Scholarly article on topic 'Hedging Strategies in Academic Discourse: A Comparative Analysis of Turkish Writers and Native Writers of English'

Hedging Strategies in Academic Discourse: A Comparative Analysis of Turkish Writers and Native Writers of English Academic research paper on "Educational sciences"

CC BY-NC-ND
0
0
Share paper
OECD Field of science
Keywords
{hedge / frequency / "native writer" / "non-native writer" / difference}

Abstract of research paper on Educational sciences, author of scientific article — Oktay Yagız, Cuneyt Demir

Abstract Given the use of appropriate uncertainty, the right strength of claim for the data, politeness and proper positioning oneself, though hedging is an important component of western academic discourse there appears inadequate awareness. The present study aims to reveal the tendency and preferences of Turkish and Anglo-sphere scholars in terms of using hedging strategies, and frequency of hedges and whether any differences exist in their published articles. Data corpus was composed of 100 articles published in English that equally belong to Non-native writers (Turkish) and native writers of English (Anglo-sphere). Three main parts of the articles i.e. Introduction, Discussion, and Conclusion, where hedging devices are commonly used will be analyzed through a concordance program to get the hedge frequencies. The outputs will be categorized according to a framework included 9 hedge types (Adverbs of Frequency, Quantifiers, Modal Auxiliary Verbs, Epistemic Verbs, Adjectives & Adverbs, Nouns, Conversational & Informal, Introductory phrases, Vague References). The categorized outputs will be analyzed by means of MANOVA and Mann-Whitney tests to compare Non-native writers (NNW) and Native writers (NW) in terms of the hedge frequency. Having completed statistical analyses, the function of hedges used in both groups was examined so as to interpret hedging strategies. The results will provide important insights about using tentative language strategies of NNWs and NWs of English. Further, as for nine hedging types, the results will reveal each group's hedging tendencies and differences.

Academic research paper on topic "Hedging Strategies in Academic Discourse: A Comparative Analysis of Turkish Writers and Native Writers of English"

CrossMark

Available online at www.sciencedirect.com

ScienceDirect

Procedia - Social and Behavioral Sciences 158 (2014) 260 - 268

14th International Language, Literature and Stylistics Symposium

Hedging strategies in academic discourse: A comparative analysis of Turkish writers and native writers of English

Oktay Yagiza Cuneyt Demirb

aAtaturk University, Erzurum, TURKEY bSiirt University, Siirt, TURKEY

Abstract

Given the use of appropriate uncertainty, the right strength of claim for the data, politeness and proper positioning oneself, though hedging is an important component of western academic discourse there appears inadequate awareness. The present study aims to reveal the tendency and preferences of Turkish and Anglo-sphere scholars in terms of using hedging strategies, and frequency of hedges and whether any differences exist in their published articles. Data corpus was composed of 100 articles published in English that equally belong to Non-native writers (Turkish) and native writers of English (Anglo-sphere). Three main parts of the articles i.e. Introduction, Discussion, and Conclusion, where hedging devices are commonly used will be analyzed through a concordance program to get the hedge frequencies. The outputs will be categorized according to a framework included 9 hedge types (Adverbs of Frequency, Quantifiers, Modal Auxiliary Verbs, Epistemic Verbs, Adjectives & Adverbs, Nouns, Conversational & Informal, Introductory phrases, Vague References). The categorized outputs will be analyzed by means of MANOVA and Mann-Whitney tests to compare Non-native writers (NNW) and Native writers (NW) in terms of the hedge frequency. Having completed statistical analyses, the function of hedges used in both groups was examined so as to interpret hedging strategies. The results will provide important insights about using tentative language strategies of NNWs and NWs of English. Further, as for nine hedging types, the results will reveal each group's hedging tendencies and differences.

© 2014 The Authors.PublishedbyElsevier Ltd.This is an open access article under the CC BY-NC-ND license (http://creativecommons.Org/licenses/by-nc-nd/3.0/).

Peer-review under responsibility of Dokuz Eylul University, Faculty of Education.

Keywords: hedge; frequency; native writer; non-native writer; difference

1. Introduction

Hedging can be defined as a tentative language to avoid any certainty or to mitigate the statements to able to avert possible criticism. Either you use it to distinguish facts from opinion, or "honesty, modesty and proper caution" (Swales, 1990, p. 174), hedging has been undoubtedly an important issue for all levels of academic

1877-0428 © 2014 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/).

