales' (1990) framework The former are citations that the name of the researcher appears in the sentence with an explicit grammatical role that foregrounds the researcher, while the latter are those that are outside the sentence, usually placed within brackets, and which play no explicit grammatical role in the sentence.

Brie (1988) showed that the moon is | Previous research has shown that the made of cheese moon is made of cheese (Brie, 1988)

As the examples above illustrate, only integral citations contain reporting verbs, so integral citations were copied to an excel file for further classification of reporting verbs in terms of functions, tenses, and voice It means that the verbs in citing sentences and those in the further discussions of a previously mentioned author in expressions like "this theory" or "their definition", or "he/she" or "they" referred to previously cited authors were investigated.

In terms of modal verbs, concordance searches were conducted on the corpus, using WordSmith Tools, on the modal verbs (may, might, can, could, will, shall, would, shall, should, ought to, must) Cannot and reduced forms of the negative were also searched for, and it is added to can Occurrences of similar word strings, such as the month ‘May’ were removed from the concordance output, and sO were occurrences in a quoted text (parts of the text that were quoted from other sources) The frequency of occurrence was then quantified for each modal verb in each move in the corpus, entered into a spreadsheet, and these figures were also adjusted to show the average occurrence of the modal per 1,000 words of the text so that comparisons could be made between moves/chapters of differing lengths.

SUMMALY 0A

This chapter describes the research method chosen for the present study and the process of gathering data for analysis Then the research procedure is proposed to provide the frameworks and procedures for move analysis and language analysis.

In this study, the mixed methods research employs the explanatory sequential approach to explore and understand the generic structure and the language features of the thesis literature review written in English by Vietnamese graduates In the first stage, the qualitative data is collected and analysed based on Kwan’s (2006) modified CARS model In the second phase, the quantitative data and analysis are conducted to assist in the interpretation of the qualitative findings Finally, the procedures for coding and quantifying a search word in a thesis literature review are reported as a pilot analysis before carrying out the whole corpus.

THE GENERIC STRUCTURE OF THE PHD THESIS

Introduction and conclusion texts of the LR chapters . -‹- ô+<<ô 85

The study of the LR structure shows that many of the LRs begin with an introductory text and end with a summary text The detailed results and findings of these two functional units in these chapters are presented as follows.

Most of these introductory texts are quite short compared with the length of themes in the chapters, varying from 1 sentence (16 words) to more than 2 pages (747 words) It was found that 70% (21 out of 30 cases) of the LR chapters in the corpus open with an introductory text but only 7 of which were headed with

‘Introduction’ specifically, so the researcher has to base on the location as well as the function of the text to classify them as the introductory text The analysis of the semantic elements of the texts suggests that many of these introductions mainly preview the overall structure of the chapter, and orient the reader toward the aims, the number, and the topics of themes to be discussed in the chapters. These semantic features are illustrated in Examples 4.1 cited below (see the underlined parts).

The chapter provides theoretical background relevant to _The_first_section discusses , followed by the second section about The third section provides The fourth section reviews Next, are given The final section reviews

This chapter reviews_a number of relevant concepts and studies to the present study Section 1.1 begins with Then, a number of are presented in Section 1.2. This_is followed by Section 1.3 and Section 1.4, in which the concept of are discussed The next two_sections critically review previous studies on the _The final_section_summarizes this chapter and highlights the research gap that this study aims to fulfill.

This chapter presents comprehensive theoretical issues related to the field of study.

It starts with, which is followed by and by Finally, the theoretical framework which guides the process of this study is also demonstrated.

This Chapter is to review the theoretical bases relevant to the present study: genre analysis and the framework of systemic-functional grammar As a way of starting, we will take a look at the and the Then we will discuss, which provides the theoretical framework for our analyses in the Chapters that follow.

It was also observed that two writers introduce the chapter by stating either the aim or theme of the chapter that are illustrated respectively in Examples 4.2 and 4.3 below:

This chapter will present the relevant theories and research concerning

This chapter provides an overview of different approaches to and presents my viewpoint on the appropriate approach to study in the context of Vietnamese culture.

The literature review presents the theoretical background related to the definition of English pronunciation, pronunciation from the communication view and from

86 the teaching view, the research on English pronunciation with the focus on linking and assimilation, and the research on pronunciation.

In some of the introductory texts, the themes are previewed in a sophisticated way by applying a rhetorical structure like the CARS model suggested by Swales (1990) This example is demonstrated in Example 4.4 below.

The two concepts important for translation evaluation are Move 1 translation and translation quality Making topic generalization Translation evaluation normally goes together with a theory of Move 2 translation Therefore, different views of translation itself lead to | Indicating a need for different notions of translation quality, and different ways of further research evaluating it.

In this chapter, first, the will be reviewed; and second, the Move 3 and will be discussed in three different schools of thought In Announcing the the third section will be described In the fourth section, will | structure and the be presented in detail themes of chapter

Example 4.4 begins with a brief segment stating the topic of the current research (Move 1) The segment is then followed by a need for further research (Move 2) and then an announcement of the structure and the themes of the chapter (Move 3) The use of the CARS model in writing the introduction text helps present the argument clearly and convincingly (Swales, 1990), and the reader can easily understand the author’s communicative purpose better from the outset.

Table 4.1: The frequency distribution of the realizations of introduction texts

Realizations of introduction texts No of Percentage in the distribution corpus Aim + Theme 13 43.3%

Rhetorical moves (resembling CARS 4 13.3% model)

Table 4.1 displays the frequency distribution of the different realizations of the LR introductory sections Nearly a half of the authors start the LR chapter by mentioning both the aim and the themes at the same time, whereas about 13% of the

87 writers introduce the chapter by using rhetorical moves However, a few only begin the chapter by naming or listing the themes and the aim of the chapter The reason why both the aim and the theme are mentioned in most cases is quite common in the writing process in which the introduction tells the reader what to expect and alerts the reader to the central issue of the section.

It was found that 83.3% (25 out of 30 cases) of the LR chapters in the corpus end with concluding texts mostly marked as ‘summary’, ‘conclusion’, ‘concluding remarks’, or ‘conclusion and theoretical frame’ However, three concluding texts of which are not headed or numbered in separate parts, so they are classified based on the semantic meaning and their function in the chapter as in Example 4.5 below:

In conclusion, although here or there some researchers have dealt with hedging, most of the studies in written language are on separate linguistic items considered as hedges There is a gap in studying hedging devices in scientific discourses, especially in economic discourses Besides that, there has not been any comparison about hedging between English and Vietnamese, as well as hedging functions in the two languages This is the gap this thesis must fill in.

The length of the concluding texts also varies, with the longest spanning more than two pages (853 words) while the shortest occupies only 4 sentences (73 words) The concluding text primarily summarizes the gist of the chapters or reiterates the purposes of reviewing the themes in the chapters as illustrated in Examples 4.6 below:

Thematic rhetorical structures in the body parts of the LR chapters

It is observed that the body of the LR chapters in the corpus is made of several thematic units which discuss a particular aspect of the writer’s research topic The result shows that132 thematic units were identified in the body parts of

30 LR chapters It means that on average each LR chapter consists of 4.4 themes It is also found that many of the thematic unit's model features the three moves in the CARS model and in particular those proposed by Kwan (2006) for PhD thesis LR texts This structure resembles Bitchener’s (2010) when he states that a literature review is organised with ‘a series of themes or topic units’ in the body (p 61).

The frequency distribution of the three moves shown in Table 4.3 indicates that no move appears in the thematic units 100% of the time, suggesting that none of them are obligatory In terms of the move frequency distribution identified in the body parts of these LR chapters, there are high occurrence rates of Move 1 (Establishing one part of the territory) and Move 2 (Creating a research niche) It means that the PhD writers prefer to include Move | and Move 2 in their LR chapters much more than Move 3 (Occupying the research niche).

Table 4.3: Frequency counts of the three moves in the LR chapters

Move Individual counts Number of units with the move

It is, however, worth considering why Move 3 is least preferred by the Vietnamese writers in the LR texts One possible explanation is that Move 3 can be realized elsewhere, for example, in the chapter conclusion sections as an overall response to the review carried out in the chapter as discussed in Section 3.1.2 or even in another chapter (e.g., the research methodology chapter) at points where it is more appropriate to occupy the niches created (e.g when introducing the research procedures conducted in the writer's work) The presence of Move 3 in different parts of LR texts is an important point for writers to bear in mind.

Move combinations in the thematic units reveal a variety of schematic patterns, and many of the patterns are also highly recursive as illustrated in Table 4.4 and Figure 4.1 below.

Table 4.4: The configurations of moves in the thematic units in the corpus

* ‘n’ stands for the number of counts - ‘n’ times, (1-2)3 means 1-2-1-2-1-2

The specific frequency distribution of the move patterns is illustrated in Figure 4.1, indicating that the most common pattern is the pairing of Move | and Move 2 which is expressed in the formula (1-2)n, in which 'n' refers to the number of times the pairing recurs It is followed by the patterns of irregular 3-move, (1-

2)n-1, and Move | only The statistics above also show there are not any obligatory sequential patterns made up of three moves because none of the patterns occur 100% of the time in the thematic units.

It is noted that Move I and Move 2 tend to be combined in a relatively regular 1—2 order that is identified in the patterns of (1—2)n, (1—2)n-1, (1—2)n-3, and 1-2(1-3)n in 57.6% of the units In fact, 12 of the 21 irregular-3move units also have a high concentration of pairings of Move 1 and Move 2 (1-2) such as 1-2-1-2-3-2, 1-2-1-2-3-2, 1-2-1-2-1-3-1, which outnumber 50% higher than the other pairings (1- 3-2-3-1 or 2-3-1-3) within the same unit These frequency counts do not qualify the 1-2 patterns as obligatory, though they do imply that Vietnamese writers often set the context for their research, provide necessary background on the discussed topic, and establish a niche in the existing research.

Although 29.5% of the themes in Vietnamese PhD literature review chapters contain all of the three moves, relatively few of them (only 6 out of 132 themes or

4.5%) have sequencing of Move 1, Move 2, and Move 3 as suggested in original CARS model.

