VERB TENSES A N D H EDGING IN PUBLISHED A N D UNPUBLISHED APPLIED LINGUISTICS RESEARCH PAPER ABSTRACTS

INTRODUCTION

Abstracts are short summaries of longer academic papers. They include references to the research topic, aims and methods, and often also indicate previous research, own results and conclusions. It is important how these pieces of information are constructed linguistically to ensure clarity and unambiguity, to show the author’s intentions and to indicate the certainty of their assertions. Lores (2004, p. 281) concludes that “there seems to be general agreement on three ways in which research abstracts differ from RAs [research articles]. They differ in their function, in their rhetoric structures and in their linguistic realizations”. In the last forty years many researchers have analyzed the structural and linguistic contents of RAs and other published or unpublished academic papers. Studies which have compared papers across disciplines have pointed to some universality, but mainly discovered discipline-specific rhetorical features (e.g., Pho, 2008; San & Tan, 2012; Stotesbury, 2003).

Due to this specificity it is difficult to compare results based on corpora drawn from different fields, journals, authors and languages. Relatively few studies have analyzed research article abstracts (RAAs) in applied linguistics and less is known about the differences between abstracts written by expert and novice writers. Ac­cording to the APA manual, a research article abstract which is “accurate, succinct,

qu ick ly comprehensible, and informative w ill increase the audience and the future retrievab ility” of an article (APA, 2003, p. 15). G ate-keeping bodies, such as ed i­tors, reviewers, conference organizers, m ay decide on the dissem ination of a re­search project based on a qu ick screening of the abstract, and published articles also find their readers through the abstracts that are often open-access even when the articles themselves are not. It m ay be argued that thesis abstracts serve a differ­ent purpose as they do not exist independently from the main text, and their aud i­ence is restricted to the supervisors and in-house review ers. Nevertheless, students are expected to follow academ ic conventions and summarize their thesis in an ab­stract sim ilar to those of published papers. As Koutsantoni (2006) righ tly points out, RAAs and thesis abstracts m ay be considered two sim ilar but distinct genres, especially due to the status of the authors in the academic discourse community of their fields. She also backs up this proposition b y showing the power asym m etry between student w riters and supervisors.

However, I w ould argue that a sim ilar k ind of power asym m etry exists between authors of RAs who w ish to publish and the editors or independent review ers who m ay reject their papers. I believe that abstract w riting is a m arginal part of academ ­ic w riting instruction at all levels, and both novice and experienced w riters learn it im plicitly w h ile they read published articles. I assume that most authors p ay less attention to abstracts com pared to other sections, and that abstract w riting remains the last step, often done w ith hustle due to final deadlines. For these reasons ab­stracts are good indicators of authors’ ta rget language academ ic w riting skills and their ab ility to see and summarize the k ey points of their papers by following rhetorical and linguistic conventions (Doro, 2013a). S tudent w riters are expected to follow these conventions in a ll sections of their papers, even w hen their w riting reaches a ve ry small audience and their experience as researchers and as academic text producers is v e ry lim ited and often insufficient. For the reasons mentioned above, I believe that these two abstract genres are close enough to be comparable and their system atic review m ay shed light on the sim ilarities and differences in the linguistic realization of abstracts w ritten b y professional and student w riters. Ap­plied linguistics papers have not been w id e ly studied from this point of view, and a comparison of published papers and undergraduate E FL theses is hoped to serve as basis for a more explicit academ ic w riting instruction of abstracts.

BACKGROUND

The next sections provide an overview of the literature concerning the following areas: (a) genres characteristics of abstracts and the differences between RAAs and thesis abstracts, (b) verb tense usage in RAs and (c) hedging in academ ic discourse.

