Generic Structures and Linguistic Features of TESOL Master ’ s Thesis

Acknowledgements are a widely used genre in academic discourse to express gratitude towards various types of assistance and contribution of individuals and institutions, and they have been reported to be contextually and culturally interwoven. Despite a number of acknowledgement studies in various settings, little is known of how these texts are composed by EFL writers in Vietnam and whether or not the culture has any influence on the composition of acknowledgements as reported in previous studies of acknowledgements. Following Hyland’s (2004) and Hyland and Tse’s (2004) frameworks, this study investigates the generic structures and linguistic elements of acknowledgements in 202 TESOL master’s theses written by Vietnamese postgraduates. The findings were compared with those in the literature to explore cross-cultural variations in acknowledgement writing. Semi-structured interviews were also conducted with actual thesis writers. The results revealed that these writers generally followed Hyland’s (2004) three-tier structure of writing thesis acknowledgements, but they were frank and less reserved in expressing their gratitude than their Chinese-speaking counterparts. Moreover, socio-cultural expectations, personal dispositions, and individual writing styles mainly affected their move constructions, inclusions of acknowledgees and linguistic choices.


INTRODUCTION
Acknowledgements have become a common practice in writing theses and dissertations, and they have received increasing attention from language researchers (thesis and dissertation are respectively used to refer to the work written by master's (M.A.) and doctoral (Ph.D.) students).Hyland and Tse (2004) state that thesis/dissertation acknowledgements (DAs) are a distinct genre with a clear purpose and physical location.According to Giannoni (2002), the pioneer analyst in the genre of acknowledgements, acknowledgements are staged texts coherently constructed with rhetorical devices to map credits.Hyland (2003Hyland ( , 2004) ) defines acknowledgements as a 'Cinderella' genre as they are rhetorically sophisticated and formatively complex, neither completely academic nor entirely personal.They are seemingly peripheral to academic research but still have significant socio-pragmatic connections.In fact, acknowledgements offer authors space to construct their personal identity and professional engagement, and show their disciplinary connections with those they wish to publicly recognise (Hyland 2004).For thesis/dissertation authors, therefore, DAs are not merely a list of individuals acknowledged for their assistance during a long and demanding research process, but a means to both demonstrate their awareness of central academic values, such as modesty, gratitude, and appropriate self-effacement and display their active membership, their interpersonal relationships, and their immersion in scholarly networks (Cheng 2012, Hyland & Tse 2004).DA sections are, thus, short but important pieces of text in student theses/dissertations, which orient the reader to what the student has done as well as where the student is placed in various scholarly and social networks.Like the table of contents, DAs are often the last thing a student writes, but among the first thing an examiner reads (Paltridge & Starfield 2007), which can help make the first impression a good one (Finn 2005).
Although acknowledgements in research articles (RAs) are described as a 'part-genre' (Giannoni 2002) or a 'para-text' (Salager-Meyer, Ariza & Berbesí 2009) because they occupy a little space, DAs are seen as "a significant genre in their own right" (Hyland 2004, p. 306).As defined by Swales (1990), genres include a series of goal-oriented communicative events formed out of schematic structures, and those who take part in these events share the same communicative purposes.Furthermore, a genre is highly structured and conventionalised, and has specific constraints on lexis and moves exploited by the members in a community to achieve communicative purposes (Bhatia 1993, Swales 1990).Analyzing a genre hence can help ESP practitioners identify how texts are structured and distinguished in conventional and socio-cultural contexts in order to realise their communicative purposes (Hyland 2004).Moreover, analysing texts in the genre-based approach offers researchers "explicit and systematic explanations of the ways language functions in social contexts" (Hyland 2004, p. 18), which also helps new disciplinary members acquire the disciplinary cultures (Bhatia 2002).
There are two main approaches in conducting genre-based analysis on academic texts, namely generic moves (at a macro-level) and linguistic features (at a micro-level).A large number of studies have relied on macro-level analysis to investigate moves/steps of different English RA sections (Holmes 1997, Kanoksilapatham 2005, Nwogu 1997, Peacock 2011, Swales 1990, R. Yang & Allison 2003).However, studies employing genre-based approaches on Ph.D. dissertations and M.A. theses tend to be scarce.To the author's knowledge, a couple of studies (Afful & Mwinlaaru 2010, Cheng 2012, Zhang 2012, Zhao & Jiang 2010) have investigated DAs written by M.A. students, and they mainly focused on the texts written by Chinese-speaking ones.What is apparent is the scarcity of studies on texts written by other EFL writers, and research work on Vietnamese writers is virtually limited.Employing the genre-based analysis, the present study, therefore, explores how Vietnamese students compose the acknowledgements of their TESOL M.A. theses.