Figure 4.1: Frequency distribution of move patterns across thematic units in the corpus (Total thematic units: 132)

* ‘n’ stands for the number of counts - ‘n’ times, (1-2)3 means 1-2-]-2-1-2

It is also noted that the highly frequent occurrences of Move 1 alone in the present corpus show that these Vietnamese PhD students are trying to display their familiarity with the knowledge in their field, but not arguing for their studies to be situated into the body of knowledge In other words, by just referring to previous research (Move 1) without synthesizing or indicating the relevance of the reviewed literature (Move 2) for the niche to be established (Move 3), the literature reviewed by these Vietnamese writers tended to be less effective and logical It is, therefore, necessary for these Vietnamese PhD students to be made aware of how this chapter should be effectively written.

Besides the move configurations, the findings on the body parts of the LR chapters in this PhD thesis corpus also revealed some interesting information First, forty-four thematic units (33.33%) begin their parts with an advanced introduction of the sub-themes and their organization (Example 4.10 below) This opening part

94 at the beginning of these themes is similar to the introductory section at the very beginning of an LR chapter because it provides a justification and subthemes to be reviewed in the accompanying texts Furthermore, nearly a similar number of thematic units (43 out of 132) end with the concluding segments which summarize the review of the themes (Example 4.11 below) This could be attributed to the three-part structure of an essay (Introduction-Body-Conclusion) that these Vietnamese English major students have been trained at university (Ha, 2011).

1.1 Concept of translation in three different approaches Translation is defined differently depending on the theoretical approach that one takes In this section, the concept of translation will be reviewed in three different schools of thought, namely literature-oriented approaches, skopos-related approaches and linguistics based approaches The discussion on the concept of translation will centre on the two issues: the relationship between the source text (ST) and the target text (TT), and relationship between the features in the text itself and how they are perceived by human agents.

To summarize, in this section, the concept ‘translation’ has been discussed in three different schools of thoughts Firstly, in the literature-oriented approaches, a translation is viewed as a text in its own right, and the original text is of secondary importance Secondly, in the Skopos-related approaches, a text is considered to be its author’s ‘offer of information’ to its potential recipient, and a source text and its translation may move apart depending on the purpose of translation Thirdly, from the perspectives of linguistically oriented approaches, translation is regarded as inter-lingual transfer This study adopts House’s (2001a) definition: translation is the substitute of a text in one language by a semantically and pragmatically equivalent text in another language.

This study also identified a new strategy within a thematic unit, which is not reported in Kwan (2006) That is the strategy of indicating transition or movement to the new theme This new strategy is found in twenty thematic units (15.15%) in

95 the corpus, and it is always accompanied by a theme concluding text that is illustrated in Examples 4.12 and 4.13 below:

After discussing some basic background of news language, approaches from which news has been examined, and previous studies related to news, the author would like to move on to a detailed discussion of the theoretical background related to the study (Theme concluding text)

In the following section, first of all, the dissertation will focus on reviewing the literature on cognitive linguistics and then, the specific models used in this research This will provide a basis for the elaboration of the analytical framework of the study (Theme transitional text) (4.13) [26]

Move structures in the thematic units of the LR chapters

The study reveals that the thematic units in the body parts of LR chapters in the corpus comprise a series of moves and strategies as suggested by Kwan (2006) and Bitchener (2010) I will go into detail about the moves and strategies identified in the corpus as follow.

4.3.1 Move 1: Establishing one part of the territory of one's research

The majority of the thematic units start with segments that establish particular themes related to the writer’s research These segments fall roughly into three categories, each of which displays semantic and functional characteristics that correspond to those in a respective strategy of Move 1 in Kwan’s (2006) model.

4.3.1.1 Surveying the non-research-related phenomena (Strategy 1A)

Based on the function and semantic analysis of the surveys, Strategy 1A mainly mentions 3 typical types of knowledge: definitions and explanations of terms, theories, non-research practices, and non-research phenomena.

Definitions and explanations of terms

Some of the texts in Move 1 introduce the definitions or explanations of key terms in the writers’ studies One example of defining and explaining key terms can be found in Example 4.14 cited below, which is taken from one main thematic unit of the LR by Writer T5 whose thesis examined the conference interpretation in the Vietnamese context from a pragmatic perspective It can be seen that the text presents an explanation of the notion of 'interpreting' and the distinction between the term ‘interpreting’ and 'translation'.

1.1.1.1 Interpreting vs translation According to most common dictionaries, ‘interpreting’ is usually equated with ‘oral translation’ or, more precisely, with the ‘oral rendering of spoken messages’. However, following this definition would exclude interpreting in signed (rather than spoken) languages from our purview, and would make it difficult to account for the less typical manifestations of interpreting, such as sight translation or on- line (written) translation of Internet chats occasionally used in the European Commission, among others Instead, by elaborating on the feature of immediacy, one can distinguish interpreting from other forms of Translation without resorting to the dichotomy of oral vs written.

Kade (1968) defined interpreting as a form of Translation in which: (a) the source- language text is presented only once and thus cannot be reviewed or replayed, and (b) the target-language text is produced under time pressure, with little chance for correction and revision.

A more detailed explanation of key terms can be found in Example 4.15 by Writer T30, in which she reviews the different dimensions of the key term

‘conversation’ used in her thesis.

2.1.1.1 The concept of conversation Conversation is regarded as one of the most popular and effective forms of communication The concept of conversation has been approached and discussed broadly as well as understood variedly by numerous linguists The variety in the ways of defining conversation is because it is approached from different perspectives with different purposes For example, Hornby et al (1988) defined conversation broadly as the exchange of language through language or Bublitz (1988, p 151) regarded conversations as entities which exist, develop and proceed in time and constantly grow and change More particularly, Collins (2002) claimed that conversation is a friendly and natural talk in which participants exchange information, ideas, and emotion to one another Regarding conversation as a form of “sociability”, Have (1999, pp 3-4) regarded conversation as any activity of interactive talk, regardless of its purpose or just for talking This definition implies that any forms of interaction occurring when people talk with each other can be a conversation Conversation, in this definition, can be performed via telephone, online chats, and so on.

However, in this study, conversation is understood following Goffman’s (1966) definition According to Goffman, a conversation is a kind of interaction occurring only when participants get close together and directly cooperate to sustain a single focus of attention with the exclusion of non-present parties and inclusion of co- present parties (p 24) It can be inferred from this definition that only direct interaction is regarded as conversation and telephone talks, online chats or email exchanges are not Classified as a form of conversation The choice of Goffman’s (1966) definition on conversation is because the present study focuses on face-to- face or direct interaction rather than other channels of communication.

The definitions/explanations identified in the corpus are given for a variety of purposes In some cases, they are provided for explaining purposes, as obvious in the case of Example 4.14 cited above, to help the reader understand the terms and

98 concepts that will be used in the thesis In other cases, the definitions are introduced in the text to justify the choice of a specific definition or interpretation among the confusing ones as what is observed in Example 4.15 written by Writer T30 The reader can clearly see the deliberation of Writer T30 in her selection and presentation of the characterizing of the five definitions of ‘conversation’ to critique the first four while justifying his choice of the fifth one.

Definitions or explanations of terms and concepts mentioned in the text can also serve as a specification of a set of the terms and concepts studied in some parts of the writer’s research That is to say, the writer systematically collects and evaluates phenomena that cannot be observed directly For instance, the detailed explanation of the concept of conversation is cited in Example 4.15 Writer T30 surveys at the beginning of her chapter to review previous literature on the same concept to discern the most relevant variables for her throughout the study Such was also the case in the LR written by Writer T21 (see Example 4.16) The descriptions of the six roles to the definition of ESP practitioners not only make clear to the reader what the concept of ESP practitioners mean and provide details to justify the theories of ESP practice that ESP teachers should also be equipped in order to carry out the job in ESP teaching profession.

1.2.2 ESP practitioners The concept of “ESP practitioners” was first introduced by Swales (1988) with the purpose to emphasize other responsibilities of ESP teachers besides their teaching role Regarding the roles of an ESP practitioner, Dudley-Evans and St John (1998) identified five key roles namely a teacher, a course designer and materials provider, a collaborator, a researcher and an evaluator According to the authors, as in the role of a teacher, an ESP practitioner is suggested to “draw on students’ knowledge of the content in order to generate genuine communication in the classroom” (p 13) and to negotiate with students in suitable ways to reach lesson goals Besides, an ESP teacher in many cases should be a course designer and provide materials by selecting, adapting and even writing materials to suit the learners’ needs By the role of a collaborator, an ESP practitioner should be

99 interested in working with content departments, subject matter teachers and specialists of the field The fourth role refers to the ability to carry out research on learners’ needs analysis, target situation analysis and on genre analysis or the discourse of the text types that are related to the discipline Another important role of an evaluator is also mentioned referring to the assessment on students’ learning, the evaluation of the courses and teaching materials.

Creswell (2013), in his examination of the connection between traditions of inquiry and research design, suggests that it is important to consider how one

‘frames the study within the philosophical and theoretical perspectives’ in the beginning stage of design The LR texts in the corpus come with accounts of theoretical perspectives of the various types which Creswell refers to For instance, in Example 4.17, Writer TI devotes almost two pages to reviewing theories of hedges used in the field to account for a complete understanding of hedges in different contexts Some parts of the discussion of the theories are cited in Example 4.17 below.

2.1.3 Toward a Working Definition of Hedging Hedging, as mentioned above, can be broadly defined as a genre-specific, interactional, multi-functional phenomenon which integrates semantic, pragmatic, social and cognitive factors Depending on the genre, hedging is used to meet certain macro-level expectations, or to accomplish certain micro-level intentions, but always attends to particular norms and limitations Genre also decides certain type of hedging strategies employed, functions of hedging and certain lexical items used to recognize a hedge Semantically, hedging helps identify the typical statement component which is often signaled by characteristic linguistic items which serve to carry out specific discourse strategies Pragmatically, hedging assists the writer/speaker to obtain both macro- and micro-level purposes simultaneously Socially, hedging serves to meet the demand and expectation of a certain discourse community A good way to understand hedging concept is to consider all these elements of the phenomenon In other words, the definition of hedging will be incomplete or even incomprehensive if these elements are not

Generic modeling for a thesis literature review written in English by

In this chapter, I have presented the findings generated from the analysis of

30 LR texts Here, I will summarize some of the major findings and relate them to research question 1 in this study:

What is the generic structure of the thesis LR chapters written in English by Vietnamese PhD students?