Abstracts serve an important role in academic discourse and are considered a distinct genre. Studies such as Sw ales (1990) and Askehave and Sw ales (2001) point out that genre analysis should include both bottom-up (text based) and top- down (discourse community practices, values, beliefs based) identifications. Based on their information content, structure and communicative purpose, two main types of abstracts are distinguishable, nam ely descriptive and informative (M artin­M artin , 2005). The second one is more detailed and provides information also about the results, outcomes and conclusions of studies. The rhetoric structure of research articles and abstracts has been broadly studied in the past. In 1981, M alcolm Sw ales designed the so called Create a Research Space (C A RS) model which consists of the following main moves: establishing the territory, establishing the niche and occupying the niche (Sw ales, 1990). This CA RS model has been applied by a large number of studies to investigate the rhetoric structure of RA introductions (e.g., Hirano, 2009; Ozturk, 2007; Sam raj 2002, 2005; Sheldon, 2011). Apart from research on the introduction sections, attention has also been paid to the analysis of RA abstracts. Santos (1996) created a model for RAA, which includes the following five moves: situating the research, presenting the research, describing methodology, summarizing the findings, and finally discussing the findings. This five-step model has formed the basis for the analysis of move structures of both published RAAs and theses written by student w riters (e.g., Doro, 2013a, 2013b; Lon, Tan & Abdullah, 2012; Pho, 2008; Tseng, 2011). A sim ilar framework, referred to as the IM RD or IM RaD , has also been used by researchers (e.g., Bhatia, 1993; Lores, 2004; Serholt, 2012; Swales, 1990). These acronyms stand for the introduction, methods, result and discussion sections of abstracts. The CA RS and IM R D frameworks are almost identical, but the first two moves in the CA RS model are combined in the IM RD framework; this ensures the com parability of research conclusions of studies that use either of the two models or an y of their modified versions. Lores (2004) analyzed abstracts published in applied linguistics journals and compared the thematic distribution of moves according to both the IM RD and the CA RS models. She found that 61% of the abstracts followed the IM RD model, 30.5% resembled the CA RS structure and 8% of them a combination of the two.

Several studies have pointed out the variab ility in the number of moves and the order of these moves employed in RAAs. Not all moves are obligatory, but when major moves (aims, methods and results) are missing, the reader m ay feel that the abstract is incomplete or less informative. In an earlier study (Doro, 2013a) I compared BA-level thesis abstracts (BTAs) written on applied linguistics, literature and culture topics. W hile some kind of reference w as given to the research topic in all the theses abstracts and methodology w as introduced in at least 90% of the papers in the three sub-corpora, move 4 (findings) showed a significant difference between linguistic and non-linguistic theses. This means that results w ere not summarized in more than half of the literature, culture and history theses, while

358

Abstract,) and their structure

90% of the authors of linguistics papers felt the need to indicate results in their abstract. The non-obligatoiy nature of referring to the findings in abstracts has also been reported in other studies (e.g., Doro, 2013b; Lon et al., 2012; Lores, 2004; Santos, 1996).

Verb tenses in academic research papers

Compared to the boom in the move structure analysis of research articles, much less has been system atically reviewed in terms of the verb tense usage in research papers, and most of these analyses concentrated on a lim ited number of science a r­ticles. M alcolm (1987) reviewed studies from the 1970s and 1980s that point to the functional use of tenses in academic discourse. He highlights that the choice of tenses, on the one hand, reflects the genre traditions and, on the other hand, the rhetorical choices of authors to express their intentions. Moreover, he cites exam­ples of ob ligato iy and optional constraints of tense usage in various parts and sec­tions of a paper. For example, references to past research are done in present sim­ple, present perfect or simple past, while methods and results are explained either in the present or in the past. Research manuals often provide clear guideline as to which tenses and aspects should be used for certain sections or purposes. Never­theless, they often fail to discuss the flexib ility of tense usage found in research a r­ticles. For example, the fifth edition of the APA manual suggests the following:

Use verbs rather than their noun equivalents and the active rather than the passive voice. Use the present tense to discuss results with continu­ing applicability or conclusions drawn; use the past tense to describe specific variables manipulated or tests applied (pp. 12-13).