Peer-review under responsibility of Dokuz Eylul University, Faculty of Education. doi:10.1016/j.sbspro.2014.12.085

writing. Hyland (1994) tried to draw attention to that importance by indicating the necessity of its inclusion in even textbooks on top of academic writings. That importance makes hedges as the most frequent features of writer perspective, as Hyland confirmed (1998). Even hedging devices are seen as if they were suggestive or desirable, Jordan (1997) urged its existence in academic writings as a requisite.

While the importance, even requisite, is blatantly apparent, authors' use of hedging devices still needs to be searched especially in terms of nativeness of writers because in non-Anglo-sphere academe, rhetorical persuasion does not connote hedging necessarily, and hedging the statements or claims is not an obvious consideration for many non-native writers (NNWs) (Hinkel, 2003) when compared to native writers (NWs). That difference obviously deserves to be addressed. Second, many studies used a taxonomy of hedge mostly including rhetorical devices, lexical and referential markers and syntactic markers and structures. However, as stated by Hinkel (2003) as well, prevalence of various hedges have been underestimated despite large amount of corpora of English language. Therefore, it should be kept in mind that hedging is not a device confined to a fixed pre-determined structure, and that is why different hedge types also wait to be dealt with. Another issue is as regards formality of hedging from the aspect of nativeness. Hedging, mostly used in academic writings, is also employed in casual conversations. In informal registers, informal or conversational hedges are generally represented through lexical hedges characterized by vagueness (Channell, 1994), and claimed that they have not been found in written academic corpora (Hinkel, 2003). Because informal hedging devises are mostly regarded improper for academic writings, whether informal and conversational hedges are used in academic studies or not; if so, either by native or non-native writers of English, are issues to be addressed. That NNWs have a restricted lexical repertoire of hedging devices in their L2 writings was first alleged by Hyland (2002) and then by Hinkel (2003b). But from then on, it does not sound that any study focused on the subject as regard hedge numerical variances in terms of writer nativeness.

Having completed a thorough literature review, it was concluded that even hedging is regarded "vitally important" (Jordan, 1997) for academe, it is either examined unidirectional or its data included culture-specific conclusion which could not be generalizable to other situations. For example, a study conducted with a chinese population (e.g. Hinkel, 1997, Chang, Luo, & Hsu, 2012) would provide conclusions which were valid for Chinese people because hedging and culture are intertwined (Bloor & Bloor, 1991). So, it is necessary to sustain a research in the culture where you would like to get a result associative. The literature provides a paucity of data on Turkish academics' hedging tendencies, and almost none regarding the comparison with NWs. One exception is the study which aims to reveal whether research articles (RAs) of Turkish academics written on different subjects employ different hedging strategies (Ekog, 2010). A similar study was held by Doyuran (2009) in order to compare the purposes, distribution and major forms of hedges used in RAs of Turkish engineering and ELT. However, both studies had compared only non-native writers' hedging devices. On the other hand, Algi (2012) preferred to compare hedging strategies employed in the writings of L1 and L2. What is challenging conducted to see the discrepancy between NNWs and NWs in terms of hedging usage was of Uysal (2014). But, she had only examined conference abstracts of Turkish, Anglo-American, Indian and Japanese. However the present study examined the whole article in detail by dividing it into three: Introduction, Discussion, Conclusion, which allowed readers to see sectional differences as well as hedge frequencies.

The present study acquaint the reader with hedging strategies of Turkish writers and native writers, differences in terms of hedging device frequencies, and hedge types which NWs and NNWs are prone to.

1.1. Research questions and purposes

The present study aims to achieve the following research objectives: (a) to examine the overall frequency of hedges used in the study corpora; (b) to compare the frequency and distribution of hedges in the research articles (RAs) of NNWs and NWs of English from the point of Introduction, Discussion and Conclusion sections; (c) to reveal the frequency and distribution of various forms of hedges used in the study corpora in the sense of NNWs and NWs of English.

Specifically, the study contains three research questions;

(1) What is the distribution pattern of hedges across different sections in the RAs of NNWs in

contrast with NWs of English?