Most of the LR chapters are named 'Literature Review’, Review of the Literature’, 'Theoretical Background and Literature Review', 'A Review of ' or the name of the discussed topic Most of the LR chapters follow a potential generic Introduction-Body-Conclusion structure That is, the LR chapter is structured with an introductory move and a concluding move, and between them are theme cycles. The introductory section of the chapter serves two possible purposes: providing an advanced organizer to facilitate reading and justifying the themes pursued in the chapter The concluding part carries a summarizing statement providing a brief overview of the themes reviewed in the chapter and the intention of the review The majority of conclusion texts were found to display features of Swales' CARS model, especially Move 2 and Move 3, which respond to claims made in the chapter and at the same time announce the writers' research In two cases where there is no marked concluding text to the LRs, the final thematic units of the LR texts function as the overall Move 2 and Move 3 of the LRs Both of the final thematic units are titled

‘Theoretical framework of the study' or 'Contrastive framework’, suggesting that one primary aim of the two LRs is to establish the theoretical frameworks for the writers’ research Since there are fewer chapter introductory texts than chapter conclusion texts, the structure of (Introduction)-Body-Conclusion is proposed to represent the schematic organization of the LR chapters.

In the body of the LRs, discussions are divided into thematic units systematically numbered and marked by section heads The multi-thematic sectioning in the LRs reveals that the rhetorical exposition presented in the genre can be sophisticated and complicated Multi-themes in the LRs are arguably needed to establish complicated conceptual and available literature outlines of the writers' research topics.

4.4.2 Move structures in thematic units

4.4.2.1 Presence of Kwan’s (2006) CARS-elements

The thematic sections in the LR chapters of the present corpus display different types of text units which find corresponding semantic and functional

129 matches with the majority of the strategies in each of the three moves of the revised CARS model posited by Kwan (2006) for the thesis literature review chapters This result also confirms that they have the same genre.

Several thematic units begin their parts with a theme introduction and end with the theme concluding segments which summarize the review of the themes and theme transition text to forward the reader to the next part of the discussed topic Each thematic unit is constructed with (Introduction)-Moves-(Conclusion)-(Transition). This could be attributed to the three-part structure of an essay (Introduction-Body- Conclusion) that these Vietnamese English major students have been trained at university (Ha, 2011) and the rules of good writing which is made up of a clear organizational structure with an introduction, conclusion, and transitions between sections suggested by Swales and Freak (2009) and Ridley (2012).

Moreover, this study identified two other new strategies in the move of creating a research niche, which is not reported in Kwan (2006), that is, the strategy of concluding a part and the strategy of indicating a transition to a new part within

4.4.2.3 Non-obligatory status of moves and move elements

Since none of the three moves and their respective elements occurs in 100% of the thematic units nor do they occur in any fixed pattern, they can only be given the optional status while the move elements can only be treated as strategies as they are named in Kwan’s (2006) study Nonetheless, some of the moves and strategies appear more frequently, and they are thus considered to be moves and strategies of strong preference (e.g., Move 1 and 2, Strategies 1A (surveying the non-research- related phenomena or knowledge claims), IC (surveying the research-related phenomena), 2A (Counter-claiming), and 2F (Concluding a part in a theme)).

4.4.2.4 A scheme posited for the move pattern per thematic unit

A move structure is postulated below to describe the generic structure of the Vietnamese PhD LR chapters (see Table 4.11) with newly identified strategies in italics The strategies in their respective moves are renamed according to strategy- numbering systems to avoid being mistaken between different revised CARS

130 models The numbering system means that the first number indicates the number of the move; it is separated from the numeral that indicates the descending order of frequency of occurrence (i.e., the strength of preference) The arrangement, therefore, does not imply any sequential structuring.

Table 4.11: A generic structure posited for a thesis LR written in English by

Vietnamese doctoral candidates in social sciences research

Move 1 Establishing one part of the territory of one’s own 97.73% research by:

Strategy 1.1 | surveying the non-research-related phenomena or knowledge 54.80% claims Strategy 1.2 | surveying the research-related phenomena 54.24%

Move 2 Creating a research niche(in response to Move 1) by: 78.79 %

Strategy 2.2 | Concluding a part in a theme 26.45%

Strategy 2.3 | Synthesizing the theoretical framework/position 15.58%

Strategy 2.6 | Indicating transition to a new part in a theme 12.68%

Move 3 Occupying the research niche(in response to Move 2) by: 38.64%

Strategy 3.1 | Announcing the adoption of terms or definitions of terms 38.46%

Strategy 3.2 | Announcing aims of investigation 30.77%

Strategy 3.3 | Announcing the theoretical position or the theoretical 23.08% framework Strategy 3.4 | Announcing the research methodology 12.31%

*: ‘n’ stands for recurring for ‘n’ times of theme

The moves tend to be realized in a variety of recursive manners with the 2 most common configurations being (1-2)n and (1-2)n-1 Recursion also occurs at the strategy level (e.g., 1.1/1.2, or 3.2/3.3) in Move 1 and Move 3 but does so in much less predictable patterns; strategy recursion occurs rarely in Move 2.

In summary, this chapter attempts to investigate how Vietnamese PhD students composed the LR chapters of their theses in English Regarding the overall text divisions, all 30 LR chapters in the current corpus were separated from other chapters in the theses This result can be attributed to the guidelines provided by the universities with the PhD programs in Vietnam In addition, it is found that the LR chapter structure is mostly organized in the form of an Introduction-Body- Conclusion-Transition pattern, which is similar to those written by native authors (Kwan 2006) The introductory texts employed in the LRs serve two purposes: to provide the aim and to preview the themes to be organized and discussed in the chapter Most of the conclusions carry brief reviews of the themes, often accompanied by transitional texts that signal the readers to the upcoming chapter and improve the coherence between chapters.

It is also concluded that the body of all LR chapters is divided into several thematic sections which are made up of different moves and strategies In terms of moves, Move | and Move 2 were conventional, and Move 3 was optional in the body texts of the LR chapters of the PhD theses Regarding the strategies, Strategies 1.1 (surveying non-research-related phenomena) and 1.2 (surveying research activities) were conventional while all other strategies in the three Moves were optional Moreover, Strategies 1.1 and 1.2 occupied the largest textual space of these LR chapters while the other steps were a few lines long With regards to move sequences, the regular sequences of Moves 1-2 were the most frequently used, followed by the irregular pattern of Moves 1-2-3, the regular Moves (1-2)n-1, and Move 1 alone Infrequent use of Move 3 and the frequent occurrences of Move 1 alone are likely to show that these Vietnamese postgraduate students were trying to display their familiarity with the knowledge of their discipline without synthesizing the reviewed literature.

THE TYPICAL LEXICOGRAMMATICAL FEATURES OF

Reporting verb (RV) classification in the present COFDUS

Based on the Hyland’s (2002) framework of process categories of RVs, the classification of reporting verbs in the corpus can be analysed and illustrated in Table 5.3.

Table 5.3: Distribution of reporting verbs in process categories in the corpus

As can be seen from Table 5.3, a total of 1973 RVs were identified in the body parts of LR chapters The figure shows that on average there are 65.75 RVs per chapter, and the density of RVs (per 1,000 words) is 5.16 This density matches nearly perfectly with those in LR chapters written by Vietnamese MA TESOL students (Nguyen and Pramoolsook, 2015), in which 5.33 RVs were found per 1,000 words (p 203) The analysis of all reporting verbs in the present corpus revealed a preference for reporting information as Discourse Acts, with 53.23% of

135 all the cases in this category Most of the remaining instances were reported as Research Acts, and only 7.00% as Cognition Acts The statistics of RVs use in PhD theses written by Vietnamese graduates confirm the trend of using RVs in applied linguistics research articles by Hyland (2002) with the most popular type is the Discourse Acts (59%), and he claims that the prominent use of Discourse act verbs in the humanities and social sciences is considered suitable in an argument schema relating to the explicit interpretation, speculation, and complexity as “accepted aspects of knowledge” (p 126).

As can be seen in Table 5.4, the majority of Research Act verbs are procedure verbs, which occupy 21.69% of the total cases in the corpus, and finding verbs account for about 18% This proportion may be inferred that procedural aspects of previous studies and their findings were both discussed almost equally in Vietnamese PhD LR chapters.

Table 5.4: Overall distribution of reporting verbs in the corpus

It is also noticeable to find the prominent use of non-factive verbs, which neutrally report on the research findings and inform readers of the author’s positions

> The first figure indicates the number of instances of the RVs and the figure in brackets expresses the percentage of this type used in the current corpus.

136 on the cited claims in these LR chapters They were found in 33.95% of the total RVs It could be better understood concerning the communicative functions of these chapters that these Vietnamese writers had in mind when composing them In fact, in LR chapters, reviewing non-related research (1A), which provided a general discussion of a theme, without focusing on specific aspects of previous studies such as subjects, features more than half of the total corpus (54.80%) In this part, writers often display their familiarity with their field knowledge such as definitions and explanations of terms, and theories, non-research practices, and non-research phenomena, but do not establish the credibility of their studies The non-factive RV preference by these Vietnamese PhD thesis writers in the LR chapters could be due to these writers’ assumptions about the communicative functions of the LR chapters, which simply review previous research without arguing for the need for their studies to be conducted.

As the following examples show, procedure verbs carry no evaluation in themselves but simply report research tasks neutrally (see Examples 5.1a), and non- factive verbs comment on research findings with no clear attitudinal signal as to their reliability (see Examples 5.1b):

Bondi (1999) only examined the nature of academic persuasion in two genres She developed a framework for the analysis of

Lindeberg (1994) studied rhetorical conventions in the various disciplines

Doan, T.T.H (2000) investigated meaning of idioms expressing epistemic contents in Vietnamese.

Bloor and Pindi (1987) used hedging as a tool to draw schematic structure of economic forecasts.

