PROJECTING GENDER IDENTITY THROUGH METADISCOURSE MARKING : INVESTIGATING WRITERS ’ STANCE TAKING IN

The present study aimed at investigating gender identity through the expression of interpersonal metadiscourse stance marking. The current study investigated male and female authors' pattern of stance markers utilization, focusing on totally 60 English and Persian articles, and English articles written by Persian speakers. Based on Xu and Longs'(2008) classification, five categories of stance markers (textual, epistemic, attitudinal, deontic and causation) were identified and the frequencies of their occurrences were computed. The differences in each group were investigated separately through running chi-square tests. Regarding English articles, it was found that both male and female writers used the same pattern of stance taking except the epistemic markers. Another finding of this study was that both male and female writers followed the same pattern of stance taking in Persian articles except the deontic ones. In English articles written by Persian speakers, female writers used the same pattern as their native counterparts, while male ones were affected mostly by their native language. Attending to stance taking patterns, this article provides an informative picture which illustrates the common preferences of disciplinary community especially between male and female writers. Hence, the implications of this study can be helpful in academic writing, in assessment, and textbooks.


PROJECTING GENDER IDENTITY THROUGH METADISCOURSE MARKING: INVESTIGATING WRITERS' STANCE TAKING IN WRITTEN DISCOURSE
It is generally agreed upon that people use language to convey both referential information and "create and sustain expressive meanings" (Malinowski, 1930, p. 231).Academic discourse, as a kind of communication, is no exception.According to Hyland (2004), academic writers do not just produce texts which represent an external reality, but use language to offer a credible representation of their work and themselves and in Hyland's view, metadiscourse is based on a view of writing as a social and communicative engagement and, in academic contexts, shows the ways writers project themselves into their argumentation in order to control their interactive intentions and signal their perspectives and commitments.In so doing, writers try to convey their personality, credibility, consideration of the reader and the relationship to the subject matter and to readers by using certain devices in their texts.The devices, according to Hyland (2005), including words, phrases, main clauses and even punctuation and typographic marks, are referred to as metadiscourse.
Metadiscourse is defined by Hyland (2004), as "self-reflective linguistic expressions referring to the evolving text, to the writer, and to the imagined readers of that text" (p.133).Metadiscourse markers are one of the rhetorical tools that make a text reader-friendly and as such enable the writer to reach the audience.Vande Koppel (1985) suggests that metadiscourse conveys textual and interpersonal meanings.Interpersonal metadiscourse "helps writers express their personalities, their evaluations of and attitudes towards ideational material, show what role in the communication situation they are choosing, and indicate how they hope readers will respond to the ideational material" (Vande Koplle, 1985, p. 2-3).Textual metadiscourse helps writers relate and connect bits of ideational material within a text and make sure the text makes sense in a particular situation.Hyland (2004) maintains that even textual devices perform an interpersonal function, i.e. they represent the writers' efforts in highlighting certain features in a text to accommodate the reader's understanding and guide him towards the writers' preferred interpretations.