(2) Do the corpora comprised of RAs of NNWs and NWs of English demonstrate any differences in the overall number and frequency of hedges?

(3) Do the types of hedges employed in the corpora disclose differences in terms of NNWs and NWs of English?

2. Methodology

2.1. Data

A corpus of total 100 research articles (RAs) written by 50 NNWs and 50 NWs of English constituted the data for the present study. The RAs were selected randomly from diverse journals on ELT. Only RAs written on ELT subjects were compiled and analyzed because hedging and other features of opinion positioning is said to differ across disciplines (Hyland, 2005). To able to see synchronical variations on the use of hedging in RAs of NNWs and NWs, RAs published in last 5 years were gathered. Not to lead any reliability concern, it was tried to compile the corpora from equi-length RAs (see table 1). Verification about author nativeness was not ensured by contacting them. Authors' status of being NNW or NW of English was presumed based on their names or nationalities. In RAs where more than one scholar is involved, the corresponding author or the first author in the affiliation was regarded as the writer of RA, hence the nationality of the first or corresponding author determined the status of nativeness of all others.

Table 1. Vocabulary numbers for each section

Introduction Discussion Conclusion Total

NNW NW NNW NW NNW NW NNW NW

Tokens 22542 22798 30147 32718 15831 16898 68520 72414

Words 3521 4072 3756 4275 2544 2917 9821 11264

2.2. Data Analysis and procedure

Although different taxonomies have been used for the categorization of indirectness and hedging (e.g. Skelton, 1988; Myers, 1989; Hinkel, 1997; Crompton, 1997; Koutsantoni, 2006) in the literature studies, hedging devices employed in the present study were determined mostly based on their function as well as Crompton's suggestions (1997):

1. Adverbs of frequency: frequently, usually, often, occasionally, weekly, rarely...

2. Quantifiers: some, a few, a bit, a good deal, many...

3. Epistemic modality verbs: can, may, might, could, be able to, must, should, need to, to be to, will, would.

4. Epistemic lexical verbs: to seem, to appear, to believe, to assume, to suggest, to estimate, to tend, to think, to argue, to indicate, to propose, to speculate...

5. Adjectives & Adverbs:

a) formal: slightly, presumably, relatively, somehow, merely, actually, partly...

b) informal (diminutives): almost, (a) little, (a) few, enough, only, pretty, quite, basically, at least...

6. Nouns: assumption, claim, possibility, estimate, suggestion...

7. Conversational & Informal: anyway, in a way, kind of, more or less, like, maybe, sort of...

8. Introductory phrases: it is our view that, we feel that...

9. Vague references: as we all know, as is known, as people say, as the reader knows...

To able to demonstrate whether any differences existed in the overall number and frequency of hedges in RAs of NNWs and NWs of English, a concordance program was employed. The RAs were inserted into the program, and lexical frequencies for both groups -NNWs and NWs- were investigated. Because the program was able to work only on lexical level, the analysis for "introductory phrases and vague references" were conducted through scanning by the researchers. In addition, it was obvious that another scanning was necessary to descend to particulars because concordance program would categorize the vocabularies only on account of frequency but not semantics or pragmatics. So each frequency was checked to prove whether the word was used as a hedge. One of the other research purposes of the present study is to purport the distribution pattern of hedges across different sections, namely Introduction, Discussion, Conclusion, in the RAs of NNWs in contrast with NWs of English. To achieve the aim, each section was analysed separately in terms of nine hedging devices. Then, all indirectness and hedging devices detected in the corpora were compiled and interpreted by the authors in order to disclose differences between NNWs and NWs of English in terms of employing types of hedges.

3. Results

3.1. Hedging across RA sections

Having examined and analysed the study corpora, the hedges used sectional across the whole data were figured (figure 1). The figure shows the general tendency of writers across 3 sections, namely Introduction, Discussion and Conclusion in using hedges in RAs of NWs and NNWs.