Claxton and Murrell (1987) analyze learning styles at four levels

Wiseman (2002) analyses the reason why people laugh in

Bell and Attardo (2010: 430) identify 7 levels of failed humour of ESL learners

For instance, Smoak (2003), from her experience as an ESP practitioner, found that “ESP is English instruction based on actual and immediate needs of learners who have to successfully perform real-life tasks unrelated to merely passing an English class or exam”

It is also interesting to identify the very rare presence of critical verbs (in Cognition acts and Discourse acts) or counter verbs (in Discourse acts) In fact, among 1973 reporting verbs there were only eight instances of critical verbs in the form of negation (not accept, not equip with, not establish, not make clear distinctions, not make point ) (see Example 5.2a below) and sixteen examples of counter verbs (ignore, deny, refute, question and criticize/ critique ) were found (see Example 5.2b below).

However, not all scholars have received politeness function of hedging in the same way Hyland (1996) did not accept the politeness function of hedging in RAs because according to him, politeness phenomenon is a way to deny peer critiques which are essential for maintaining standards and constructing reputations.

Nonetheless, Reif (1971) and Newmark (1988) do not establish specific procedures of comparing between ST and TT.

However, the examination of activities and behaviors performed in this sequence reveals that this sequence shares many similarities with the pre-closing sequence. Especially, on introducing these twelve strategies for closing sequences, Okamoto (1990) herself could not make a clear distinction between activities employed in the pre-closing sequence and ones employed in the leave-taking sequence.

The speech act theory has, however, been criticized by a number of scholars, including Levinson (1983) and Geis (1995), for its insufficiency as a basic framework for exploring the pragmatics of verbal interactions

Firstly, Fish (1981, p 59-64) criticizes House s model (1977) for the inflexibility of one-to-one matching of structure and meaning.

In addition, these critical and counter verbs are often accompanied by contrast and concession transitions such as however, nonetheless to supplement opposite ideas in an effective way (see Examples 5.5 a, b) That the writers always use negative forms (not) to express their disagreement or dispute with the author as euphemistic language may be inferred that these writers avoid criticizing the mentioned authors directly and explicitly In addition, the avoidance of counter verbs could be due to these thesis writers' awareness of the power inequality between themselves and examiners (Koutsantoni, 2006) and partly reflects Vietnamese cultures in which all forms of confrontation should be avoided as taught by the philosophies of Taoism and Confucianism (Nguyen, 2002).

It can also be observed in Table 5.3 that about 12.21% of tentative RVs report both the authors’ tentative views (Cognition tentative) (Examples 5.3a) and the writers’ tentative uncertainty (Discourse tentative) towards the reported messages (Examples 5.3b).

Swales (1990), for example, believes that the term ‘genre’ has been maladroitly used and genre remains a fuzzy concept, a somewhat loose term of art (Cognition Tentative)

Omar (1992), in his study on conversational opening in Kiswahili, supposed verbal recognition like calling out somebody's name as the first sequence in an opening section (Cognition Tentative)

Cohen (2003), in agreement with these statements, suggests that learning will be facilitated by making the learners aware of range of strategies from which they can choose during the language learning and use (Discourse Tentative)

For example, Posteguillo (1999) reporting on the results of a move analysis indicated the schematic structure of computer science RAs (Discourse Tentative)

Gardner and Lambert (1959) hypothesized that a social motivation involving ‘a willingness to be valued members of the L2 community’ would result in high levels of L2 achievement (p.271) (Discourse Tentative)

In addition, a small number of Cognition positive and neutral RVs were also found in these LR chapters (2.58% and 2.94%, respectively) (see Examples 5.4a and 5.4b, respectively).

Robinson (1991) also agrees that learners’ present learning needs and target situation needs should be considered with equal weight and simultaneously in syllabus design (Cognition positive)

Thus, similar to Sornig (1977) and Rees-Miller (2000), Wierzbicka thinks that a disagreement is a response to a stimulus which is a prior utterance from the first interlocutor (Cognition positive)

Similarly, Markkanen & Schréder (1989; 1992) see hedging as a means to accomplish the writer’s intention, a modifier of the writer’s responsibility for the truth value of the propositions expressed or as a modifier of the weightiness of the information given, or the attitude of the writer to the information (Cognition neutral)

Ladefoged (1975: 92) regards that words can be affected when they occur in connected speech (Cognition neutral)

The preference uses of positive, tentative, and neutral RVs employed in the current corpus is likely to indicate their preference for ascribing a tentative view to the cited authors or characterizing themselves as holding a tentative stance when they were uncertain about the cited materials.

Reporting verb forms - -s- 5 + TT HH Tu nh HH Hư 141

Table 5.5: Reporting verbs in Vietnamese PhD LR chapters

6 argue (63) 23 divide, focus, indicate, compare, discuss (16)

10 study (53) 27 criticize, confirm, stress, categorize (12)

28 make, make a distinction, contribute, declare, illustrate, suppose, denote, note draw attention, insist (11)

29 refer, employ, differentiate, review, include, collect, discover, group, reveal, contend, maintain, see, clarify, report (6)

30 depict, adopt, organize, pay attention to, make use of, demonstrate, list, base, survey on, take, postulate, recommend, highlight, cite, deal with, have analyse about (4)

31 advance, adapt, address, express, specify, characterize, outline, apply, utilize, put forward, take the view of, create, generate, revise, comprise, establish, think, support, prove, distinguish, offer, bring about, combine, set out, administer, realize, posit, admit, judge, affirm, acknowledge, advance, pinpoint, supplement, admit, judge, examine (23)

14 use, affirm, do research on, make a contribution to, complain, emphasize (25) conceptualize, synthesize (2)

15 conclude, 32 stipulate, inform, delineate, appear, deploy, go deep into, probe, share (24) come to conclusion, come up with, deliberate, lay out, rate, ask, edit,

16 carry out, concentrate on, talk of, applaud, select, correspond, raise a concern about, broaden, represent, correspond, devote, contrast, discriminate, have an overview of, involve, complain, gather, relate, separate, design, understand, hold, conceptualize, take, open, attitude, interpret, not accept, recognize, appeal, evaluate, imply, not equipped with, not establish, not make a clear distinction, reduce, recall, reckon, make a model, make a study, hypothesize, synthesize, question, pose the questions of, ignore, guide, sequence, coin, acquire, pioneer, catch, modify, uncover, start, formulate, measure, constitute, construct, take advantage of, fulfill, announce, inherit, prioritize, refine, label, govern, influence, publish, recall, reckon, do a study on, have research on, have a study about, make a study, research, take, have more emphasise on (1)

Table 5.5 shows that the most common verbs in the corpus were the discourse acts state (155 cases) and claim (87), suggest (78), define (72), followed

3 The number in brackets indicates the times of verb occurrences in the PhD LR chapters written in English by Vietnamese learners

141 by the research verbs conduct (66), and the discourse verb argue (63) Other verbs belonging to the top ten verbs in the corpus are the research act verb find out (60) and then the discourse verbs propose (57) and point out (53) and the research act verb study (53) The most commonly used verbs in the three mentioned corpora are shown in Table 5.6 below.

Table 5.6: Top ten reporting verbs in Vietnamese PhD LR chapters

No | Reporting Verbs | Total number | Percentage

It is noted that the most common RV in Vietnamese corpus is state, which neutrally informs readers of the author’s position The preference use of this non- factive verb one more time affirms the claim that confrontation should be avoided in Vietnamese cultures (Nguyen, 2002).

In addition to these ten most common RVs, which constitute over a third of all cases, the other 207 verbs (see Appendix 7) (among 217 verbs) had a low frequency of occurrences and more than half occurred once or twice in the corpus The analysis of reporting verbs has shown that the total number of 214 reporting verb forms in the Vietnamese corpus is equal to only half of Hyland’s (2002) study identifying more than

400 different reporting verbs to introduce citations (p.124) Such a wide range of reporting verbs from the native writers causes the selection of a particular reporting verb to be rather delicate and challenging for the L2 writers The use of limited ranges of reporting verbs among Vietnamese writers reveals that Vietnamese writers have not been fully familiar with different rhetorical functions and that they often face many challenges in their attempts to comply with the conventions and negotiate their claims effectively through reference to other sources.

Since some of the RVs found in this corpus have not been included in Hyland’s (2002) classifications, they are added into his category, depending on their denotative and evaluative meanings (Appendix 8) However, these added words are limited to their occurrences found in this corpus only.

In terms of Research acts, besides verbs with clear semantic denotations (Finding factive: demonstrate, establish, show, solve, confirm; Finding non-factive: find, identify, observe, obtain; Procedures: conduct, examine, study, evaluate, analyse, compare, use, investigate), the other verbs are added to their corresponding groups (Appendix 8) with the following illustrated examples:

The most common referenced perspective of teacher cognition in the recent studies in this decade has been contributed by Borg (2006).

Hofstede et al (2010) proved that in individualist societies, the relationship between employers and employees is just a business transaction or work-orientated relationship.

The normative approach is supported by Japanese researchers, especially Hill et al (1986), Matsumoto (1988) and Ide (1989)

Hewings (1990) also focused on specific items of language (mainly nouns) which assist economic writers to transfer from the real world to a hypothetical or idealized world in their writing.

Discussing language in ESP, Dudley-Evans and St John (1998) have paid more attention to the questions of grammar, vocabulary and discourse, together with developments of genre analysis.

The underlined RVs in Examples 5.5 are classified as Finding factive verbs in this corpus because the writers would like to show their acceptance of the authors’ achievements or conclusions.

New Finding non-factive verbs found in this corpus are illustrated in Examples 5.6a, and 5.6b below These verbs comment on research findings with no clear attitudinal signal as to their reliability (see Example 5.9a), but some of them occur only once in the total corpus such as caught, acquire, uncover, sequence, modify, pioneer (see Example 5.9b) It is challenging to allocate these verbs into finding non-factive verbs because of their multi-meaning such as the verb catch, uncover, SO it is necessary to consider the context in which these verbs occurred.