Salager-M eyer (1992) pointed to a close relationship between rhetorical functions (moves) and finite verb forms and modals used in medical RAAs. He reported that the three most common tenses (simple present, past and present perfect) made up 89.8% of all finite verbs. W ithin these, past dominated w ith 51.4%, while the present constituted one-third of all verbs (32.8% ). The purpose, methods, results and case presentation sections used m ainly the past tense, while general truth, in­troductory notes, conclusions, recommendations and data synthesis were predomi­nantly referred to in the present. The present perfect w as used to refer to previous research. Tseng (2011) analyzed 90 applied linguistic abstracts and concluded that the present tense w as applied m ainly to present the background, aims and conclu­sions, while the past tense w as used to discuss the methodology and results. This is in line w ith the results of the Salager-M eyer (1992) study. Tseng also reported some variation across the three journals he used. Authors accepted for Applied L in­gu istics referred to their methods and results more often in the present than in the past. W hile analyzing differences between native and non-native English speaking

359

first authors, Tseng concluded that native speakers used the present tense more for aims and results than did non-natives.

L i and Ge (2009) reviewed the structural and linguistic changes in m edical arti­cles in a 20-year time span and found that the frequency of tenses in the various sections of articles had significantly changed. The overall frequency of the four most frequent tenses had not changed much, but their roles had shifted. Simple past increased in the presentation of new research, w hile the use of simple present increased in the discussion sections.

In a recent study de W aard and Pander M aat (2012) exam ined how readers classify statements taken from various sections of biology research articles (prob­lems, hypothesis, methods, results and implications). Their aim w as to see w hat happens if verb tenses are modified in these sentences. Their results demonstrate that verb forms serve as m arkers of intention and segment types of research pa­pers, and b y changing verb tenses the reader m ay classify the same sentence differ­ently. For example, fact statements and hypotheses used with the past tense are in­terpreted as results. Result sections, in contrast, are understood by and large as re­sults either in the present or in the past. The authors of this study found that modal auxiliaries also have a crucial role, as hypothesis-type statements w ithout modals are understood as facts or results, w hereas results w ith modal auxiliaries m ay be interpreted as hypotheses or w eak statements.

Hedging in academic ducourje

W hile academic w riting is often referred to as impersonal and objective, especially when it reports on facts and results, research papers go beyond the mere recitation of definitions or general truths. Authors analyze previous research and interpret their own results. W hile doing so they take personal responsibility for their claims, show their attitudes or the strength of their claims. H yland (1994, p. 240) states that “effective academic w riting actually depends on interactional elements which supplement propositional information in the text and alert readers to the w rite r’s opinion”. This does not mean a h igh ly individual or subjective voice, but is treated as a natural part of academic w riting. Authors provide tentative analyses through the use of epistemic modality which refer to their (lack of) confidence in their propositions. The degree of confidence is most often expressed through hedges (e.g., JuggeJt, Jeern to, Likely) and boosters (e.g., àtrongly, clearly). Hedges are one form of metadiscourse that authors use to modify their arguments in accordance w ith the needs and expectation of possible readers (H yland, 2004). Indeed, hedges have been found to be the most frequent category of m etadiscourse in a large corpus of doctoral and masters level dissertations (H yland, 2000). H yland (1996a, 1996b, 2000) w arns us that the usage of hedges is troublesome for most authors, especially non-native researchers, who do not c learly feel the meaning and function of hedges or boosters. He states the following:

360

Effective academic writing always carries the individual's point of view. Writers also need to present their claims cautiously, accurately and modestly to meet discourse community expectations and to gain accep­tance for their statements. Such pragmatic aspects of communication however are vulnerable to cross-cultural differences and L2 students are rarely able to hedge their statements appropriately. (2000, p. 477)

M ilton and H yland (1999) compared native and non-native students' essays written in English and concluded that non-natives used epistemic modifiers (such as usually, actually, m ay) less often and that 75% of such words w ere restricted to the ten most often occurring ones. On the contrary, Koutsantoni (2006), while analyzing engineering journal articles and theses, found that student w riters hedge more often than experienced authors, but also documented that students use different types of and a more restricted repertoire of hedges. Abdollahzadeh (2011) compared the conclusion sections of applied linguistic RAs published in English by native and non-native, Iranian researchers. He pointed to a “rem arkable tendency by both w riter groups towards hedging their propositions” (p. 288), but found no significant differences in the overall number of hedges between the two groups. Nevertheless, the Iranian researchers used fewer attitude markers (e.g., I feel it it even more important, it Li interesting) and emphatics (e.g., certainly, tru ly ). He concluded that the “differences are attributed to the degree of rhetorical sensitivity to and awareness of audience, purpose, cultural leanings, and the proclivities of the discip linary genre” (p. 288). V ariab ility in the reviewed research results are partly due to the categorization problems of metadiscourse markers, the proficiency level of the authors, and the section of articles under review in the different studies. Another important reason could be the fact that most studies concentrate on predeterm ined fists of markers (e.g., Hyland, 2000, 2010; Serholt, 2012), while others do a manual search of their corpora and include all signs of hedges or boosters (e.g., Abdollahzadeh, 2011).