Despite being an indispensable section in published texts and student theses/dissertations, acknowledgements have received less research attention compared to other academic texts, namely the Introduction, Literature Review, Methods, Results, Discussion and Conclusions (Cheng 2012, Hyland 2004, Zhang 2012).Previous research on acknowledgements can be divided into three groups.While the first and second groups of studies focus on the acknowledgements written by native speakers of English (Davis & Cronin, 1993) and non-native speakers of English (Al-Ali, 2010;Hyland, 2003Hyland, , 2004;;Hyland & Tse 2004, Jaroenkitboworn 2014, W. Yang 2012a, 2012b, 2013, Zhao & Jiang 2010), the last group centers around the cross-cultural similarities and differences in acknowledgements (Giannoni 2002, Lasaky 2011, Salager-Meyer et al. 2009, Zhang 2012).However, most studies in these three groups were guided by the theoretical framework posited by Swales (1990Swales ( , 2004) ) in terms of generic patterns and linguistic realizations.Giannoni (2002) analysed 100 RA acknowledgements in English and Italian scholarly journals and found that their generic structures not only reflect the disciplinary differences but are also affected by the national patterns of disciplinary communities.A similar finding to Giannoni's (2002) which states that acknowledgers from different linguistic backgrounds, even though they study in the same field, express gratitude in different ways was reported by Salager-Meyer et al. (2009) in their study of 150 medical English and Spanish RA acknowledgements by writers from Venezuela, Spain and the USA.The three studies that have opened a window for subsequent research to examine acknowledgements in more detail are Hyland (2003Hyland ( , 2004) ) and Hyland and Tse (2004).These studies established a three-tier structure of DAs from a corpus of 240 M.A. and Ph.D. DAs written by Hong Kong Chinese-speaking postgraduates from six academic disciplines.In this model (Figure 1), DAs mainly consist of one obligatory move, namely the thanking move (Move 2), and two optional moves (reflecting and announcing moves), and their subsequent steps.Hyland's (2004) generic structure of DAs Hyland and Tse (2004) furthered the study with a focus on the linguistic devices employed to realise the thanking move.The results showed that all thanking steps contained five patterns of gratitude expressions: nominalization (My sincere thanks to Prof. A or my deepest gratitude goes to Dr. B), performative (I thank (someone), adjective (I am grateful), passive voice (Appreciation is given), and bare mention (X has been helpful).Nominalization patterns were the commonest expression used to convey gratitude, followed by performative, adjective, passive voice and bare mention.In addition, the acknowledgers tended to use adjectives and adverbs to intensify their gratitude, varying from participant to participant.Besides these, most authors used the first person pronoun (I, my), while some thanking acts contained non-authorial subjects and a few used the subject "the author".The exclusive use of the first person pronoun, and the decrease in the use of the third person and passive voice in Hyland and Tse (2004) is in line with the finding by Scrivener (2009) who studied 219 DAs written from 1930 to 2005 by history Ph.D. students.
Following these pioneering studies of DAs, some researchers have searched for new moves/steps while others have integrated discussions of the influences of cultures and society on DA language use.For example, Zhao and Jiang (2010) employed Hyland's (2004) and Hyland and Tse's (2004) models to explore the generic structure and lexico-grammatical patterns of DAs in applied linguistics, English language and literature written by students from China.Their findings on the generic structure, gratitude expressions and modifiers used in the thanking acts were compared with those written by Chinese speaking writers in Hong Kong.Their conclusion was that divergences exist in the academic practice between mainland and Hong Kong Chinese graduates.The absence of Reflecting and Announcing moves, and the excessive use of bare mention form and modifiers (sincere, special, heartfelt, hearty) in the thanking acts were found in DAs written by students from China.Zhao and Jiang (2010) attributed these variations to the cultural, mental and academic diversities in the two contexts.Three similar studies (W.Yang 2012aYang , 2012bYang , 2013) which compared Ph.D. DAs from three disciplines written by Taiwanese writers in Taiwan and in the USA, and M.A and Ph.D. DAs by Chinese-speaking writers in Taiwan, Hong Kong and China confirmed that contextual contexts, such as academic, socio-cultural or geographical differences affect the construction of moves and choices of linguistic elements in DAs.For instance, a new step 'Making a confession' which was not reported in the other two settings (Hong Kong and China) was identified in DAs by Taiwanese.Moreover, the rhetorical language in DAs written by Taiwanese in Taiwan tended to be more direct, emotional, and definite than that of DAs written by Taiwanese studying in the USA.In a similar project on 45 DAs written by Filipinos, Chinese and American M.A. students, Zhang (2012) found variations in the texts due to the differences in these writers' cultural backgrounds.In particular, while Chinese writers preferred nominalization and performative verbs and American counterparts relied on performative verbs and bare mentions to express gratitude, Filipinos made frequent used of parallel constructions of thanking acts.