Figure 1. Hedging distribution across sections 600

•NNW ■NW

Introduction

Discussion

Conclusion

As seen from the figure, the general tendencies of both NNWs and NWs in using hedges are similar. The least hedge employment is in Introduction part for NNWs while the situation is the same for NWs. Despite the similarity, hedging carries some major variety in terms of hedge number. NNWs used hedging devices over 100 while the number is over 200 for NWs, which means that NWs are more prone to hedging in Introduction part in their RAs. Meanwhile, we understand that Discussion is the section in which both writer groups used hedging the most; over 400 for NNWs and over 500 for NWs. As occurred in Introduction, that indicates comparatively disposition of NWs in employing hedge structures in Discussion parts when compared to NNWs. As for Conclusion part, the case does not differ. Again NWs are over NNWs in terms of hedge number. But this time the difference in Conclusion is not as much as in other sections. As regards hedge number, table 2 provides the exact figures.

Table 2. Sectional hedge numbers

Section NNW NW Total

Introduction 177 262 439

Discussion 456 522 978

Conclusion 259 287 546

Total 894 1071 1965

As understood from the table, NNWs lag behind of NWs in hedge number through all sections; especially in Introductions. Comparatively, the difference is slighter in Discussion and Conclusion parts. The least difference in number appears in Conclusion part: 259 to 287. There exists a difference of 177 hedge devices in favour of NWs in total. If will be talked sectional, Discussion takes the lead in including hedge devices while Introduction brings up the rear for both writer groups.

3.2. Hedge types across sections

Each section in RAs of NWs and NNWs was analysed through a concordance program, then checked by the researchers, and categorized top-down in number according to their hedge types as below. Table 3 shows the different means by which NNWs and NWs show their detachment to their RAs as well as hedge frequencies.

Table 3. Total number and variances of hedges through sections

Introduction Discussion Conclusion Total „ , ,

Hedge Type NNW NW NNW NW NNW NW NNW NW

Epis. Modality Verbs 35 99 136 221 102 137 273 457 730

Adjectives & Adverbs 43 53 98 90 38 31 179 174 353

Introductory phrases 25 22 101 72 59 48 185 142 327

Quantifiers 26 23 41 39 15 28 82 90 172

Epis. Lexical Verbs 17 21 36 34 22 19 75 74 149

Conver. & Informal 17 17 24 35 11 13 52 65 117

Adv. of Frequency 7 14 11 21 4 3 22 38 60

Nouns 5 13 9 10 3 8 17 31 48

Vague References 2 0 2 0 5 0 9 0 9

Total 177 262 456 522 259 287 894 1071 1965

Thanks to the table, we understand that NWs employ hedges in their RAs more than NNWs; i.e. 1071 to 894. While NWs use hedging types like epistemic modality verbs, quantifiers, conversational &informal, adverbs of frequency and nouns more, NNWs use adjectives&adverbs, introductory phrases, and vague references. Epistemic lexical verb type appears slightly more in favour of NNWs. Epistemic modality verbs gains the upper hand with a use of 730. Among all, it is vague reference which is used the least as a hedging type; only 9 and all by NNWs but none for NWs. Both groups hold a similarity in frequency of hedging type. What we mean is both NWs and NNWs used epistemic modality verbs the most while vague reference the least. In addition, the groups were near to one another in hedge types like adjectives&adverbs, quantifiers and epistemic lexical verbs. More pertaining to adjectival&adverbial hedges, these types had been divided into two as formal and informal. Informal types are also called as diminutives in other hedging studies (Hinkel, 1997, 2003, Uysal, 2014). 78 out of 179 adjectival&adverbial hedges of NNWs were diminutives while 88 out of 174 of NWs' adjective&adverb hedges were comprised of diminutives.

Sectional speaking, aforementioned situations do not change much. While NNWs contained adjectival&adverbial hedges the most in their Introduction parts (43 times), it is epistemic modality verbs that comes the first for NWs (99 times). As can be foreseen, the least used hedge type in Introductions is vague reference for both NNWs (2) and NWs (0). Referring Discussion, both group utilized epistemic modality verbs as their premier hedging devices (136 for NNWs and 221 for NWs). Again the least one for Discussion is vague reference with the same numbers for same groups: 2 to 0. As to Conclusion, once more epistemic modality verbs come into prominence as the most used hedge type for both NNWs (102) and NWs (137).