It is false to hold that a good writer in the native language can absolutely be an efficient writer in the target language (Kaplan, 1984).Kaplan believes that foreign students employ rhetoric and a sequence of thought while composing term papers, theses and dissertations, but they would violate the expectations of the native writer.Effective argument involves a community-oriented application of appropriate linguistic resources.That is, the way writers present themselves, negotiate an argument, and engage with their readers is closely linked to the norms and expectations of particular cultural and professional communities.Consequently, contrastive analysis of features of rhetoric is offered as one possible approach to reveal the differences between writers to discover how they tone down and organize their writings.Such an approach has the advantage that it can help writers who intend to publish in high-stakes journals in a foreign language form standards of judgment consistent with the expectation of the target language discourse community.
One of the most interesting features discussed in metadiscourse studies is stance taking.Stance is commonly understood as the expression of a speaker/writer's attitude, perspective, point of view, opinion, or position towards feeling about entities or propositions that he or she is talking about (Hunston & Thompson, 2000).In other words, stance refers to the speaker's or writer's personal judgment, assessment, and sometimes the way of persuading listeners or readers, drawing upon his own knowledge, beliefs, and/or immediate perception ( Xu and Long, 2008).Biber (2006) and Hyland (2005) have revealed that stance is used in writing more than in speech, and it is possible to infer about writers' stance and attitudes through various linguistic forms.Thus, because of the significance of stance taking in expressing beliefs and viewpoints, a great deal of research has been done so far to investigate stance taking and stance-supporting behaviors (Biber, 2006;Hyland, 2005;Xu& Long, 2008), the use of stance adverbs in press reportage (Hsieh, 2009), colloquial conversation (Englebretson, 2007), and asking questions (Alphen, 2004).
In addition, in the field of second language writing, stance taking has been examined (e.g.Camiciottoli, 2004;Ai, 2012).The importance of stance taking among nonnative speakers of English has been examined in the study done by Pishghadam and NorouzKermanshahi (2012).They compared writers' stance taking in discussion section of EFL articles written by Persian, English, and EFL speakers.
Since in academic writing, there are culturespecific patterns which reveal writers' different cultures (Swales, 1990) and stance taking is a cultural concept, it is necessary to examine the similarities and difference between stance taking used by writers of two different cultures of English and Persian represented in texts.It can reveal the impact of English as L2 or foreign language on the Iranian (EFL) academic writers' use of stancetaking in the EFL articles.Moreover, some researchers (Xu & Long, 2008) believed that advanced ESL/EFL learners are not perfectly capable of projecting their stance in English.So, concentrating on stance-taking and stancesupporting construction among ESL/ EFL writers is of importance.However, there is little research in this area in the Persian culture and there are relatively few efforts made in examining writers' stance among Persian, English, and nonnative speakers.
Looking from another perspective, stancetaking and gender identities are also closely intermingled.Based on recent theories, individuals' social identity is constructed through interactions (Coupland, 1996).Communication Adaptation Theory (Giles, Coupland, & Coupland, 1991) asserts that individuals change their linguistic choices depending on the situation and their communicative goals.They may emphasize or de-emphasize particular aspects of their identities to become closer or distance from their interlocutors.Gender schematic information is a particularly influential aspect of social identity formation (Yaeger-Dror, 1998).
Few studies have been done to investigate the relationship between stance taking and gender in written discourse.Therefore, the purpose of the present study was to investigate whether there are any statistically significant differences between male and female writers' use of interpersonal metadiscourse subcategories of stance markers in English and Persian articles, and English articles written by Persian speakers since it is hypothesized that writers' gendered role schemata may affect their linguistic choices regarding stance-markers choices.