In addition to the research in the settings of Chinese culture, a number of DA studies adopting Hyland's (2003Hyland's ( , 2004) three-tier model in other contexts have been conducted.In these studies, besides analyzing the generic structures and linguistic features of DAs, much emphasis is placed on how social and cultural factors affect the formation of moves/steps and identities.For instance, Al-Ali's (2010) study of 100 Ph.D. DAs written by Arabian students in several Middle East countries identified a new move, namely Thanking Allah (God).Al-Ali claims that due to religious beliefs and the local academic and social conventions, the Arabian acknowledgers tended to use specific contextualised components to realise their thanking acts.Similarly, in his study of Ph.D. DAs written by native English speakers (NS) and nonnative-English speaking (NNS) Iranian students in applied linguistics, Lasaky (2011) identified Thanking Allah (God) in the latter group's DAs.The presence of this new step in DAs by Muslim students (Al-Ali 2010, Lasaky 2011) indicates that genres are dynamic, and can be shaped and appropriated to accommodate newly accepted practices and generate new constructions (Al-Ali 2010, Bhatia 2004).
Furthermore, the absence of Reflecting move and Accepting responsibility step in the DAs of the NNS group was also found in Lasaky's (2011) study whereas both appeared in the DAs of the NS group.Lasaky explained that this difference stems from cultural issues, as Iranian students usually perceive writing a dissertation as their own duty, and accountability is not given importance.In Cameroon, moreover, writers used nativised deferential strategies and nominal phrases to express gratitude to supervisors and seniors (Nkemleke 2006).This DA writing practice by Cameroonians was claimed to be influenced by the cultural belief that one's success or achievements come from the collective efforts of an extended community.Mkemleke's finding, therefore, demonstrates that culture plays a significant role in shaping non-native English texts and "understanding texts produced in a specific non-English culture setting may thus involve a proper appreciation of the expectation of the people there" (Nkemleke 2006, p. 181).In a case study of one DA by an M.A. Ghanaian student, only Thanking and Reflecting moves and a hybrid discourse were discovered (Afful & Mwinlaaru 2010).Although different linguistic items used to express gratitude to different acknowledgees in thanking acts in this single DA was confirmed in other studies (Hyland & Tse 2004, Zhao & Jiang 2010), the presence of two moves in a Ghanaian DA cannot be generalised since this could be this single informant's personal preference.
In Thailand, where English is a foreign language and Buddhist followers account for almost 95% of the population, DAs were structured in the way that reflects Thai culture (Jaroenkitboworn 2014).With three moves (Thanking, Announcing and Signing off), the generic structure of Ph.D. DAs by Thai writers are different from Hyland's (2004).Furthermore, that 'parents' were thanked in every DA for giving life to the writers in Jaroenkitboworn's (2014) study illustrates the Buddhist belief that giving birth to a person is considered as a great virtue and a benevolence that children have to be grateful for.Moreover, being indirect in expressing gratitude is considered as "appropriate, formal, polite, and respectful" in Thai socio-cultural contexts (Jaroenkitboworn 2014, p. 124).
In summary, most of these studies adopted Hyland's (2003Hyland's ( , 2004) three-tier model to explore the generic moves and linguistic features of DAs.However, socio-cultural factors, which were not mentioned in Hyland's model, have received much attention from genre researchers, and as a result, a new move "Signing off" and steps, such as "Thanking Allah (God)" and "Making a confession" have been discovered.Such findings echo the claim that genres are dynamic and the lexical and generic elements dictated by cultural preferences will shape a new generic form (Bhatia 2004).Furthermore, it is argued that identifying moves is no longer a valid strategy for genre analysts if the roles of writers and the expectations of their communities are not considered (Nkemleke 2006).Moreover, research on the extent to which English texts have become nativised in the home culture is claimed to be an important area of research in the "outer circle" of English (Nkemleke 2006).
There seems to be, however, a dearth of interest in this domain of study in Vietnam.In fact, though thorough and systematic research on what authors write in their acknowledgements, how they construct them, and how contextual factors affect the formation of the DAs in different contexts is documented in the literature, little has been known about how these texts are structured by EFL writers in Vietnam.Whether or not Vietnamese cultures have any influence on the composition of DAs is likely to be of great interest to ESP practitioners.In fact, an analysis of DAs written by Vietnamese authors is an opportunity for a comparison of cultural differences on writers and genres, as opposed to Chinese speaking ones (Hyland 2004, Hyland & Tse 2004, W. Yang 2012a, 2012b, 2013, Zhao & Jiang 2010) and Thais and Filipinos (Jaroenkitboworn 2014, Zhang 2012, respectively).This study, therefore, aims to investigate the generic structure and linguistic features of acknowledgements written by Vietnamese M.A. students to explore cross-cultural variations in acknowledgement writing.