Akindele (2007) classified greetings into different types such as

Chamot (1987, cited in Wenden and Rubin, 1987, p.77) introduces twenty-two strategies which can be used in reading

For instance, Johnson (1994) observed that “when pre-service teachers

, Welo (2009) caught the thought of students that English jokes could improve their reading comprehension skills, but they failed to find solutions to the problems or the difficulty for better reading comprehension.

Drawing on Hutchinson and Water’s view, Chen (2009) acquires two important points.

By using factor analysis, Clement and Kruidenier (1983), Kruidenier and Clement (1986) and Ely (1986) uncovered different dimensions of language learning motivation: instrumental and integrative motivation.

Skehan (1996) also sequenced tasks in three steps known as forms of task-based teaching (p 39).

In terms of the Research Procedure RVs, as compared with those classified by Hyland (2002), several new verbs were employed by this group of Vietnamese writers to refer to the procedural aspects of previous researchers’ investigations (see

Example 5.6c) The writers only report the research procedures neutrally without evaluation.

Especially, Kellermann et al (1991) made use of a survey questionnaire to gather the data for analysis.

Following the work of Black and William (1998a/b), Hattie (1999), after synthesizing 500 meta-analyses of 180,000 studies related to assessment purposes and strategies, developed a model of feedback to enhance learning.

Tenses and voice of reporting Verbs - - ss HH Hiệp 145

In addition to the denotative and evaluative functions of RVs, tenses, and voice of RVs also play a part in showing the writer's opinion towards the reported information (Swales, 1990) This claim is also confirmed by Shaw (1992) when he suggests that in reporting, tense-alternation is often studied together with voice.

The result of tenses of RVs used in Vietnamese PhD literature review chapters is shown in Figure 5.1 below.

Figure 5.1: Tenses of RVs in the corpus: percentage

It can be seen in Figure 5.1 that the three main tenses used in the corpus are present simple, present perfect, or past simple tenses, though the proportions of tenses are quite different among the reporting verbs This result is in line with Shaw’s (1992) suggestions that ‘sentences reporting others' findings and including a reporting verb can have a variety of forms such as the present simple, present perfect, or past simple tenses’ (p 302).

The majority of the RVs (49.18% of the total) are in the simple present tense, and 41.30% are in the simple past tense Only 3.13% of the total RVs are in the present perfect tense, and the rest, 6.39%, are in other tenses The present tense and the simple past tense are the most frequently used tenses in the corpus The reasons may be due to the special features of the genre characterizing this chapter In the three moves of LRs, Move | is for the writer to survey the topic generalizations, and claim the centrality of the research topic In this move, the writer uses the simple present tense of the reporting verbs which can mark his/her general comments on a subject under investigation (see Examples 5.7), and at the same time, he makes generalizations about the topic (see Examples 5.8).

Besides, Buck (2001) states it is also important that the test tasks engage the listeners in using those competences.

Bachman (2004) suggests that it is essential that we be able to estimate their effects on test scores in order to use these scores for the purpose for which they are intended.

Buck (2001) states that in a listening test, it seems reasonable not to penalize mistakes if the response is intelligible and clear since after all, we want to know whether test-takers have understood a text, not whether they can produce correct language.

Hambleton and Pitoniak (2006) state that it is common to select panels that demonstrate balance in terms of gender, ethnicity, geography .They also state that panelists’ knowledge about the targeted subject area is the most important criterion in panel section.

Chiaro (1992) proposes three elements, namely the linguistic, the socio-cultural and the poetic that a speaker should have to appreciate humour in English jokes.

Dalton & Seidthofer (1995, 3) define pronunciation in general terms as

4Move 1: Establishing one part of the territory of one’s own research; Move 2: Creating a research niche;

Move 3: Occupying the research niche

In Move 1, the writer also reviews the previous studies on a subject, and at the same time, shows his attitudes towards these studies It is a convention that the writer employs the simple past tense in reporting the previous studies because by the time he writes his LR, the information he has about the subject has become part of established knowledge (see Examples 5.9).

Brown, Lake, and Matters (2011) administered a questionnaire to examine 784 primary school teachers’ beliefs about the purposes of their assessments in Queensland, Australia The findings indicated that the teachers preferred assessments serving formative purposes of improving teaching and learning to assessment serving summative purposes of accountability and certification.

Gibbs and Lucas (1997) found that large class size prevent teachers from implementing authentic and performance-based assessment, which leads to their beliefs in disregarding such types of assessment in their classroom.

Simon-Vandenbergen (1996) found that positive politeness hedges are used by politicians when they have to answer face-threatening questions.

Nittono (2003) mentioned to effects of social factors in using positive politeness hedges such as the social status of the speaker in relation to the listener, or the topic for discussion.

Hyland (1996) did not accept the politeness function of hedging in RAs because according to him, politeness phenomenon is a way to deny peer critiques which are essential for maintaining standards and constructing reputations Hyland said that a failure to consider politeness would not leave the writer unprotected because the choice of linguistic form in science partly shows the writer's responsibility and respect for the readers.

In more details, Holmes and Stubbe (2015) claimed that small talk in the workplace functions like knitting.

Thus, he may acknowledge this by referring to it in the simple present tense (see Examples 5.10)

Halliday (1994) describes the three metafunctions of language as follows

Based on the expressing functions, Tú ( 1983) divides Vietnamese idioms into three categories

Thus, Wierzbicka (2013, p 307) complains that the concept of asking pain has become an “immeasurable” issue in doctor-patient interaction While Fermandez and Wasan (2010, p 450) find out that “pain” only ranks as “the fifth vital sign” of language that occurs during doctor-patient interaction

In Move 2, the writer indicates research gaps or counter-claiming after he comments on the previous studies of a subject in detail This element is based on Move 1, but different from them in that the writer’s focus here is on research gaps, flaws, drawbacks or weaknesses of the previous studies In doing so, the writer may choose the simple present tense of the Critical verbs or the negative forms of other reporting verbs (see Examples 5.11).

It can also be said from this difference that setting goals and purpose of reading is considered important in Oxford (1990) system while O'Malley and Chamot (1990) ignore this.

However, Reif (1971) does not establish specific criteria for the analysis of ST and TT.

Nonetheless, ReiB (1971) and Newmark (1988) do not establish specific procedures of comparing between ST and TT.

Especially, on introducing these twelve strategies for closing sequences, Okamoto (1990) herself could not make a clear distinction between activities employed in the pre-closing sequence and ones employed in the leave-taking sequence.

In announcing the present research (Move 3), the writer employs also uses the simple present tense or the present future tense of reporting verbs (see Examples (5.12).

Sharing similar viewpoints with the authors (Nida, 1964; Newmark, 1988; Toury, 1995; Venuti, 1995; Nord, 1997), House (1997, 2015) adopts the terms of overt and covert translation to signify “SL-oriented translation” and “TL-oriented translation”.

Hardin (2001, 155) shows that positive politeness contributes to the force of the advertising message because it allows the advertiser and the consumer to share the same desire

In general, the simple present tense of reporting verbs is the most commonly used tense in LRs In addition, the simple past tense signals the methods used and the results obtained in a given experiment or research project in Move 2 and Move

Modal verbs in Vietnamese PhD LR chapt€rs -. - 55+ +ss<++s++ss2 152

Table 5.7: Occurrence of the modal verbs in the corpus n = " = 3 = 3 3 s |e |e |B IEE (Ble |S Š§ '-

| || || || P| all can may should will could might would must

Figure 5.3: Occurrence of the modal verbs per 1,000 words All occurrences of 3121 modal verbs including can, could, may, might, must, should, ought to, will, shall, would, and must were counted in the raw number in the corpus of the total of 382,047 words (see Table 5.7), and the density of model verbs (per 1000 words) is 8.17 (see Figure 5.4) Such statistics have been done without considering their functions in the texts to find out merely the general pattern and distribution of modal auxiliaries in the three moves of the LR texts Counting and recording the total number of the frequencies of modal verbs in the discourse, the author calculated their relative frequency per one thousand words so that this data can be compared with another corpus of unequal sizes of samples.

The simple aggregates in Figure 5.3 indicate that Vietnamese thesis writers use modal verbs quite frequently, with more than eight for every 1,000 words of

152 text which is higher than that rate in the thesis Agricultural Botany (natural science) corpus (nearly 8 for every 1,000 words of text), but it is a bit lower than that of Agricultural and Food Economics thesis corpus of over 12 for every 1,000 words of text (Thompson, 2001, p.152) This data may reveal that social science disciplines contain more modal verbs than natural science writings The two most frequent modal verbs employed by Vietnamese thesis writers are can and may The dominant use of modal verbs can, may in PhD texts may carry a preliminary hypothesis that

PhD writers make more predictions, possibilities, and hedged claims.

In order to identify the proportion of each auxiliary in the corpus, the percentage of every auxiliary out of the total number of auxiliaries was computed (see Figure 5.5) Because the number of ought to and shall is too small, so this study only focuses on 08 core modal verbs can, could, may, might, will, would, should, and must The low frequency of shall in academic texts was also confirmed by Biber et al (1999). would, 4.2_ Must, 3.88 might, 4.33 could, 4.58

Figure 5.4: Percentage of modal verbs in the corpus

As the pie chart in Figure 5.4 demonstrates, there is no balance in the use of modal verbs The most frequent modal verb is can with 45.34 % occurrences At the other extreme, the modals must (3.88%), would (4.2%), might (4.33 %), and could

(4.58%) were found to be significantly underused in comparison to the other modal

153 verbs in LR texts The remaining modal verbs including may (15 %), should (12.34

%), and will (10.29 %) were located in the middle.

Table 5.8 shows the relative frequencies of use of each of the 8 modals in the corpus As can be seen, there is considerable variation in the density of modal verb use, with the heaviest use appearing in LR numbers T26, and T16 (about 11 modal verbs per 1,000 words) while LR T13 contains only 4.2 modal verbs per 1,000 words Table 5.8 also marks variation among different LR texts within this corpus.

