Projecting Gender Identity in Argumentative Written Discourse

Generally speaking, writers use various resources for introducing themselves to their readers. Among these, stance and engagement discourse markers are fundamental properties which manifest the underlying interaction process between writers and readers. The present paper sought to investigate whether male and female Iranian EFL learners performed differently in terms of using stance and engagement features in their writing assignments. To this end, a corpus comprising 80 argumentative essays written by advanced learners (40 males and 40 females) were collected and analyzed respectively. Hyland’s (2008) framework of stance and engagement features including hedges, boosters, attitude markers and self-mention as well as readers’ use of engagement markers such as pronouns, directives, questions, and shared knowledge served as a tertium comprationis for comparing and contrasting the written corpora created by male and female students. For analyzing the data, a software named Hermetic Word Counter was utilized to determine the frequency of the targeted tokens. The findings revealed that male and female writers made a differential use of stance and engagement features in writing argumentative essays. The significance of the differences was further attested by the application of a chi-square statistical technique. Regarding stance-taking, it was found out that both groups followed the same patterns of stance-taking except for the use of hedges and boosters. Moreover, compared to male students, the female writers tried to create reader engagement by asking questions.


INTRODUCTION
The past few decades have witnessed increasingly rapid advances in area of research concerning academic writings in which writers tend to interact with their readers through the application of different strategies based on the appraisal of the circumstances evoking a specific kind of emotional or affective response.It has conclusively been demonstrated by concerned practionaires such as Hyland (2000) and Swales (1999) that linguistic resources utilized in a rhetorical event like writing argumentative essays can reveal the writers' stance and engagement, which play a pivotal role in our understanding of an interactive writing process.
More specifically, one of the main objectives of argumentative wring is for the writers to convey their opinions or intentions towards the targeted audience through establishing what Thompson (2001) calls solidarity and alignment and has turned into one of the most significant current research areas in academic contexts (Halliday, 1994;Biber & Finegan, 1989;Hyland, 1999;Hunston & Thompson, 2000).
So far, however, far too little attention has been paid to the area of argumentative essay writing by EFL and/or ESL student writers and the ways by which they try to actualize their linguistic resources in order to articulate their intended interpersonal meaning related to the register variables dominating the unfolding of a given text (Hood, 2004& Mei, 2008).
The actualization of interpersonal metafunction requires an intelligent use of features of text or texture.Metadiscourse markers, as an indispensable component of texture, have an important part in argumentative writings.Hyland (2009) focuses on the interactive function of metadiscourse markers which help writers to convey the intended message within a particular context of situation where the reader's cognitive context is attracted towards a particular perception (Anderson, 2003).Such a conscious attempt to direct the readers' attention along carefully planned rhetorical paths requires the application of crucial properties of argumentation such as stance and engagement.
Stance taking plays a paramount role in argumentative writing and refers to speakers or writers' attitude, perspective, point of view, or position towards what they are talking about (Hunston & Thompson, 2000).According to Hyland (2008), stance taking includes four main elements: (a) hedges, (b) boosters, (c) attitude markers, and (d) self-mentions.
Hedging is the expression of tentativeness and possibility in language use.Hyland (1995) has categorized lexical items expressing hedging into lexical verbs (e.g., appear and suggest), adjectives (e.g., likely and possible), adverbs (e.g., almost and usually), modal verbs (e.g., would and may) and modal nouns (e.g., possibility and assumption).
Alternatively, boosters are classified into three main lexical categories.First, words like obviously, clearly, and demonstrate which show the certainty of writers.Second, attitude markers like fortunately, hopefully, and remarkable which arte affective rather than epistemic.Finally, self-mention devices which refer to the use of first parson pronouns and possessive adjectives (Hyland, 2001).
On the other hand, engagement refers to bringing readers into discourse and engaging them in an appropriate way.It enlists four elements: reader pronouns, appeals to shared knowledge, directives and questions.Hyland's stance and engagement permit writers to make a balance between authority and concession with their audience (Poos and Simpson, 2002).
Although extensive research has been carried out on stance and engagement in academic writings, however, there have been no controlled studies which compare and analyzes gender differences in applying these two strategies.In other words, interactive metadiscourse has been examined in light of gender differences.On this basis, the present study aimed to investigate whether there are significant differences between male and female writers and their use of stance and engagement strategies.