3.3. Statistical analysis results

A MANOVA was carried out to investigate whether nativeness of the writer have an effect on hedging types categorized as: Adverbs of Frequency, Quantifiers, Epistemic Modality Verbs, Epistemic Lexical Verbs, Adjectives & Adverbs, Nouns, Conversational & Informal, Introductory phrases, Vague References. MANOVA revealed a significant multivariate main effect for nativeness of the writers (Wilks' X — .202, F (9, 90) = 39.508, p <. 001). Therefore, a series of univariate ANOVA analyses were computed in order to probe the significant main effect for each dependent variable. The results of the analysis for each hedge type are as follow:

• epistemic modality verb, there is a statistically significant difference between NNWs and NWs with respect to epistemic modality verb usage as a hedge device (F(1,98)=153.294, p< .001).

• adjective&adverb, there is not a statistically significant main effect for nativeness of the writer regarding adjectives&adverbs (F(1,98)=.548, p=.461).

• introductory phrases, there is a statistically significant difference between the groups (F(1,98)=20.310, p< .001).

• quantifiers, there is not a statistically significant difference between NNWs and NWs regarding quantifiers as hedges (F(1,98)=3.213, p=.076)

• epistemic lexical verbs, again there is not a statistically significant difference between NNWs and NWs regarding epistemic lexical verbs (F(1,98)=.039, p=.843)

• conversational & informal structures, there is a statistically significant difference between NNWs and NWs with respect to conversational & informal speech structures as hedges (F(1,98)=4.548, p=.035).

• adverbs of frequency, there is not a statistically significant difference between NNWs and NWs regarding adverbs of frequency as hedging devices (F(1,98)=3.007, p=.086).

• nouns, there is a statistically significant difference between NNWs and NWs regarding nouns as hedges (F(1,98)=8.351, p<.01)

• vague reference, there is a statistically significant difference between groups as regards vague reference as hedges (F(1,98)=10.756, p=.001)

Having completed a series of multivariate ANOVAs, significant associations were examined further by non-parametric testing (Mann-Whitney). The results prove a statistically significant difference for hedge types of modality verbs (U= 312, Z= -6,868, p<. 001), introductory phrases (U= 631, Z= -4,445, p<.001), nouns (U= 900, Z= -2,788, p=.005) and vague reference (U= 1025, Z= -3,129, p=.002) but not for conversational & informal (U= 1121, Z= -1,792, p=.073).

3.4. Hedging Types and usage

The number of different hedging words or structures used by the writers was examined and the results was provided in table below.

Table 4. Number of hedge kinds

Hedge Type NNW NW Total

Epis. Modality Verbs 7 9 16

Adjectives & Adverbs 19 23 42

Introductory phrases 59 39 60

Quantifiers 11 14 25

Epistemic Lexical Verbs 10 11 21

Conversational & Informal 7 6 13

Adverbs of Frequency 4 10 14

Nouns 3 4 7

Vague References 9 0 9

Total 129 116

Apparent from the table, hedge type which the biggest different exists is introductory phrases for both NNWs and NWs, respectively 59 to 39. Then, adjectival&adverbial hedges follow. According to the analysis results, NNWs used 19 different hedge types by using adjectives & adverbs while the figure was 23 for NWs. What is interesting is that while NNWs used nine vague references in order to make a hedging statement, NWs used none. The least hedging alternative use was with noun use, seven diversities of noun in total. In summary, it is understood from the table that NNWs used more distinct hedging words or structures when compared to NNWs. But that superiority in favour of NNWs comes mostly from Introductory phrases. If it was excluded, NWs would be ahead of NNWs.

4. Conclusion and Discussion

The present study examined general hedge tendencies of NNWs and NWs, sectional differences i.e. Introduction, Discussion, Conclusion, of NNWs and NWs on hedge usage, hedge types across data composed of a hundred RAs, and statistical differences in terms of nativeness of the writers.

As regards hedge using tendency traits, NWs use more hedges in an attempt of weakening their statements or for other rhetorical purposes when compared to NNWs. The difference may stem from some culture-specific or other pragmatic reasons. Because hedging in academic writings may not be desired by some cultures (Bloor & Bloor, 1991), hence may change proportionally from culture to culture. Although hedge using in total is in favour of NWs, that is not the case for each hedge type. It is found that NWs are ahead of NNWs in some hedge type while they were equal or beneath in some types. Accordingly, conversational&informal hedge usage, mentioned as 'discourse particles' in some studies, did not concur with Hinkel (1997) who found that native speaker and non-native speaker did not differ significantly in the use of discourse particles. That NWs used more conversational&informal hedges in the present study may purport the easiness or liberty in using English language, meanwhile NNWs are more confined to formality of the language.