LR T26, for example, makes far greater use of can and should than other theses, while LR T30 makes proportionately higher use of may This can be explained by reference to the nature of the research being done in each case: LR T26 explores cut scores of an English test and is concerned with validity, different validation approaches, and standard-setting Therefore, the author uses can indicate deontic permission for what can and what cannot be applied through different approaches and should to indicate epistemic necessity; LR T30 is an investigation of the linguistic, pragmatic, and cultural knowledge in making communication, and an assessment of extensive theories, which requires lengthy discussion of possible outcomes Therefore, may is used significantly to moderate writers’ assertions or statements.

Table 5.8: Relative occurrence of 8 modal verbs per 1,000 words in the corpus

LR_|_may | might | can | could | will_| would | should | must |_ all

Table 5.8 also shows the relative occurrence of the same modal verbs amongst the LR sections It can be seen, then, that there is variation between the LRs This can be attributed to the differing rhetorical purposes of the writers, and also to their research orientations Research that is strong on theory and which has strictly controlled variables is capable of making stronger predictions, with more use of will and can, for example, than research work, like that in T26, which is conducted in the field, and which has to consider complex interactions between variables and make greater use of hypothetical and qualified statements.

It must be acknowledged, too, that another factor at play in the variation between theses is that of different personal backgrounds LR T25 has a remarkably low use of modal auxiliaries at 4.32 per 1,000 words, with can, must, and may accounting for the majority of these instances The research investigated the learning strategies affected by different factors from the learners The data of the research was collected through a questionnaire and semi-structured interviews The writer may be, therefore, requested to make adjustments to the strength of her conclusions for an academic audience, and also to speculate on possible future directions for research in the area.

LR T26, by contrast, has approximately four times the density of modal verbs per 1,000 words, with can, should, and will appearing quite frequently, especially will is used much more than any other LRs though it is not the most three common modal verbs in the corpus It can be inferred that the author uses will as one of the boosting devices to create trust and certainty in readers and establish an impression of confidence, and assurance in her research about the cut score of a test.

Variation is again clearly evidenced with much greater use of can in T26, T8,and T7, for example, or of should in T26 and T21, or may in T15 In other words,the frequency of occurrence of particular modal auxiliaries can differ considerably from one LR to another.

Modal use in different rhetorical MOVES 5 Ăn 1x re 156 b9 20100 in

Another aspect of variation that can be investigated is that of modal use within different moves of an LR Table 5.9 shows the dispersion of modals across the conventional rhetorical moves of an LR chapter, as found in Thesis T20 This

LR chapter is typical as it follows a simple Move 1-Move 2-Move 3 pattern of organization, with each theme following this cyclical pattern Therefore, it provides a clear illustration of the dispersion of modal verb use across rhetorical moves.

Table 5.9: Modal verbs per 1,000 words in each move in Thesis T20

T20 Movel | /1000 | Move2 | /1000 | Move3 | /1000 | Whole | /1000 can 40 3.68 6 7.56 9 4.64 55 4.04 could 4 0.37 0 0 3 1.55 7 0.51 may 13 1.20 4 5.04 2 1.03 19 1.40 might 0 0 2 2.52 1 0.52 3 0.22 must 3 0.28 0 0 0 0 3 0.22 should 19 1.75 0 0 4 2.06 23 1.69 will 10 0.92 0 0 3 1.55 13 0.96 would 3 0.28 1 1.26 0 0 4 0.29

As can be seen from Table 5.9, the density of modal verbs in Move 2 is the highest with 16.39 modal verbs per every 1000 words of the text This density is quite reasonable because Move 2 conveys evaluative elements so using modal verbs to indicate the author’s stance (see Example 5.15).

It is, equally, clear that Borg’s category cannot incorporate the situation in this case; the reason may be assumingly possible that Borg might get used to modern systems of education like the USA, Hong Kong, the UK, Canada, and Australia, where teachers persist the right to act according to their thoughts

Borg might not have sufficient information about the top-down educational policies in other parts of the world.

The use of modal verbs decreases significantly over Move 1 (8.47), which includes what is known about the subject already and claims the centrality of the theme reviewed, and Move 3 (11.33), in which the author reveals the solution to help fill the gap, answer the research questions, or continue the research tradition that has been presented in Move 2 Can, may, and should are the three most common modal verbs and the popularity of these three verbs is in line with the general trend of the whole corpus According to Lorenzo (2008), these modal verbs (can, may, and should) are categorized as hedging devices, which is the representation of authors’ tentativeness, and it is also one of the main features of academic writing in which assertions are seldom made with an objective appraisal of their reliability (Hyland, 2006).

Table 5.10: Occurrence of the eight modal verbs per 1,000 words in the corpus

Modals | Move 1 | /1000 | Move 2 | /1000 | Move 3 | /1000 | Whole /1000 can 1188 3.89 167 3.98 60 3.65 1415 3.70 could 117 0.38 17 0.41 9 0.55 143 0.37 may 397 1.30 62 1.48 9 0.55 468 1.22 might 109 0.36 21 0.50 5 0.30 135 0.35 must 109 0.36 10 0.24 2 0.12 121 0.32 should 304 1.00 63 1.50 18 1.10 385 1.01 will 251 0.82 49 1.17 21 1.28 321 0.84 would 101 0.33 24 0.57 6 0.37 131 0.34 all 2576 8.44 413 9.85 130 7.92 3119 8.17

To determine the use of modals in different rhetorical moves in greater detail, I then extracted all the moves of the LR chapters that were labeled Move 1 (establishing one part of the territory), Move 2 (creating a research niche), and Move 3 (occupying the research niche) and made a small sub-corpus of them Table 5.10 shows the frequency of use of the 8 modal auxiliaries in three sub-corpora. Densities of use, as might be expected, are quite similar for the modals in all three moves, in which Move 2 is higher than they are in the other two.

5.2.2.1 Modal verbs in establishing a niche (Move 2)

The slightly higher use in the Move 2 texts in comparison with other two may be due to the tendency for some writers to criticize weakness or inadequacies of the previous studies or to relate deficits of knowledge, assert the relevancy or confirmative claims in Move 1 (Swales, 1990) In other words, in this move, the writers seek to find a gap in the existing literature in the field under review in order to justify their current research, or they indicate their acceptance of the citations mentioned in Move 1 In so doing, the author may be led to express disagreement, criticism, or affirmative strategies in such a way as to show respect for fellow researchers As shown in Table 5.10, the writers rarely used the modal verbs must, preferring can, should, and may to realize the second move ‘creating a research niche’ Should is much more popular in Move 2 than the rest.

Consequently, although it can be practically applicable to some empirical studies, especially of English-speaking societies, to a certain extent, it can hardly be a universal mode.

Due to very limited preparation time and resources, ESP teachers in Vietnam can hardly follow the complex steps for designing

Swales argued that such classifications have impressive intellectual credentials and considerable organizing power However, the propensity for early categorization can lead to a failure to understand particular discourses on their own terms.

However, there have been criticisms about House s model (1997) that should be taken into account.

In Examples 5.16 and 5.17, the positive features of the alternative are conceded in the first half of the sentence, but the words although, however signal the change to the conclusive drawback of the method A modal auxiliary verb can be used in a directly concessive sense, where can modifies the statement of potential benefit, which is then countered by the opposing negative A modal auxiliary verb should also be employed in Example 5.10 as a necessity to consider different theory model evaluations.

The modal verb may in the following examples (5.18) in Move 2 expresses hedges when the writers make negative comments on the discussed topic or problems relating to the existing research.

Due to different types of learning styles, teaching methods that are used by instructors may vary Some instructors prefer giving lectures at classroom; other may focus more on rules, some use demonstration, while some prefer memorization As a result, the mismatch between the individual’s learning style and the instructor's teaching style may lead to failure of learners.

However, in examining different types of conversations, other researchers have realized that in this sequence, different enquiries and responses are produced rather than just the question of “how are you?” According to Taleghani-Nikazm (2002), this sequence may be lengthy in some languages but rather short in others.

However, these studies have some issues within the scope of an M.A thesis This shortage may make it difficult for ESP teachers and course designers to improve their practice in this increasingly vital area in ELT.

The phrases may + lead to/ negative adjectives in the above examples are used as hedged statements (Hyland, 1999) ‘Hedging’ is taken here with the modal verb may to mean to diminish “writers’ responsibility,’ and to “shift the66. responsibility for inferring contextual meanings to the reader” (Hinkel, 2005) after gaps or problems are pointed out with phrases such as due to, the mismatch,

159 however, be lengthy, rather short, shortage Such hedging can be used as a defensive strategy; the writer qualifies a statement to reduce the strength of a claim, and prevent the claim from being vulnerable to a strong challenge In a thesis, where the writer is being assessed by examiners, this is an important concern.

The above hedges are clear examples of the epistemic sense of the modal verbs, where the uncertainty is identified with the writer They are in line with Palmer’s (1990) description of the epistemic and deontic functions of the modal verbs as being discourse-oriented, an indication of the writer/speaker’s attitude towards the subject of the discourse Another use of modal verbs in discourse- oriented statements is in meta-discourse, which will be discussed next.

CONCLUSION ng HH nh ht 171

Pedagogical implications .- c5 2113311189101 1191119111 11 1n ng nếp 175

In this section, I will discuss how some of the findings generated in the study can be turned into useful insights for students, supervisors, and EAP instructors. First, I will review how the schematic patterns of LRs as identified in the study can be applied in PhD LR writing instruction Then, I will suggest designing writing courses that help novice non-native writers to be familiar with typical linguistic resources used in the LR chapter of a PhD thesis.

6.2.1 How the schematic patterns of LRs can be applied in PhD LR writing instruction

The main pedagogical implication of this study relates to the teaching and introduction of the genre of Vietnamese PhD LR chapters to PhD candidates, Vietnamese novice researchers, and scholars from different disciplines Given that the social science PhD literature review constitutes a coherent genre and that students and novice researchers need to be able to control this genre, it is important to consider how information about the PhD literature review genre can be integrated into teaching Therefore, the common approaches used for the teaching of the genre will be reviewed as follows.