Literature Review
While taking stance effectively in order to evaluate the works of others and argue for a position is essential for second language writers (Hyland, 2004), utilizing the engagement strategy is equally a great challenge for them to achieve readers' expectations of solidarity (Hyland, 2001).Stance taking has various theoretical and methodological orientations.According to Biber (1999), stance is the writers' or speakers' feeling, attitude, or assessment.Many researchers (Xu & Long, 2008) believe that stance can be materialized and implied through the use of linguistic signs and discourse markers.In point of fact, successful writers are primarily required to analyze their readers' needs by making intelligent use of linguistic resources.
A number of studies have tried to investigate the appropriate use of stance strategies in different contexts (Hood, 2006;Lancaster, 2014;Hyland, 2005).Exemplifying this would be the work by Hyland (1997), which revealed that Cantonese L2 writers lack the required resources for making effective use of boosters to show certainty.In another study, Hyland (2012) found out that the application boosters are more frequent than that of hedges in published academic writings.In a study on investigating doctoral students, Cheng and Schleppegrell (2011) also found that the targeted participants employed more assertion compared to tentative markers.
In a comparative study, Coffin (2002) and Wu (2007) have reported that stance-taking patterns are quite different in high-graded and low graded papers claiming that students whose essays had received a low rating used an incoherent evaluative stance in their writings.Similarly, Swain (2009) focusing on argumentative essays maintained that appropriate use of stance markers in such essays resulted in higher ratings.According to Eslami Rasekh and Dousti (2016), females have a higher tendency to employ hedging devices. in the same vein, Mirzapour (2016), using a corpus of 60 articles, found that hedges are more frequent in females' academic writings.In a different study, Fahy (2002) claiming there is a positive relationship between gender and boosters (e.g., amplifiers and intensifiers) reported that male students use boosters more frequently than women writers.Similarly, using a mixed method research, Ghoreyshi and Yeganeh (2015) found out that males were more inclined to use boosters in their academic writing compared to females.
Additionally, it is suggested that social identity formation is highly dependent upon gender schematic information (Yaeger-Dror, 1998).Alphen (2004) discussed the issue of stance taking and indicated that women's questions are out of submissiveness.Yazdani and Ghafar Samar (2010) have revealed that males and females use different strategies in encoding the relation between writer and reader at the level of sentences, paragraphs, and texts.However, the difference in the use of signifiers was not significant, which was in contrast with the results reported by Winn and Rubin (2001) who found that gender role affects writing style.
In addition to stance taking, engagement has also received considerable attention.Writers attempt to establish the presence of readers or what Kroll (1985) calls " a second voice".Hyland (2005) believes that the interaction between readers and writers involves a great deal of effort on the part of the writer to persuade the reader.In other words, acts of writing are not faceless and impersonal anymore.
There are some studies (Hyland, 2002& Webber, 1994) which have investigated the purposes of engagement: (a) understanding readers' needs and expectations for the purpose of solidarity and inclusion, and (b) predicting possible problems and guiding readers to particular interpretations.The existing research on stance and engagement has proliferated in recent years.So far, however, little attention has been paid to the interaction between gender and stance/engagement strategy use (Hyland, 2005).On this basis, the present study sought to address two different but complementary research questions: a) Is there any significant difference between the type and number of stance markers employed by male and female writers?b) Is there any significant difference between the type and number of engagement markers employed by male and female writers?

Participants
Eighty students (40 males and 40 females) participating in IELTS preparation courses in Gooyesh Institute, Isfahan, Iran were recruited randomly for this study.The sample was chosen because of their familiarity with metadiscourse application regarding IELTS instruction.The students were required to write an argumentative essay in four paragraphs.

Instruments
The data come from eighty argumentative essays written by males and females comprising a total number of 19845 words.In order to count metadiscourse markers more precisely, an advanced version of a software called Hermetic Word Frequency Counter was utilized.Furthermore, Hyland's (2008) taxonomy of stance taking and engagement, which is illustrated in Figure 1, was employed as the analytical framework in this study.

Data Collection Procedures
To satisfy the purpose of the study, eighty argumentative essays (40 written by males and 40 by females) were collected.The students were asked to write an argumentative essay on the following topic: With the help of technology, students nowadays can learn more information and learn it more quickly.Do you agree or disagree?
In addition, they were asked to use specific reasons and provide relevant examples to support their written responses.Factors such as transparent test instruction, time, and length of writing were strictly controlled.In order to identify the frequency of each discourse marker a software named Hermetic Word Frequency was utilized.The classification of discourse markers and specific examples are shown in Table 1.