CORPUS AND PROCEDURES
Due to the inaccessibility to the libraries of the other four universities with the TESOL M.A. program in the Central and the North of Vietnam, only 202 TESOL M.A. thesis acknowledgements from three universities in the South of Vietnam were collected with writers' consent, making a corpus of 41,552 words with an average of 206 words each.Theses in the library database of these universities are regarded as standard and accepted ones in TESOL discourse communities in Vietnam because they are the final revised versions submitted to the school libraries as required by the universities after the candidates have passed the oral thesis defense.Since generic structures are subject to variations across time, the acknowledgments of theses written during the years from 2005 to 2015 were selected in the hope that it would reflect the writing practice of this genre by a group of M.A. students in this part of Vietnam.Hyland's (2004) and Hyland and Tse's (2004) models were respectively employed to explore the generic structure and linguistic features of DAs in the current corpus.The DAs were first randomly coded from number 1 to 202 for the ease of data management and analysis.They were then manually analyzed for their move structures by the researcher and her colleague, who holds a doctoral degree in English Language Studies.To ensure the intercoder reliability, the analysis of 50 randomly-selected texts were examined and compared, yielding high inter-rater reliability rates (97%).After the moves and steps were identified, their frequency was recorded in order to verify the extent to which a particular move or step was employed.Secondly, the use of acknowledgees, linguistic patterns, modifiers, and authorial subjects in the thanking acts was studied with the help of Antconc software.Finally, the documented data on move structures and linguistic features were compared with those in the literature (where possible) to explore cross-cultural similarities and differences in acknowledgement writing.Besides the genre-based analysis of these texts, the semistructured interviews with ten actual thesis writers were conducted to provide an insightful understanding of text analysis and of discovering these writers' preferences and thoughts.The interviews were conducted in Vietnamese as this encouraged them to express what they really thought about the questions.However, only the information that helped clarify the issues related to the writing of these texts were translated and included as excerpts throughout the texts.