THE STUDY

Research questions

Based on the literature reviewed above the following main research questions w ere formulated:

1. W hat is the proportion of finite verb tenses used in published and unpublished applied linguistics research papers?

2. W hat are the hedges used in these abstracts and what is the frequency of their occurrence?

3. Are there significant differences between applied linguistics research article abstracts and BA thesis abstracts in terms of verb tenses and hedging?

361

Corpus and methods

For this study English studies BA level theses, w ritten and defended b y students at a large H ungarian un iversity in the ye a rs 2011 and 2012, w ere considered. Since a previous study using these papers (Doro, 2013a) found results sim ilar to those voiced by other authors, nam ely that m arked differences exist between research papers w ritten in various disciplines, only one field of studies w as chosen. Applied linguistics paper abstracts w ere found to adhere more to the rhetoric structures and linguistic conventions of research papers than theses w ritten in other fields (Doro, 2013a) and showed the most consistency, therefore they w ere chosen to be the ones used in paralle l w ith published RAAs. The first corpus for this study consists of 30 applied linguistics BA thesis abstracts and the second one of 30 RAAs. The 30 student papers cover the m ajority of theses w ritten on various applied linguistic topics in these two academic years . The RAs w ere published in 2012 and random ly retrieved from three journals (Applied L inguistics, Jo u rn al of Second Language W riting and System ). A ll three are peer review ed and prestigious, but cover sligh tly different topics and include different authors. These selection criteria resu lted in a thesis corpus (referred to as BTA) of 6,046 w ords and a RA abstract corpus (referred to as RAA) of 5,333 words. This shows that, on average, the student abstracts are sligh tly longer. A ll student authors used English as a foreign language and wrote their papers as a final step for the completion of their 3-year BA English studies. As the native/non-native dichotomy w as not the ta rget of this research, it w as not a selection criterion, although most of the published articles have at least one non-native author.

The two corpora w ere first m anually screened for finite verb phrases and w ere categorized in the following tense and aspect combinations: present simple, past simple, present perfect, past perfect, present continuous, past continuous and fu­ture w ill. Cases of unusual or infrequent verb form choices w ere also screened for the move functions they perform in order to better understand the authors’ inten­tions. As a second step the corpus w as searched electronically for tw en ty items w hich commonly perform hedging functions in academ ic w riting (see Table 1 for the list) using AntConc, a text analysis and concordance program . W ith the help of the concordance function, a ll instances w ere carefu lly analyzed ind iv idually to en­sure that they w ere perform ing m etadiscoursal functions. Results w ere normalized per 1,000 words to allow comparison across the two corpora of sligh tly different sizes. F inal figures w ere calculated as number of occurrences, percentages, and also as proportions.

Table 1. L inguistic devices of hedges used in this study

Based on H yland (2000) and Serholt (2012)

Taken from the BAT and R A A corpora

can tend

may tiycould seek

might relatively

seem sometimes

appear mostly

assume

believe

speculate

indicate

likely

possible/possibly

probable/probably

generally

RESULTS AN D DISCUSSION

Verb teruej

The frequency counts show the ru ling position of simple present, followed b y sim­ple past. This is in line w ith the research review ed above (e.g., de W aard & Pander M aat, 2012; Li & Ge, 2009). Table 2 shows the different patterns of verb use. S im ­ple present and simple past together accounted for 90% of the finite verbs in the BTA corpus and 98% of the RA A corpus. F ive published articles and one student paper used the present tense exclusively. Present perfect w as identified much less often (accounting for 5% and 3%, respectively) and the other tenses occurred only rare ly . There is, however, difference in the num ber of finite verb phrases and the distribution of verb forms between the two corpora. The BTAs used between 9 and 27, overall 453 (75/1000) finite verbs, w hile the RAAs em ployed between 7 and 19, overall 366 (69/1000) finite verbs. This suggests that on average, student w riters d id not only w rite longer summaries, but w hile doing so they used fewer non-finite verb phrases and subordinate sentences than the RAAs.