RESULTS
The instances of stance-taking and engagement markers among EFL learners in both male and female classes were counted and listed and the F and Ps were calculated respectively.Then chi-square test was then run to investigate whether there is a significant difference between gender and stance-taking and engagement in argumentative writings of the targeted students.The results are presented in Table 1 and Figure 2.
According to Table 1, hedges were used 114 times (68.4% by females and 31.6% by males) and there was a significant difference between gender and use of hedges (p< 0.05).moreover, boosters were used 117 times by writers (35.0% by females and 65.0% by males) and difference be-tween gender and booster was significant (p< 0.05).Also, the results indicate that there is no significant difference between gender and both attitude markers and self-mentions.
According to Table 3, 24 questions were asked in the writings (68.6% by females and 31.4% by males) and there was a significant difference between gender and employing questions in writing (P< 0.05).the difference between gender and both reader pronouns and shared knowledge was not significant (p> 0.05).Use of directives was not observed at all.

DISCUSSION AND CONCLUSION
To address the first research question which addressed the preferred ways of stance-taking among males and females, Hyland's ( 2008) framework was employed.The findings of the study showed that female writers employed hedges more than males.Moreover, males were inclined to utilize boosters compared to the males.
This finding is in line with Eslami Rasekh and Dousti (2016) who found that female writers prefer to use hedging devices more frequently in comparison with the male writers.Additionally, Mirzapour ( 2016) conducted a research on

Questions
Questions asking readers' viewpoints a corpus of 60 articles and found that hedges were relatively more frequent in females' academic writing.Moreover, the findings further revealed that male writers preferred to use boosters more frequently than those employed by females.Notably, these findings are supported by Fahy (2002), who claimed that men used boosters (e.g., amplifiers and intensifiers) more frequently than women.Similarly, according to a qualitative and quantitative research conducted by Ghoreyshi and Yeganeh (2015), males were inclined to use more boosters in their academic writing compared to females.
To address the second question which was focused on different engagement strategies employed by males and females, the obtained data were analyzed based on Hyland's (2008) framework of engagement.Analyzing the frequencies related to the four types of engagement indicated that female writers engaged the readers in writing by use of more than males.This finding is consistent with Yazdani and Ghafar Samar (2010), who found out that males and females use different strategies in encoding the relation between writer and reader at the level of sentences, paragraphs, and texts.
Generally speaking, the current study adds substantially to our understanding of preferred strategies employed by male and female writers in order to take stance and engage readers in their argumentative writings.In other words, discourse markers are not single and monolithic entities, differentiated merely by the topic of writing.Instead, they play a pivotal role in projecting gender identity.Taken together, successful academic writing depends on the writers' projection of a shared context and his or her ways of taking stance effectively thorough use of appropriate discourse markers.It is generally agreed today that academic writing has lost its traditional tag, being merely regarded as an impersonal type of discourse.In point of fact, academic writing is a conscious and purposive endeavor involving a carefully designed and well-thought interaction between authors and their readers.Under such perspectivization, academic writing is not considered merely a form of text production whose main objective is to portray external reality, but rather, it is an approach to using language to acknowledge, build, and negotiate interpersonal relations.As such, writers tend to offer a credible representation of themselves and the work they produce by creating a plausible rapport with their expected readers.In other words, they intend to assess and evaluate their content knowledge by offering additional, alternative views in order to control the level of personality for making their argument as convincing as possible.
Notably, the current study is affected by certain methodological limitations.For one thing, it only addresses a relatively small sample of argumentative writings in TEFL.It would have been much better if a larger sample of participants had been investigated.Moreover, according to Tannan (1982), gender-related issues should be investigated in relation to other dynamics like age, background, class, and etc. Regardless of such shortcomings, however, the results of the current study may have important implications for both EFL teachers and learners as well as material developers.According to table 2, hedges were used 114 times (68.4% by females and 31.6% by males) and there was a significant difference between gender and use of hedges (p< 0.05).Moreover, boosters were used 117 times by writers (35.0% by females and 65.0% by males) and difference between gender and booster was significant (p< 0.05).Also, the results indicate that there is no significant difference between gender and both attitude markers and self-mentions.

Figure 2 .
Figure 2. Descriptive Analysis on Stance-taking Markers Alphen, I. (2004).How to do things with questions: international power and stance taking.Paper presented at Proceeding of Workshop on the syntax, Semantics and Pragmatics of Questions, European Summer School on Logic, Language, and Information, France.

Table 1 .
Classification of DMs with related examples

Table 3 .
Chi-square test on gender and engagement

Table 2 .
Chi-square test on gender and stance-taking