6.2.1.1 Introducing the organization of PhD LR chapter

PhD candidates should be informed on how they can organize their literature review chapters because the generic structure proposed by this study will provide a template for non-native speakers of English to arrange the rhetorical moves in good orders Specifically, PhD candidates need to be aware of the following structural characteristics of LRs:

PhD candidates should be instructed on the introduction-body-conclusion pattern in organizing their PhD LR chapters They should be advised to produce introductory texts that preview the overall organization of the chapter and orient the

175 reader toward the aims and themes to cover in the chapters They should also be informed to state either the aim or the structure of the chapter They should also be advised that the introductory texts can display a rhetorical movement like the CARS model suggested by Swales (1990), which may include the topic of the current research (Move 1), a need for further research (Move 2), then an announcement of the structure and the themes of the chapter (Move 3).

PhD candidates should also be informed that the body of the LR can be divided into major thematic units, each of which discusses one particular aspect of the writer’s research topic They should also be shown that several thematic units begin their parts with an introduction and end with the concluding segments which summarize the review of the themes and transition text to forward the reader to the next part of the discussed topic Therefore, each theme in the body of LR chapters can be constructed with Introduction-Moves-(Conclusion)-(Transition) Moreover, PhD candidates should be aware that in the body of each theme are located the strategies that review a part of a theme (the strategy of concluding a part of) and which indicate movement to a new area within a theme (indicating the transition to a new part).

PhD candidates should also be advised that the theme body can be structured based on the schematic pattern worked out in this study (see Figure 5.1) The very common occurrence of pattern Move | in the present corpus shows that the writers mostly focus on their familiarity with the knowledge in their field However, the lower frequency of Move 2 and 3 in the present research indicates that the writers do not always claim the niche by synthesizing or indicating the relevance of the reviewed literature (Move 2) and rarely fill this niche (Move 3) in the same theme. Therefore, it is suggested that Move | should always be accompanied by Move 2 and Move 3 so that the reader can understand how the research topic and evaluation may eventually result in one crucial aspect of the writer's research.

Moreover, PhD candidates should also be guided through different strategies they can apply to realize the moves in the themes by introducing typical discourse markers and semantic features characterizing these strategies in each move They need to be informed about the different possibilities of combining these strategies, regardless of any sequential pattern.

Finally, the schematic structure of the LR chapter proposed in this study is worked out in the fields of social sciences and humanities, such as English linguistics and English language teaching Cautions should be taken to apply this model in other disciplines, as disciplinary backgrounds can affect the organization of LRs and the way to use citations (Swales and Lindemann, 2002) Therefore, PhD candidates can be trained to practice simple move analysis of LRs in theses so that they can use the model as a starting reference to guide the analysis in the disciplines of their research work.

As stated by Swales (1990), each type of academic writing is a highly conventionalized genre with distinctive lexical, lexicogrammatical, and rhetorical features that show variance depending on the disciplines, tasks, and registers. However, there have been complaints about literature review (LR) texts: 'Not sufficiently theme-based, 'not structured according to the issues’, ‘insufficiently informed by the research hypotheses’, 'merely a list, 'boringly chronological’, or

‘just describes each piece of research one by one without adequate linkage’ (Swales

& Lindemann, 2002, p 107) In fact, several studies have been carried out to investigate the structure of LR sections, but most of them are done for the neighboring genres such as research articles, and undergraduate or master dissertations, which are much shorter and simpler than what is found in theses (Starfield, 2003) In some studies, it is supposed that the structure of the LR chapter is like that of the introduction chapter, so the writers use the introduction's model to examine the schematic pattern of LR chapters (Swales, 1990; Bunton, 2002).

The difficulties encountered by non-native writers may be more serious because they not only have to prove to have a sufficiently critical perspective and knowledge of the theories, concepts, and current state of knowledge in that area of research but also advanced English proficiency Moreover, the findings from this study show that the schematic pattern of LR chapters can be very complicated and confusing Therefore, academic writers should be instructed to write literature reviews by being exposed to authentic LR texts, considering them in their thesis

177 entirety, which will help students appreciate the complexity, sophistication, and deliberation involved in the organization of LR chapters These sample texts can provide a rich resource of language input, accompanied by the instructions of rhetorical moves characterized by a set of features, which can greatly aid novice researchers in being well aware of the schematic structure of the chapter Finally, writing instructors are advised to be mindful to provide students with samples in different disciplines and have them discover if there are any alternative discipline-specific patterns This can be implemented by training students to analyse the simple and typical moves of LR texts in theses.

6.2.2 Writing courses for novice non-native writers in English

This study reveals that writing skill is a complex construct including a wide range of knowledge and processes, and there are two levels of understanding and writing a thesis literature review: generic structure and linguistic resources to operationalize a communicative purpose or a rhetorical move Therefore, apart from the application of schematic patterns in writing LR sections, the writer needs to be aware of the required knowledge of a particular genre through formal training in their postgraduate programs The curriculum and instruction programs should provide explicit genre instruction on rhetorical moves and linguistic features to help shape non-native English writers' knowledge of writing for their thesis Based on the problems relating to linguistics features identified in this study, the writing course should include the following issues:

- Teaching reporting verbs by explaining the various functions of RVs, raising their awareness of strategic choices in using RVs, the rhetorical feature of tense and voice of RVs.

- Teaching different citation types and functions by explaining the citation importance and conventions in academic texts, especially in LR texts, demonstrating the rhetorical functions of citations, describing the classification of citations; getting the writer to use wide diversity of citations to avoid repetition The thesis writer should be aware that learning and applying a greater variety of citation practices is definitely a step towards becoming more competent and successful writers.

Limitations of the Study - c1 11111 HH TH TH gệp 179

Although this study has brought about some significant insights into the investigation of the genre of LR chapters in PhD theses written by Vietnamese researchers, during the implementation of the research, it might not avoid some limitations.

Firstly, regarding the data collection tools, as stated clearly in Section 3.4, the researcher planned to collect as many PhD theses as possible, but due to the limited number of theses available in the library and the copyright policy, it is very time-consuming to get the data To create the corpus, the LR chapters of these theses were copied, scanned, converted to, and saved as plain text files for running in the software In addition to that, the researcher spent quite a lot of time familiarizing herself with the use of the software.

Secondly, like many other qualitative studies, the findings of this study might not be used for generalization because it aims to explore the phenomenon in specific

179 contexts with specific characteristics However, with the transparent procedures of data collection and analysis presented in Chapter 3 of this thesis, the applicability and transferability of the research process and the findings in some related contexts are possible.

Thirdly, despite several efforts to avoid subjective bias in the position of a qualitative researcher, the researcher’s experience as a PhD student might impose certain personal ideas on the interpretation of the data.

Finally, since LR chapters in PhD theses are complex and dynamic, the findings of this study might not fully cover all aspects related to the procedure for genre analysis of such academic texts.

Future research recommendations - - - 5 + + £+kE+vkEkeeeeessereeeree 180 W) v0 90)00ỡỡù10i1 c1 nh e

Taking the aforementioned limitations of this study into consideration, some future research avenues are suggested, as follows:

First, extending the scope of the study to other chapters in a PhD thesis, such as the Introduction, the Methodology, the Findings, Discussion, or the Conclusion chapter, might help to deepen understanding of the genre analysis of a PhD thesis. Moreover, the selection of theses from different universities might bring about more interesting findings regarding the impacts of different contextual factors.

Secondly, interviewing in person with the PhD writers and their supervisors should be conducted so that the interview data may provide a deeper understanding of the practice of writing LR chapters in a PhD thesis as well as enrich the findings of this research.

Thirdly, more lexicogrammatical features in the LR texts might be analysed. Such studies will help to characterize the overall linguistic features of such academic and sophisticated texts.

Fourthly, linguistic resources in academic writing are a rich and promising field worth further investigation This study dealt with advanced academic writing only Therefore, one potential area that requires further research is using linguistic features among Vietnamese writers on different academic levels For example, the

180 results from this study could be compared with those of less demanding groups, like those at the master's or undergraduate levels.

Fifth, it is also a good suggestion for genre analysis of PhD theses written in the Vietnamese language to identify if there are any differences in terms of genre between the English corpus and the Vietnamese corpus.

Finally, analyses of generic structure and lexicogrammatical problems made by Vietnamese writers in the LR texts might be another possibility These studies will help the researcher to work out some gaps in the literature and the writer to avoid poor standards of language and presentation in their theses.

As interest in facilitating the PhD writing process grows, renewed attention is being directed to the genre analysis of the literature review chapter for its potential to enhance our understanding of how a thesis literature review works.

Initially, it took me quite a lot of time to read instruction books for thesis writing, but I was rather confused because there was considerable variation in expectations across disciplines After a while, I decided to conduct this research to explore what characterises the genre of a thesis literature review written in English by Vietnamese PhD candidates The objectives were to investigate the macrostructure and the linguistic characterisation used to address the communicative purposes of this chapter.

The LR texts analyzed in this study were drawn from 30 accepted PhD theses that I could access at a university in Vietnam in the fields of social science and humanities, such as English linguistics, applied linguistics, translation, cross- cultural studies, and English language teaching methodology The choice of the five disciplinary areas could be considered representative scientific discourse written in English by Vietnamese graduates in the social sciences In addition, I was more familiar with these five disciplinary areas than with other disciplines from my previous education and working experiences, and knowledge of the thesis genre in these disciplines is very useful for my teaching at my university The choice of

181 exploratory sequential mixed method research design is also particularly relevant for this study because it integrates the qualitative and quantitative data analyses in a sequence of phases to create a more reliable model.

During the research process, several interesting findings, as mentioned previously, satisfied the aim of this study The study demonstrates that the combination of move analysis and linguistic analysis can enhance our understanding of how a thesis literature review is structured both at the macro level (the generic structure) and the micro level (linguistic features) The present study also shows that although the rhetorical move and strategy organisation in the suggested generic structure for a thesis LR written in English by Vietnamese learners can be broadly understood to resemble Kwan’s modified CARS model, this model needs to be adapted to include some new moves and strategies to meet the requirements of a conventional thesis literature review and educational organisations This study’s findings contribute to the existing literature on genre analysis It is hoped that this study may help to raise PhD students’ awareness of the genre of PhD thesis literature reviews, ease the research process for novice researchers who are seeking advice on how to write literature reviews properly, and benefit researchers, language teachers, and learners in several ways.