363

Table 2. F inite verb usage in the two app lied linguistics corpora

Tenses Simplepresent

Simplepast

Presentperfect

Pastperfect

Presentsprogressive

Pastprogressive

Futurewill

Total

BTAs 295 123 23 2 l 2 7 45 3

65 .1% 24.6% 5% 0.4% 0.2% 0.4% 1.5%

RAAs 212 139 11 0 3 0 1 366

57.9% 40% 3% 0.8% 0.3%

W hen tak ing a closer look at the functions of these verbs, results indicate that stu­dents use tenses log ica lly and system atically and show good know ledge of the aca­dem ic w ritin g conventions of abstracts. This w as also found in an experim ental study (Doro, 2013a) in w hich ten app lied linguistics student abstracts w ere com­pared w ith non-linguistic abstracts. No unclear verb choices w ere detected in the RAAs and only a few cases are problem atic in the BTAs. These include the sudden sw itch from simple past to simple present or present perfect in the results move, and the use of present perfect to refer to methods (see sample sentences 1 and 3). Tense m ixing m ay cause confusion in the reader, especia lly if it becomes unclear w hether the reference is made to previous investigations or to the project in ques­tion, as in sam ples 1 to 3. Sudden changes m ay be the resu lt of students’ shift be­tw een general English language conventions and academ ic w ritin g practices or the influence of some noun phrase/verb phrase collocations used as fixed expressions (e.g., JtudieJ have jhau'n ) w ithout considering their functions in the specific sections of the abstracts. These uncertainties are more typ ica l o f students who are w eaker in their English language proficiency.

1. . . . D ifferences between gender groups and faculties have been exam ined more specifically. The results show evidence for the in itial hypothesis according to which age is the most determ ining variab le in the H ungarian address system . Numerous differences have been found betw een ...

2. Color-nam ing has been investigated am ong...

3. . . . I have identified and revealed numerous beliefs and practices . . . . M ore pre­cisely, I have showed the differences . . . The resu lts showed that . . .

Hedges

The frequency of the various types of hedges is illustrated in Table 3. The 20 epis- tem ic modifiers that indicate uncertain ty appear 50 times in the BTA corpus, w hile only 36 times in the RA A corpus. O nly three of them do not occur in the two cor­pora. Their d istribution is v e ry much uneven, w ith the au x ilia ry can lead ing both

364

abstract groups (w ith 44% for BTAs and 28% for RAAs), of w hich the past form could is also used b y students, but not b y researchers. The w ord m ay occupies sec­ond position, w ith m ight appearing on ly once. A ltogether the four most frequent hedges are can, m ay, indicate and tend.

Table 3: O ccurrences of hedges in the two corpora

List of hedges BTAs RAAs Total

can 22 10 32

may 4 6 10

could 4 0 4

might 1 1 2

seem 2 0 2

appear 1 2 3

assume 0 0 0

believe 0 0 0

speculate 0 0 0

indicate 3 2 5

likely 2 2 4

possible/possibly 1 0 1

tend 4 1 5

try 1 0 1

seek 0 2 2

should 0 2 2

relatively 0 3 3

sometimes 0 1 1

mostly 2 1 3

generally 3 3 6

Total 50 36 86

H edges are frequently used to indicate results. Some illustrations for the cautious interpretation of results in the thesis abstracts are provided below. Examples 5 to 7

365

are more academic and elegant and could also come from published abstracts, while the use of can is superfluous in sentence A.

A. Findings of this paper can be useful not only in .. .5. Findings presented in this paper m ay contribute to further research . . .6. This might be the reason w h y ...7. Results seem to indicate that . . .

To compare, Salager-JVLeyer (1992) reported the leading position of m ay in medical RAs; other modals had a marginal position in his corpus. In contrast, the four most frequent hedges in the Serholt (2012) Swedish student abstract corpus are m ight, deem, could and duggedt. These differences in the results also point to the importance of using small, specialized corpora to illustrate linguistic choices in different genres, specific fields and various author groups. Large corpora m ay y ie ld impressive re­sults, but m ask the individual and group differences.