FINDINGS AND DISCUSSION
This section presents the findings and discussion on the generic structure of DAs in the current corpus, followed by their linguistic features in terms of acknowledgees, expressions of gratitude, modifiers in thanking acts, and choices of authorial subjects.

GENERIC STRUCTURE
As can be seen in Table 1, Vietnamese writers tended to follow Hyland's (2004) three-tier structure, but with a very small proportion of move 1 (Reflecting) and move 3 (Announcing) (0.07% and 0.98%, respectively).Move 2 (Thanking), in contrast, appeared in every DA with relatively different distributions of steps.This finding is in line with those of previous studies of M.A. DAs, especially those written by Chinese-speaking students in applied linguistics (Hyland 2004, W. Yang 2013, Zhang 2012, Zhao & Jiang 2010) in which moves 1 and 3 are optional while move 2 enjoys the obligatory status.The underrepresentation of Reflecting and Announcing moves in the TESOL DAs, however, as reported in the interviews with all ten actual writers, was mainly due to their infrequent occurrence in the previous DAs which these writers relied on.Moreover, unlike Zhang's (2012) corpus in which 'dedication' was found in a separate section of theses rather than as an optional step in DAs, there is no sign of standalone dedication sections in the current DA corpus.It can be inferred that Reflecting and Announcing moves have not yet become conventional in theses at the TESOL community in Vietnam.Unlike Chinese-speaking writers' DAs, however, the current DA corpus consisted of 'Signing-off' move, identified by previous researchers (Al-Ali 2010, Jaroenkitboworn 2014, Zhang 2012), and two instances of 'Thanking God'.Different from the 'Signing-off' move of Arabian and Thai graduates in which only writers' full names were found or Filipino writers with signatures and dates, this newly identified move consists of the place and date to write the DA, followed by the writer' full name in the following line.Although this closing structure is conventional in writing formal documents in Vietnamese language, the writer of one DA with this move revealed that he simply followed the DA structure of a senior's thesis.The interview data tend to suggest that these writers simply mimicked the existing DAs.In summary, the generic structure of DAs by Vietnamese students has one obligatory move (Thanking) and three optional ones, namely, Reflecting, Announcing and Signing-off moves and these moves follow the same order as presented in Table 1.Regarding the obligatory move, 64 DAs had all four steps, accounting for 31.7% of the present corpus, and only step 2.2 (Thanking for academic assistance) was found in each DA, followed by step 2.4 (Thanking for moral support) and step 2.3 (Thanking for resources) (96% and 95%, respectively).Step 2.1 (Presenting participants) was found in almost one third of the corpus (Table 1).Unlike those written by Chinese students (Zhao & Jiang, 2010) in which there were no distinct boundaries among the steps in the Thanking move, DAs in the current study had clear demarcation as described by Hyland and Tse (2004) because each paragraph was allocated for one acknowledgee.In fact, with a total of 949 paragraphs of move 2 (an average of 4.7 paragraphs per DA), step 2.2 was the most recursive with a frequency of three times each, while steps 2.3 and 2.4 were seen nearly two times (Table 2).The prevalent presence of step 2.2 in this study is likely to reflect the ethics of gratitude to teachers in Vietnam, as known in the saying "nhất tự vi sư, bán tự vi sư", which literally means that students should appreciate a single word or half the word that their teachers have taught.Moreover, as seen in Table 2, step 2.3 occurred more frequently than step 2.4, which is not in line with those reported in previous DA studies.A closer look at the texts showed that in step 2.3, these Vietnamese writers employed many thanking acts for different acknowledgees depending on various types of their assistance while in step 2.4 it is common to have many individuals in one thanking act (Example 1). ) is that step 2.4 was not found in every TESOL DA.Although the finding is in accordance with that in Hyland (2004), Vietnamese writers seemed to have their own reasons for not including this step in their DAs.From Excerpt 1, it can be seen that this writer was frank in including who to be thanked in his DA.
(Excerpt 1) "I did the study by myself and I had to overcome all difficulties when writing the thesis on my own.And I am the only one in my family to have education and I am the one who paid for my study; my parents and siblings are sales people, they did not care about my study." Besides the distributions of steps, Table 2 also shows the difference in the average number of steps per text among four corpora.Vietnamese writers employed relatively more steps than writers from Hong Kong, China, and the U.S., but fewer steps than those from Taiwan and the Philippines.The excessive use of steps by the Filipinos is explained by their 'debt of gratitude' and 'high uncertainty avoidance' cultures, which encourage displaying explicit gratitude to many different individuals (Zhang 2012, p. 151).