Review of Literature
Writing is one of the channels of communication and academic writing, as one of the forms of written discourse, involves interpersonal relationship between author and addressees in an academic discourse governed by certain conventions.In this regard, Hyland (1999, cited in Bonyadi, Gholami, & Nasiri, 2012) maintains that features of discourse are always relative to a specific audience and social purposes, and the effectiveness of writers' attempts to communication depends on how much they are successful in analyzing and accommodating the needs of readers.
According to Hyland (2004), one of the important ways of representing the features of an underlying community is through the writer's use of metadisourse.Metadiscourse has been defined as discourse about discourse, intended to direct rather than inform readers (Williams, 1981, cited in Bonyadi, Gholami, & Nasiri, 2012).
A great deal of studies has examined the notion of metadiscourse in academic research articles disciplinary and cross-disciplinary (Abdi, 2002;Abdollahzadeh, 2011;Harwood, 2005;Vazquez & Giner, 2008).Harwood (2005) conducted a qualitative corpus-based study of self-promotional "I" and "we" in academic writing across four disciplines while Vazquez and Giner (2008) carried out a cross-disciplinary study of the use of epistemic doi: dx.doi.org/10.17509/ijal.v6i2.4915stance markers as hedging rhetorical strategies in research articles in English.Abdollahzadeh (2011) worked on hedges, emphatics, and attitude markers as three types of interactional metadiscourse markers in 60 conclusion sections of applied linguistics research articles.
Metadiscourse is defined here as those aspects of the text which explicitly refer to the organization of the discourse or the writer's stance towards either its content or the reader.Various definitions have been given for stance.Biber, Johannson, Leech, Conrad, and Finegan (1999) considered stance as the speakers' or writers' personal feelings, attitudes, value judgments, or assessments.The term 'writer stance' involves, among other things, the communication of assessments and value judgments concerning the described situation by appeal to evidence, expression of degree of certainty or likelihood, as well as arguments regarding the necessity or desirability of the situation obtaining.According to Biber (2006), stance expressions can convey many different kinds of personal feelings and assessments, including attitudes that a speaker has about certain information, how certain they are about its veracity, how they obtained access to the information, and what perspective they are taking.Xu and Long (2008) believe that there exists a similarity between 'stance markers' and 'linguistic signs' defined by Maynard (1993) as the information which is put forward in the events or prepositions and is coded through some exclusively functioning devices in order to describe the world or express oneself.Xu and Long identified four categories of stance markers (i.e.epistemic, deontic, attitudinal, and textual).Their findings indicated that learners showed the meaning of certainty or assertion in English argumentative writings epistemically, of causation and permission deontically, and of evaluation attitudinally in a much similar fashion with that in Chinese writings.They organized the English essays textually in a same logic with that in Chinese essays.
In another study, Lim (2009) scrutinized the use of the Chinese epistemic phrase 'Wo Juede' in conversations, concluding that it represents speaker's epistemic uncertainty.Keisanen (2006) also analyzed the forms and functions of tag questions and negative yes/no interrogatives in American English conversations to come up with how these are involved in the construction of stance and in displaying people's evaluative, affective, or epistemic point of view.Akatsuka (1999) has examined the way speakers use conditionals to indicate the relative desirability, or positive versus negative value, of a particular entity or state of affairs.More often, however, linguistic forms are studied in regard to their indication of epistemic stances.These are usually studied in reference to the speaker's certainty of the truth of a proposition (Field, 1997;King & Nadasdi, 1999), but sometimes also in reference to the speaker's evidence for a proposition (Mushin, 2001).Lastly, in conjunction with these analyses of epistemic stance, the term affective stance is often used, usually in reference to the speaker's mood or feeling (Field, 1997;Rangkupan, 2001).
In the study by Conrad and Biber (2000), they apply corpus-based methods to study the ways in which speakers and writers use adverbials to mark their personal stance and define three major domains: epistemic stance which comments on the certainty, reliability or limitations of a proposition; attitudinal stance conveying the speaker's attitudes, feelings or value judgments; style stance describing the manner in which the information is presented.In another study, Alphen (2004) also discussed the issue of stance taking, especially regarding women's questions, claiming that women not always ask questions out of dependency or submissiveness.
In addition, in the field of interpersonal stance, previous investigations of stance in student academic writing have found that argumentative texts written by more advanced writers in certain disciplinary contexts tend to be more dialogically open or interpersonally engaged than those written by less advanced writers.Derewianka (2009), for example, found that through strategic use of attributions, concessions, and counters, more advanced student writers constructed stances that were explicitly open to other voices and possibilities.Corroborating this finding, Coffin (2002) and Wu (2007) have revealed differential patterns in stance-taking between higher and lower graded papers.Less successful papers may develop a stance of unwarranted assertiveness, as Wu found in lower-rated essays written by L2 writers in geography, or they may construct an incoherent evaluative stance, as Swain (2009) found in her comparative analysis of one successful and one unsuccessful argumentative essay.
Regarding the relationship between gender identity and the use of stance markers, research studies are extremely rare.In a study conducted by Yazdani and Ghafar Samar (2010), convincing evidence was found for different strategies used by males and females in encoding the relation between writer and reader at sentence, paragraph and text levels; however, the difference in the use of specifiers by either gender was not significant; also, pronouns in nonnative females gained higher frequency of usage.
Moreover, in a study conducted by Winn and Rubin (2001), it was found that writers changed their writing styles to complement (rather than converge toward) the apparent gender role orientation of their interlocutors.
To fulfill the purpose of this study, the following research questions were addressed: 1. Is there any statistically significant difference between male and female writers' use of interpersonal metadiscourse subcategories of stance markers in English articles?2. Is there any statistically significant difference between male and female writers' use of interpersonal metadiscourse subcategories of stance markers in Persian articles?3. Is there any statistically significant difference between male and female writers' use of interpersonal metadiscourse subcategories of stance markers in English articles written by Persian speakers?