CONCLUSION AND IMPLICATIONS

This study analyzed two small, specialized applied linguistics corpora to reveal the linguistic realization of abstracts w ritten by expert w riters and novice, BA student writers. It focused on the sim ilarities and differences found in the linguistic realiza­tion of 30 abstracts published in applied linguistics journals and 30 BA level ap­plied linguistics thesis abstracts written by H ungarian students of English. The re­sults indicate that thesis abstracts incorporate more finite verb clauses, but students are aw are of the genre conventions. The few instances of misuse w ere discussed in context and in terms of the rhetoric functions they have. The concordance analysis of 20 epistemic modifiers revealed that modals have a major role in indicating un­certainty in both groups of abstracts, while other hedges appear only m arginally.

The literature has called for the need for a careful selection of texts to be com­pared, and these two groups of abstracts, although having different functions, proved to be good choices. The analyses of verb tenses and hedges in humanities papers have not been in the center of attention of researchers, therefore this study aimed to fill this gap. A future direction of research m ay be the extension of analy­sis to non-linguistic thesis and to the interview and questionnaire methods used by H yland (2000) and de W aard and Pander M aat (2012). H yland (2000) rightly pointed out that the analysis of specialized corpora is a valuable tool to assist novice w riters and teachers to take control over disciplinary-sensitive w riting prac­tices. O nly this w ay can students fu lly explore the meaning of word choices and learn the acceptable and appropriate means to engage w ith their readers, to m ark their attitudes and to appear confident and expert writers. As there is no time for the explicit teaching of all tricks and steps of good academic w riting in the few hours of academic ■writing classes or thesis consultations, students should be shown

366

w ays and tools to extract academic texts w ritten about the topics they w ish to ex­plore, analyze them w ith easy-to-use tools such as AntConc or the Complete L exical Tutor, so that they can check word choices in context and discover solutions that are typical of experienced writers. This w ay it is hoped that student w riters do not plagiarize sections of texts m erely due to language shortcomings, but find good a l­ternatives for academic vocabulary boosters.

REFERENCES

Abdollahzadeh, E. (2011). Poring over the findings: Interpersonal authorial en­gagement in applied linguistics papers. Journal of Pragm atics, 43, 288—297.

AntConc. Corpus analyzer k it designed by Laurence Anthony. http:/Avww.antlab.s- ci. waseda.ac.j p

Askehave, I. & Swales, J . M . (2001). Genre identification and communicative purpose: A problem and a possible solution. Applied L inguistics, 23 (2), 195—212.

Bhatia, V. (1993). A nalysing genre: Language use in professional settings. London: Longman.

Compleat L exical Tutor. Language learning and research tools designed by TomCobb http:/Avww.lextutor.ca/vp

de W aard, A. & Pander M aat, H. (2012). Verb form indicates discourse segment type in biological research papers: Experimental evidence. Journal of English for Academic Purposes, 11 (4), 357—366.

Doro, K. (2013a). Selling their research: The linguistic realization of rhetoric moves in English thesis abstracts w ritten by H ungarian undergraduates. Ro­m anian Journal of English Studies, 10 (1), 181—191.

Doro, K. (2013b). The rhetoric structure of research article abstracts in English studies journals. Prague Journal of English Studies, 2 (1), 119—139.

Hirano, E. (2009). Research article introduction in English for specific purposes: A comparison between Brazilian Portuguese and English. English fo r Specific Purposes, 28, 240-250.

H yland, K. (1994). H edging in academic w riting and EAP textbooks. English for Specific Purposes, 13 (3), 239-256.

H yland, K. (1996a). Talking to the academy: Forms of hedging in science research articles. W ritten Communication, 13 (2), 251—281.

H yland, K. (1996b). N urturing hedges in the ESP curriculum . System , 24 (4), 477— 490.

H yland, K. (2000). Hedges, boosters and lexical invisib ility: Noticing modifiers in academic texts. Language Awareness, 9 (4), 179—197.

H yland, K. (2004). D isciplinary interactions: M eta discourse in L2 postgraduate w riting. Journal of Second Language W riting, 13 (2), 133—151.