ACKNOWLEDGEES
As argued by previous authors (Al-Ali 2010, Jaroenkitboworn 2014, Nkemleke 2006, W. Yang 2013), culture-specific preferred conventions affect how writers interact with the people acknowledged in DAs.As seen in Table 3, Vietnamese writers expressed their gratitude to a broader range of people when compared with Chinese and Taiwanese students in Zhao and Jiang (2010) and W. Yang (2013), respectively.Other than those most commonly acknowledged (e.g., advisors, other teachers, participants, friends, family members, or institutions), appreciation was also expressed to 'authors/publishers' of reference materials quoted in their theses (Example 2), 'bosses' (principal/rector of the school where writers are affiliated with) (Example 3) and neighbors (Example 4).(For the sake of research ethics, any real names of people and schools in the data were substituted by X and/or Y).Although W. Yang (2013) claims that Taiwanese writers of English literature majors had to acknowledge the authors of literary works, appreciation was expressed to authors/publishers of the references cited in the theses.These writers reported that the resources of references at their universities' libraries are limited, and they could not access any databases for references outside Vietnam.Having relevant materials for their studies is thus something they appreciated greatly.Moreover, the prominence of thanking acts for the heads of the writers' workplaces (28.2 %) was not reported in DA studies in other academic settings.As explained by these Vietnamese writers, their bosses had the absolute power in granting any opportunity for pursuing higher study and acknowledging them in their DAs is 'a must' in return to their kindness.While this finding partly reflects the working culture in the state sectors in Vietnam, where Communism prevails and the head of any state organization must be a Communist-party member with extreme power within the organization, it confirms previous researchers' statement that socio-cultural preferences will shape non-native English texts, and understanding these texts may thus involve a proper appreciation of the expectation of the people there (Bhatia 2004, Nkemleke 2006).
Table 3 also displays what the acknowledgees were thanked for in the present corpus.For academic assistance (step 2.2), thesis supervisors, the single category of acknowledgees mentioned in all DAs, were always the first to be acknowledged in the longest paragraph for their academic, moral, and resource assistance (Example 5), followed by other instructors teaching the M.A. course, authors and friends (Table 3).Although this tendency is similar to previous studies of DAs in different academic contexts (Hyland 2004, Jaroenkitboworn 2014, W. Yang 2013, Zhao & Jiang 2010), Vietnamese students did not give any credit to committee members as Taiwanese in W. Yang (2013) did (1.5% versus 76%, respectively).It is surprising, however, to know from the interviews that committee members were examiners who checked whether or not the students were qualified to be awarded with an M.A. degree; yet they were not included in their DAs.This finding is likely to indicate that these writers acknowledged only those who directly assisted them in their postgraduate study.
(5) 'First and above all, I would like to express my deepest gratitude to my thesis supervisor, X (full name), Ph.D., who read my manuscript with great care and devotion, gave me thoughtful and insightful comments and provided me with valuable support and relevant materials in the preparation and completion of this thesis, asserting her indispensable role as a wholehearted supervisor.I would not have finished my thesis without her enthusiastic guidance and constructive critical questions to help me think over the problems.'(DA23) While family members, friends and colleagues were appreciated mainly for their moral support (step 2.4) in the last paragraph of a DA, their assistance for resources, such as reference materials, data collection and data analysis was also recognised.Participants, in addition, enjoyed the highest frequency of acknowledgement, accounting for 62% of step 2.3 of DAs by Vietnamese writers whereas they appeared in 35% in Yang's (2013) corpus.As reported by some thesis writers, participants deserved special acknowledgement because without them, they could not complete their research as the TESOL program required them to conduct experimental research for their theses.
Another interesting finding is these writers' intentional exclusion of acknowledgees' names.Unlike the general belief that the names of individuals who had marginal contributions to the development of the thesis are dropped (Hyland, 2004), some of these Vietnamese writers excluded the names of those who, they thought, would not read their theses, even though they had greatly assisted them, as can be seen in Excerpt 2. Among 457 names included in the present corpus, eight first names of the writers' children and close relatives were found in step 2.4, and 449 full names preceded by the title Mr./Ms.(for those who do not have any other academic ranks) (Examples 3 & 4 above), or followed by an academic title (M.A., Ph.D., or Senior lecturer, a title awarded by Vietnamese Department of Education and Training) (Example 5 above) were recorded in steps 2.2 and 2.3.
(Excerpt 2) "For my advisor and the influential people, their names were included.Although my family members gave me a lot of support, I just thanked them in a general way, no names mentioned because I thought they will never read my thesis.But for my friends, I mentioned their names because surely, they will read my work."