In English, we know that the writer can remove the "active doing" from the meaning and, thus, soften a threat to the reader's face considerably and reduce a directive to a suggestion thanks to noun hedges (Brown & Levinson, 1987). In consistent with Hinkel (1997) and Carlson (1988) who found that native speaker and non-native speaker did not differ significantly in terms of using noun hedges, present study found a significant difference between NNWs and NWs. The higher rate of employing noun hedges by NWs in the present study but not in Hinkel's and Carlson's may be partly because of inter-cultural reasons, that is to say, Hinkel had compared Chinese and Japanese speakers with native speaker while Carlson compared Chinese writers with Anglo-American. However, the present study compared Turkish and Anglo-spheric writers. We know that culture have an importance on discourse. Both Japanese and Chinese academics bear far east characteristics but Turkish academics are westerners.

Hu and Cao (2011) examined abstracts of applied linguistics articles of Anglo-American and Chinese, and their quantitative analyses indicated that abstracts published in English-medium journals featured markedly more hedges than those published in Chinese-medium journals, which is in concordance with the present study. When the hedge number in total is taken into consideration, it is apparent that NNWs fall behind of NWs. A similar study by Yang (2013) proved native speakers' superiority in using more hedges when compared to non-native speakers. However, he found that native speakers were more superior than native ones in all hedge types but introductory phrases, which agrees with the present study. In the present study it was found that there is a statistically significant difference between NNWs and NWs in using introductory phrases. That proves that non-natives tend to use phrases and fixed expressions to mitigate their statements. According to Yang, native speakers used hedges comprised of lexical verbs, modal verbs, epistemic adjectives&adverbs, and phraseological expressions more. While the present study proves the superiority of NNWs in using modal verbs, in other hedge types the results propose no remarkable outmanoeuvring on behalf of any group except for phraseological expressions in favour of NNWs.

Crompton (1997) states that compound hedges, also called introductory phrase or phraeological expressions, are quite common in academic writings as it is in the present study. Accordingly, this type of hedge, which termed as 'distancing phrase', took the first rank in a study of which data derived from RAs of selected university students (Nivales, 2011) while 'introductory phrases' takes the third highest rank in the present study, which means that both NNWs and NWs are highly prone to using introductory phrases in their RAs to make a tentative language. Her

study resulted in that adjectival&adverbial hedges are the third most preferred hedge type as it is the second in the present study. However, there exists a sectional difference between present and Nivales' study. According to her results introduction part included more hedges in total when compared to 'conclusion', which differentiated with the present study; it is the conclusion sections which encompasses more hedges but not 'introduction'. However, that difference could be due to the studied data because her data included RAs from many scientific fields like biology, psychology etc. not only English language works. More, we know that there may be variations in the use of hedges across different disciplines within academic fields, e.g. variations might exist among different subfields (Guerra & Guerra, 2008)

Modals, particularly may, are favoured form of hedging (Mojica, 2005). In that sense, that the epistemic modality verbs take the first rank as a hedge form is not indeterminate. Both NNWs and NWs employed modality verbs at the utmost. On the other hand, the conspicuous thing was that NWs were far ahead of NNWs, and while NNWs used can at most, NWs preferred may. Similarly, Vassileva (2001) stated that the most frequent surface forms of hedging are modal verbs, and the most preferred modality verbs by English writers are may and might, as it is same with our results. More, in consistent with Vassileva's study, the present study is in accord with Clyne's (1987) study as well who found similar results in favour of NWs and modality verbs.

Another significant difference between NNWs and NWs were found in using vague reference hedges. It is a type of colloquial hedging and has the role of distancing the writer by attributing the information to an external source (Brown & Levinson, 1987), so called as 'references to assumed common knowledge'. The results showed that NNWs use vague reference hedges in their academic writings while NWs use none, which validates Hinkel's claim. According to Hinkel (2002) vague reference hedges are widespread in conversational&informal registers of L2 students' academic texts. The studies she conducted to detect the difference of frequency in using vague references of NNWs and NWs provided evidence on that NNWs are significantly more frequent than NWs.