1 Hoang Thi Ngọc Diệp (2018) A genre analysis of research article introductions in Maritime industry in English by Vietanmese writers 2018 International Graduate Research Symposium Proceddings Vietnam National University Press ISSN: 978 60462 60974, pp 99-113)

2 Hoang Thi Ngoc Diép (2020) Citations in Literature Review chapters of doctoral theses written in English by Vietnamese postgraduates of Applied Linguistics 2020 International Graduate Research Symposium Proceddings. Vietnam National University Press ISSN: 978 604315 1619, pp 145-156)

6 6 AdgẤAÃ 184 9s 0))i0 0n Ầ.Ắ I APPENDIX 1: Boote & Beile’s (2005) Literature Review Scoring Rubric I APPENDIX 2: Reporting verbs classification by Hyland (2002) - IH APPENDIX 3: The source of the PhD literature review chapters written in English

Vietnamese Đại học Quốc gia HàNội (2016) Tài liệu hướng dẫn nghiên cứu sinh khóa QH.2016-2.

Ahmad, U K (1997) Research article introductions in Malay: rhetoric in an emerging research community In Anna Duszak(ed) Culture and Styles of Academic Discourse Berlin: Mouton de Gruyter, 273-301.

Ahmad, U K (1997) Scientific Research Articles in Malay: A Situated Discourse

Analysis, a PhD Dissertation in the University of Michigan, Michigan: UMI Publication.

Anthony, L (2012) AntConc (Version 3.3.5) [Computer Software] Tokyo, Japan:

Waseda University Available from http://www.antlab.sci.waseda.ac.jp/.

Ary, D., Jacobs, L.C., Razavieh, A., & Sorensen, C (2006) Introduction to research in education (7th ed.) Fort Worth, TX: Harcourt Brace College Publishers.

Bawarshi, A S &Reiff, M J (2010) Genre: An introduction to History, Theory,

Research and Pedagogy West Lafayette, IN: Parlor Press.

Bazerman, C (1988) Shaping Written Knowledge: The Genre and Activity of the

Experimental Article in Science Madison: University of Wisconsin Press.

Bazerman, C (1994) Systems of genres and the enactment of social intentions In

A Freedman and P Medway Genre and the New Rhetoric London, Taylor and Francis.

Bazerman, C (2003) Intertextuality: How texts rely on other texts In C Bazerman and P Prior, What Writing Does and How It Does It: An Introduction to Analysing Texts and Textual Practices London/ New Jersey: Lawrence Erlbaum.

Berkenkotter, C., & Huckin, T N (1995) Genre Knowledge in Disciplinary

Communication: Cognition/Culture/Power Lawrence Erlbaum Associates, Inc. Bhatia, V K (1993) Analyzing Genres: Language Use in Professional Settings.

Bhatia, V K (1997) Genre-mixing in academic introductions In: English for

Bhatia, V K (2004) Worlds of Written Discourse: A Genre-Based View London:

Bhatia, V K (2014) A generic view of academic discourse Academic Discourse.

Biber, D., Connor, U., and Upton , T A (eds.) (2009) Discourse on the Move:

Using Corpus Analysis to Describe Discourse Structure Amsterdam: John Benjamins.

Biber, D., Conrad, S., and Leech, G (2002) Longman Student Grammar of Spoken and Written English Harlow: Pearson Education.

Biber, D., S Johansson, et al (1999) The Longman Grammar of Spoken and

Written English Harlow: Pearson Education.

Bitchener, J & Basturkmen, H (2006) Perceptions of the difficulties of postgraduate L2 thesis students writing the discussion section Journal of English for Academic Purposes 5(1):4-18 DOI: 10.1016/j.jeap.2005.10.002

Bitchener, J (2010) Writing an Applied Linguistics Thesis or Dissertation: A Guide to Presenting Empirical Research New York: Palgrave Macmillan.

Bitchener, J., & Turner, E (2011) Assessing the effectiveness of one approach to the teaching of thematic unit construction of literature reviews Assessing Writing, 16(2), 123-136 https://doi.org/10.1016/j.asw.2011.02.002

Blaxter, L., Hughes, C and Tight, M (2010) How to Research (4th edition).

Boote, D., & Belle, P (2005) Scholars before researchers: On the centrality of the dissertation literature review in research preparation Educational Researcher, 34(6), 3-15.

Brewer, J D and Hunter, A (2006) Foundations of Multi-method Research:

Synthesizing styles Thousand Oaks, CA: Sage Publications.

Bruce, C.S (1994) Research students’ early experiences of the dissertation literature review Studies in Higher Education, 19 (2): 217-29.

Bruce, N (1989) The roles of analysis and the conceptual matrix in a process approach to teaching academic study and communication skills In V Bickley (ed.), Language Teaching and Learning Styles Within and Across Cultures (pp. 236-65) Hong Kong: Institute of Language in Education.

Bunton, D (2002) Generic moves in Ph.D thesis introductions In J Flowerdew

(Ed.), Academic Discourse (pp 57-75) London: Pearson Education Limited.

Bunton, D R (1998) Linguistic and Textual Problems in PhD and M Phil Theses:

An Analysis of Genre Moves and Metatext (Thesis) University of Hong Kong,

Pokfulam, Hong Kong SAR Retrieved from http://dx.doi.org/10.5353/th_b3123814

Chen, M (2009) Tense of reporting in dissertation literature reviews Journal of

Cambridge Studies 4(2), 139-150 Retrieved from http://journal.acscam.org.uk/data/archive/2009/200902-article13.pdf.

Chen, T.-Y., & Kuo, C.-H (2012) A genre-based analysis of the information structure of master's theses in applied linguistics The Asian ESP Journal, 6(1), 24-52.

Christie, F (1999) Genre theory and ESL teaching: A systemic functional perspective TESOL Quarterly 33.4: 759-63.

Christie, F (2002) Classroom Discourse Analysis: A Functional

Cohen, L., Manion, L and Morrison, K (2013) Research Methods in Education.

Connor, U., & Upton, T A (2004) The genre of a grant proposal: A corpus linguistic analysis In Connor, U., & Upton, T A (eds.) Discourse in the Professions: Perspectives from Corpus Linguistics Amsterdam/ Philadelphia: John Benjamins, pp 235-256 DOI: https://doi.org/10.1075/scl.16.10con

Cooper, H (1988) The structure of knowledge synthesis Knowledge in Society

Cresswell, J (2009) Research Design: Qualitative, Quantitative, and Mixed

Methods Approaches (3rd ed.) Los Angeles: SAGE.

Creswell, J (2015) Educational Research Planning, Conducting, and Evaluating

Quantitative and Qualitative Research New York: Pearson.

Creswell, J W & Clark V L P (2018) Designing and Conducting Mixed Methods

Creswell, J W & Clark, V L P (2011) Designing and Conducting Mixed

Methods Research 2nd ed Thousand Oaks, CA: SAGE Publications

Creswell, J W & Creswell, J D (2018) Research Design: Qualitative, Quantitative, and Mixed Methods Approaches 5th Edition Thousand Oaks, CA: Sage.

Creswell, J W (2003) Research Design: Qualitative, Quantitative, and Mixed

Methods Approaches (2nd ed.) Thousand Oaks, CA: Sage.

Creswell, J W (2005) Educational Research: Planning, Conducting, and

Evaluating Quantitative and Qualitative Research Upper Saddle River, New Jersey: Pearson Education, Inc.

Creswell, J W (2009) Research Design Qualitative, Quantitative, and Mixed

Methods Approach 3rd Edition Thousand Oaks, CA: SAGE Publications

Creswell, J W (2012) Educational Research Planning, Conducting, and

Evaluating Quantitative and Qualitative Research (4th ed.) Boston: MA Pearson.

Creswell, J W (2013) Research Design: Qualitative, Quantitative, and Mixed

Methods Approaches (4th ed.) Thousand Oaks, CA: Sage.

Creswell, J.W (2014) Research design: Qualitative and quantitative approaches

(4th ed.) Thousand Oaks, CA: Sage Publications.

Denzin, N.K and Lincoln Y.S (2008) Strategies of Qualitative Inquiry, (3th ed.).

Devitt, A.J (1991) Intertextuality in tax accounting: Generic, referential, and functional In C Bazerman& J Paradis (eds.), Textual Dynamics of the Professions: Historical and Contemporary Studies of Writing in Professional Communities (335-357) Madison WI: University of Wisconsin Press. DOI: 10.1016/j.jeap.2011.03.004

Dong, Y R (1998) Non-native graduate students' thesis/dissertation writing in science: Self-reports by students and their advisors from two U.S institutions. English for Specific Purposes 17: 369-90.

Dudley Evan, T (1986) Genre Analysis: An Investigation of the Introduction and

Discussion Sections of MSc Dissertations, in M Coulthard (ed) Talking about Text, Birmingham U.K English Language Research Birmingham University.

Dudley-Evan, T (1999) The dissertation: a case of neglect? In P Thompson (ed.)

Issues in EAP Writing Research and Instruction University of Reading: CALS. 28-36.

Eastwood, J (1994) Oxford Guide to English Grammar Oxford University Press.

Flowerdew, J & Peacock, M (2001) Issues in English for academic purposes In J.

Flowerdew and M Peacock (eds), Research Perspectives on English for Academic Purposes, pp.8-24 Cambridge University Press.

Flowerdew, J (1993) An educational, or process, approach to the teaching of professional genres ELT Journal, 47, 305-316.

Flowerdew, J (2013) Discourse in English Language Education New York:

Freedman, A (1994) “Do as I say”: The relationship between teaching and learning new genres In A Freedman & P Medway (Eds.), Genre and the New Rhetoric, pp 191-210 London: Taylor & Francis.

Freedman, A., & Medway, P (Eds.) (1994) Genre and the New Rhetoric London:

Ghadessy, M., Henry A., and Roseberry, R (2001) Small Corpus Studies and ELT

Theory and practice Amsterdam/Philadelphia: John Benjamins Publishing company.

Greenbaum, S and Quirk, R (1990) A Student's Grammar of the English

Ha, P L (2011) The writing and culture nexus: Writers’ comparisons of Vietnamese and English academic writing In P L Ha & B Baurain (eds.), Voices, Identities,