EXPRESSIONS OF GRATITUDE
The present study found all five patterns to express gratitude identified by Hyland and Tse (2004) in the thanking move (steps 2.2 to 2.4).As can be seen in Table 4, Vietnamese students preferred performatives, accounting for nearly half of the linguistic patterns identified in the corpus (43.92%).Bare mention was ranked second (21.66%), followed by adjectives and nominalization with a fairly similar frequency (16.47% and 14.17%, respectively) while fewer passives were employed (3.78%).Although this tendency is in line with Chinese M.A. writers in Zhao and Jiang (2010), it is different from that by Hong Kong counterparts whose preference for nominalizations and performatives, accounting for one third of each pattern types while bare mention was the least used one (6.8%)(Hyland & Tse 2004).Similar to Zhao and Jiang's (2010) explanations, the low usage of the passive pattern in expressing gratitude by Vietnamese students is because Vietnamese language does not have voice categories and passive voice is expressed in a covert way.These Vietnamese writers, therefore, may not feel accustomed to employing this pattern in their English DAs.However, as reported in their interviews, their use of different linguistic patterns in the DAs resulted from their single intention of avoiding the repetition of the same structure (Excerpt 3).Furthermore, these Vietnamese writers also revealed that their preferred use of bare mention was to intensify their gratitude, especially when used with "without" or "subjunctive if" (Examples 6 & 7,respectively).Their employment of this linguistic form was thus unlike that of Chinese writers whose preference for bare mention was affected by their reserved culture in expressing their feelings and emotions.Moreover, bare mention was also found to express the appreciation to the writers' advisors or family in an exaggerating way, as can be seen in Examples 8 and 9, respectively.Such information suggests Vietnamese writers are much less reserved than Chinese counterparts in expressing their gratitude.
(Except 3) "Except for my thanks to the supervisor with 'would like", I changed the ways to thank others because I did not want to repeat the same structure.So, if I already used noun phrases to thank the participants, I would use passives or adjectives for my friends and family."(6) "Without his devoted assistance, this thesis could not have been completed." (DA103) (7) "The thesis would have never been possible if the participants had not responded to the survey questionnaires."(DA109) (8) "I am blessed to have known her and to have been accepted as her supervisee" (DA178) (9) "I feel proud of being her daughter" (DA68) A closer examination of these Vietnamese writers' most favorite linguistic pattern showed that almost two-thirds of their thanking verbs were prefaced by modal verbs, such as would like to or must (Table 5), and these modal verbs occurred with the highest frequency in thanking for academic assistance.The interviews with thesis writers revealed that the modal pattern was considered as a formal language feature politely used to address people of higher social ranks like advisors or school principals.Although Hyland and Tse (2004) state that the illocutionary force of the thanking acts with such modal verbs is weakened, these Vietnamese writers' explanation for this pattern use is likely to be related to Vietnamese cultures, or perhaps Asian cultures in which being a little more indirect is considered to be formal, polite, and respectful (Jaroenkitboworn 2014).Table 6 shows the linguistic patterns used across the thanking steps in the TESOL DAs as compared with those in DAs by Chinese writers (Zhao & Jiang, 2010).The two groups of writers tended to share a similar distribution of linguistic patterns for each group of acknowledgees.In particular, performatives were found the most prominent in all three thanking acts, followed by bare mention for academic assistance, adjectives for resources and nominalization for moral support (30.89%, 20.2% and 18.6%, respectively).The passive form was mostly used in step 2.4.However, it should be noted that due to these Vietnamese writers' intentionally altering the linguistic patterns across the steps as revealed in the interviews, the sentence patterns used to express gratitude across the thanking steps in these TESOL DAs are inconclusive.