Analyses concluded with statistically significant outputs construct a fact on the assumption that in many cases L2 writers do not employ hedging structure in tune with the rules and norms of Anglo-sphere academic discourse community, as also stated by Hinkel (2003). As indicated before, this difference could take its sources from culture-specific reasons or incompetent insights in using hedges. Some culture-based and language-based pragmatic pedagogical implications may be necessarily useful in dealing with the issue (see for implications Varttala, 1999; Lewin, 2005; Uysal, 2014; Hyland, 1998).

5. Further research and suggestions

Future research could add to these findings by examining some other particular forms of hedges like passive voices, if conditionals, which were not taken into consideration because it would be too many hedge types to interpret only through one study. Present study compiled a great amount of data but they were all from the field of ELT. So, it may be interesting to see hedge usage and frequency in other departments because there is a paucity of evidence on hedging in other fields. Another propose is that hedge studies have concentrated on academic writings existed in the literature however it would be very informative to know how other sources --in which hedging is crucial such as media, politics, etc.-- use hedges and accordingly rhetorical styles. It would also be very interesting to find out whether there is any gender difference in utilizing hedges; or whether it is possible females use more tentative, indirect, and vague language when compared to males or vice-versa. Another caveat is the various uses of hedge changing from the point of culture. Although there are some studies searching culture specific differences of hedge use (e.g. Leyla & Atai, 2008, Yang, 2013, Uysal, 2014), they mostly focused on academic writing. So there is a paucity of data in exploring culture specific hedge usage in discourses but that academic writings.

Now that hedge use is vital for academic papers, academics in prospect, who are still student, should be studied and their level of awareness on hedge importance and usage should be determined so as to put forth some pedagogical implications because there exist only a few studies conducted particularly on far eastern students (e.g. Hinkel, 1997, Nivales, 2011, Chang, Luo, & Hsu, 2012) and in few Arabic countries on Arabic students (Alward, 2011, Alward, Mooi, & Bidin, 2012). As last, nearly all studies in the literature circle around the same hedge taxonomies constructed by prominent linguists. It is wondered whether new taxonomies can be created; such as a study discards some traditional hedges and adds the use of indefinite article the as a hedge form (Lewin, 2005). Another crucial thing is that whether hedges bear any importance or are cared by the readers while hedges are so important in academic writings. It is claimed that hedges often did not mattered or noticed by readers (Hyland, 2000). Although some studies aspired that issue (Crismore & Vande Kopple, 1997a,b) there are still questions

placed in mind as regards the cognitive and affective influence of hedges in academic writings. References

Algi, S. (2012). Hedges and boosters in LI and L2 argumentative paragraphs: Implications for teaching L2 academic writing. Retrieved 8 10, 2014, from Metu: etd.lib.metu.edu.tr/upload/12614579/index.pdf

Alward, A. (2011). An exploratory study of hedges used in the EFL Yemeni undergraduates' job application letter. Frontiers of Language and Teaching, 280-287.

Alward, A., Mooi, C. C., & Bidin, J. B. (2012). Hedges and Boosters in the Yemeni EFL Undergraduates' Persuasive Essay: An Empirical Study. The Internet Journal of Language, Culture and Society, 34:1-12.

Bloor, M., & Bloor, T. (1991). Cultural expectations and socio-pragmatic failure in academic writing. In P. Adams, B. Heaton, & P. Howarth, Socio-cultural issues in English for academic purposes (pp. 1-12). London: MacMillan Publishers Ltd.

Brown, P., & Levinson, S. (1987). Politeness. Cambridge: Cambridge University Press.

Carlson, S. (1988). Cultural differences in writing and reasoning skills. In Writing across languages and cultures: Issues in contrastive rhetoric (pp. 109-137). Newbury Park: Sage.

Chang, M.-H., Luo, Y.-W., & Hsu, Y.-K. (2012). Subjectivity and Objectivity in Chinese Academic Discourse: How Attribution Hedges Indicate Authorial Stance. Concentric: Studies in Linguistics , 293-329.

Channell, J. (1994). Vague language. Oxford: Oxfor University Press.

Clyne, M. (1987). Discourse structures and discourse expectations: Implications for Anglo-German academic communication in English. In L. E. Smith, Discourse across cultures. Strategies in World Englishes (pp. 73-83). London: Prentice Hall.