H yland, K. (2010). M etadiscourse: M apping interactions in academic writing. Nordic Journal of English Studies. Special Issue on M etadiscourse, 9 (2), 125—143.

367

Koutsantoni, D. (2006). Rhetorical strategies in engineering research articles and research theses: Advanced academic literacy and relations of power. Journal of Englidhfor Academic Purpoded, 5 (1), 19—36.

Li, L. J . & Ge, G.-C. (2009). Genre analysis: S tructura l and linguistic evolution of the English-medium medical research article (1985—2004). Englidh fo r Specific Purpoded, 28 (1), 93—104.

Lon, C. E., Tan, H. & Abdullah, A. N. (2012). Conveying communicative intent: M oves and metadiscourse in abstract w riting. International Journal of Applied Lin- guidticd and Englidh L iterature, 1 (7), 56—65.

Lores, R. (2004). On RA abstracts: From rhetorical structure to thematic organi­zation. Englidh fo r Specific Purpoded, 23, 280—302.

M alcolm , L. (1987). W hat rules govern tense usage in scientific articles? Englidh for Specific Purpoded, 6, 31—43.

M artin-M artin , R (2005). The rhetoric of abdtractd in Englidh and Spanidb dcientific did- courde: A crodd-culturalgenre-analytic approach. Bern: Peter Lang.

M ilton, J . C. P. & H yland, K. (1999). Assertions in students’ academic essays: A comparison of English N S and N NS student writers. In R. Berry, B. Asher, K. H yland & M . Lam (Eds.), Language analydid, dedcription and pedagogy (pp. 147-161). U ST Press. Retrieved from http://repositoiy.ust.hh/dspace/bitstream/ 1783.1/1045/l/MILHYL2.pdf.

Ozturh, I. (2007). The textual organization of research article introduction in ap ­plied linguistics: V ariab ility w ithin a single discipline. Englidh fo r Specific Purpoded, 26, 25-38.

Pho, P. D. (2008). Research article abstracts in applied linguistics and education technology: A study of linguistic realizations of rhetoric structure and authorial stance. Didcourde Studied, 10 (2), 231—250.

Publication m anual of the APA 5th edition (2003). W ashington DC: American Psychol­ogy Association.

Salager-M eyer, F. A. (1992). Text-type and move analysis study of verb tense and m odality distribution in medical English abstracts. Englidh fo r Specific Purpoded, 11, 93-113.

Sam raj, B. (2002). Introductions in research articles: Variations across disciplines. Journal of Englidh fo r Academic Purpoded, 21, 1—17.

Sam raj. B. (2005). An exploration of a genre set: Research article abstracts and introduction in two disciplines. Englidh fo r Specific Purpoded, 21, 141—156.

San, L. Y. & Tan, H. (2012). A comparative study of the rhetorical moves in ab ­stracts of published research articles and students’ term papers in the field of computer and communication systems engineering. International Journal of Ap­plied Ling uid tied and Englidh L iterature, 1 (7), 40—50.

Santos, M . B. Dos. (1996). The textual organization of research paper abstracts in applied linguistics. Text, 16 (4), 481^199.

Serholt, S. (2012). Hedged and hood terd in academic w riting: A dtudy of gender differenced in eddayd written by Swedidh advanced learnerd of Englidh. Unpublished dissertation.

368

G otensborg U niversitet. R etrieved from https://gupea.ub.gu.se/bitstream/ 2077/29526/l/gupea_2077_29526_l.pdf

Sheldon, E. (2011). Rhetorical differences in RA introductions w ritten b y English LI and L2 and Castilian Spanish L I w riters. Jo u rn al of E nglish fo r Academic Pur­posed, 10, 238—251.

Stotesbu iy, H. (2003). Evaluation in research artic le abstracts in the narrative and hard sciences. Jo u rn al of E nglish fo r Academic Purposed, 2 , 327—341.

Sw ales, J . M . (1990). Genre an alysis. Cam bridge: Cam bridge U niversity Press.Tseng, F. (2011). A nalysis of move structure and verb tense of research article ab ­

stracts in applied linguistics journals. International Jo u rn al of English L inguistics, 1 (2), 27-^39.