MODIFIERS IN THANKING ACTS
Modifiers in thanking acts include adjectives and adverbs which are often found to attach to these acts.The analysis of the present corpus shows that 87% of the thanking acts in steps 2.2, 2.3 and 2.4 were modified in various ways.Thus it can be concluded that almost all thanking acts in this DA corpus contained modifiers.This percentage is by far twice higher than that in Hyland and Tse's (2004) and Zhao and Jiang's (2010) studies (35% and 38.5%, respectively).As asserted in the interviews with the thesis writers, adjectives and adverbs were used in the DAs as their device to strengthen the appreciation to acknowlegees.This finding is likely to confirm the tendency that Vietnamese writers are less reserved in expressing their acknowledgement than their Chinese-speaking counterparts.
Besides adverbs infrequently used to modify in the performative and adjective patterns, namely greatly, heartily, deeply, sincerely, extremely, profoundly, truly, highly and gratefully, adjectives were prominently seen to precede nouns for both expressing acknowledgements (e.g., gratitude, thank(s), and acknowledgement) and what to thank (e.g., support, guidance, and encouragement).In Table 7, honorific and intensified adjectives, especially superlatives (deepest, greatest, and special) were most frequently used to thank for academic and moral assistance.Such exaggerating use is claimed to be very common in non-Western cultures (Al-Ali 2010), and this is likely to be true in Vietnam.The thesis writers, in fact, confirmed their employment of strong and emotional adjectives for emphasizing supervisors and families' enormous contribution to the success of their theses.
Table 7 also shows a high frequency of different adjectives employed to intensify various types of assistance, and they were mainly found in thanking for academic assistance (Example 5).As argued by W. Yang (2013), the teaching of the genre knowledge of language and contexts will enable writers to sensitively observe the genre nature, and then to compose personalised and honest acknowledgements with choice and constraint provided, rather than mimicking existing samples.This study, therefore, provides a list of nouns and adjectives used in thanking acts (Table 7), and it is expected to help novice writers have a reference list for constructing their acknowledgements and thanking reasons.
in the 1930s to 100% in 2005.This implies that writing a DA can be less formal and more impersonal and emotional.In addition, like Chinese writers in Zhao and Jiang (2010), the subject with the author/the writer was not found in the current corpus whereas their occurrences were found in 3.1% of Hyland and Tse' (2004) M.A. DA corpus.Similarly, non-authorial subject sentences (e.g., Sincere thanks go to my colleagues and friends for their unfailing support) were found to be present with almost the same frequency as those in the Chinese counterparts.Nonetheless, the use of the third-person form (e.g., he, she, they, the thesis or their corresponding possessive adjectives) in the present study enjoyed a relatively higher frequency as compared with those by Chinese and Hong Kong writers (17.9%, 14%, and 2.1%, respectively).The reason for this lies mainly in these Vietnamese writers' preference for these forms in the bare mention pattern.

CONCLUSION
With reference to Hyland's (2004) and Hyland and Tse's (2004) models, the present study conducted a genre-based analysis of 202 TESOL M.A. DAs and semi-structured interviews with actual writers in Vietnam with the aim of investigating their generic structures and linguistic features.The findings were then compared with those reported in the literature to explore cross-cultural variations in acknowledgement writing.Although the results demonstrate that these writers generally followed Hyland's (2004) three-tier model, differences exist in constructing moves/steps, choosing linguistic elements and interacting with people acknowledged in DAs.These findings tend to support the documented arguments that this particular genre is dynamic, changeable, and is able to be manipulated.Furthermore, academic conventions, socio-cultural expectations, personal dispositions, relationships with acknowledgees, and individual writing styles may account for this diversity.
This study also suggests that NNS writers' awareness of the genre should be raised through explicit instruction in order for them to compose purposeful and interactional acknowledgements instead of copying verbatim from the existing texts.Due to the limited samples and scope of the study, further research is needed.DAs written by Vietnamese authors in the hard disciplines can be compared to the analysis of DAs in social sciences.Furthermore, DAs written by Vietnamese graduates studying in English-speaking countries can be examined to explore whether the different settings exercise any influence on the DA composition.

TABLE 1 .
Percentages of moves and steps with reference to previous studies of M.A. DAs

TABLE 2 .
Relative frequency of steps in each text with reference to previous studies of M.A. DAs

TABLE 3 .
Acknowledgees across thanking acts and with reference to previous studies of M.A. DAs

TABLE 4 .
Occurrence frequency of linguistic patterns with reference to previous studies of M.A.

TABLE 5 .
Verb patterns of gratitude expressions

TABLE 6 .
Zhao and Jiang (2010)tude expressions across thanking steps and with reference toZhao and Jiang (2010)

TABLE 8 .
Occurrence percentages of authorial subjects with reference to previous studies of M.A. DAs