Crismore, A., & Vande Kopple, W. J. (1997b). Hedges and readers: Effects on attitudes and learning. In &. H. R. Markkanen, Hedging and discourse: Approaches to the analysis of a pragmatic (pp. 83-114). Berlin: Walter de Gruyter.

Crismore, A., & Vande Kopple, W. J. (1997a). The effects of hedges and gender on the attitudes of readers in the United States toward material in a science textbook. In A. Dusak, Culture and styles of academic discourse (pp. 223-247). Berlin: Mouton de Gruyter.

Crompton, P. (1997). Hedging in academic writing: some theoretical problems. English For Specific Purposes , 16, 4, 271-287.

Doyuran, Z. (2009). Conciliation of Knowledge through Hedging in Turkish Scientific Articles. Journal of faculty of letters , 85-99.

Ekoj, A. (2010). Analyzing Turkish MA students' use of lexical hedging strategies in thesis abstracts. Hasan Ali Yücel Egitim Fakültesi Dergisi , 13:49-62.

Guangwei, H., & Feng, C. (2011). Hedging and boosting in abstracts of applied linguistics articles: A comparative study of English- and Chinese-medium journals. Journal of Pragmatics , 2795-2809.

Guerra, C., & Guerra, J. M. (2008). Discursive analysis and pragmatic metadiscourse in four sub-areas of economics research articles. IBERICA , 16:81-108.

Hinkel, E. (2003b). Adverbial markers and tone in L1 and L2 students' writings. Journal of Pragmatics , 1049-1068.

Hinkel, E. (1997). Indirectness in L1 and L2 academic writing. Journal of Pragmatics , 27, 361-386.

Hinkel, E. (2002). Second language writers' texts. Mahwah, NJ: Lawrence Erlbaum Associates.

Hinkel, E. (2003). Teaching Academic ESL Writing. New York: Routledge.

Hyland, K. (2000). Hedges, boosters and lexical invisibility. Language Awareness , 9:179-197.

Hyland, K. (1994). Hedging in Academic Writing and EAP. English for Specific Purposes , 3:239-256.

Hyland, K. (1998). Hedging in scientific research articles. Amsterdam/Philadelphia: John benjamins publishing company.

Hyland, K. (2005). Stance and engagement: a model of interaction in academic discourse. Discourse Studies , 7(2), 173-192.

Hyland, K. (2002). Teaching and researching writing. Harlow, Essex: Longman.

Jordan, R. (1997). English for academic purposes. Cambridge: Cambridge University Press.

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

Lewin, B. A. (2005). Hedging: an exploratory study of authors'. Journal of English for Academic Purposes and readers' identification of 'toning down' in scientific texts , 4:163-178.

Leyla, S., & Atai, M. (2008). A cross-cultural study of hedging devices in discussion section of applied linguistics research articles. Teaching English language and literature society of Iran , 7:1-22.

Mojica, L. (2005). Filipino authors' ways of showing detachment/commitment. Linguistic Society of the Philippines , 511-525.

Myers, G. (1989). The pragmatics of politeness in scientific articles. Applied Linguistics , 10, 1-35.

Nivales, M. L. (2011). Hedging in College Research Papers: Implications for Language Instruction. Asian EFL Journal, 35-45.

Skelton, J. (1988). Comments in academic articles. London: CILT/BAAL.

Swales, J. (1990). Genre Analysis. Cambridge: Cambridge university press.

Uysal, H. (2014). A Cross-cultural Study of Indirectness and Hedging in the Conference Proposals of English NS and NNS Scholars. In A. Lyda, & Warchal, K. (Eds.), Occupying niches: Interculturality, cross-culturality and acculturality in academic research , 179-195.

Varttala, T. (1999). Remarks on the Communicative Functions of Hedging in Popular Scientific and Specialist Research Articles on Medicine. English for Specific Purposes , 2:177-200.

Vassileva, I. (2001). Commitment and detachment in English and Bulgarian academic writing. English for Specifc Purposes , 83-102.

Yang, Y. (2013). Exploring linguistic and cultural variations in the use of hedges in English and Chinese scientific discourse. Journal of Pragmatics , 50:23-36.