METHOD Materials
Sixty TEFL articles were selected in order to examine their discussion section.The reason for considering these articles was that, the researchers were not familiar with other fields of study.The selected articles included three groups of 20 English, and Persian articles, and English articles written by Persian speakers, half of them were written by male writers and half written by female ones.
In this study, English articles were written by native speakers of English, and Persian articles were written by Persian native speakers.Moreover, English articles written by Persian speakers included English articles written by Iranian EFL speakers.All the articles were selected from well-known national and international journals.

Procedure
In order to answer the research questions, the following procedure was pursued.A total of 60 TEFL articles were randomly selected to be compared regarding writers' gender and stance taking.
To provide a valid comparison, the first 500 words in the discussion section of articles were selected and analyzed.
Based on the model proposed by Xu and Long (2008), different types of stance markers were analyzed: 1. Epistemic stance: Appeal to assessment of the degree of likelihood concerning the described situation.In fact, epistemic stance markers represent the author's certainty, evidentiality, and likelihood.Particular examples of epistemic markers from this study's corpus are as follows: 2(1.English/epis.)This is clearly because of the larger number of students who were in higher education….
Examples of textual stance marker are as follows: (1.English/Text.)All the participants are from the same cultural background; therefore, they are expected to perform the same.
(1.Persain/Text.)‫بنابراین‬‫فعالیت‬ ‫های‬ ‫شفاهی،‬ ‫مقیاس‬ ‫کاملی‬ ‫از‬ ‫نتایج‬ ‫دانش‬ ‫آموزان‬ ‫را‬ ‫فراهم‬ ‫نمی‬ ‫.سازد‬ Together, these features convey the level of personal involvement of the writer with the text, as well as his or her moral evaluation, degree of certainty, and/or emotional perspective and response to the content of the text.
In sum, this collection of lexical items collaborates to reflect the writer's stance.This model was applied to all the 60 articles.Moreover, in order to ensure the reliability of scoring, two experts were asked to analyze data.Finally, Chi-square was used to determine the areas of differences.

FINDINGS AND DISCUSSION
To make a study of metadiscourse, this study examined native English and native Persian writers' use of stance markers in the academic articles in order to find out the differences and similarities between the use of stance markers in two different languages and culture.As a consequence, the effect of English language on Iranian EFL writers' use of stance markers in EFL articles can be revealed.
In order to find the answers to the proposed research questions, the results obtained from the analysis of English, Persian, and English articles written by Persian speakers were subjected to the relevant descriptive and inferential statistics.
The results of the study are summarized in the following tables: Table 1 shows the general pattern of utilization of stance markers in English articles.The first research question of the study asks whether there is any statistically significant difference between male and female writers' use of interpersonal metadiscourse subcategories of stance markers in English articles.In order to investigate this research question, a square test was conducted.The results are summarized in Table 2.

Figure 1.The Distribution of Stance Markers between English Male and Female Writers
A general look at the above figure and table reveals that male and female writers used "textual" stance markers more frequently than other types.Moreover, it seems that the frequency of stance markers is apparently different in male and female writers' articles.However, to analyze whether these differences are statistically significant, a chi-square test was conducted (see table 2).So far as the use of stance markers in English articles was concerned, the results of data analysis indicate that male and female English writers use stance markers (including textual, attitudinal, deontic, and causation) similarly except in the case of "epistemic" ones, male English writers used them more frequently in their writings (X2= 9.10, p< 0.05).
According to Biber, Johannson, Leech, Conrad, and Finegan (1999), "epistemic" stance markers highlight the writer's degree of certainty, which draws on the author's knowledge or belief in a proposition.Female authors do not opt for "epistemic" markers as male ones do, since they favor "relativity" rather than "certainty".This finding was based on Hyland's (2005) notion of boosters which is parallel to "epistemic" markers; their feature of allowing writers to express their certainty in what they say makes them more favorable to men rather than women.Moreover, as Crismore , Markannen, and Steffensen ( 1993) believe, in western culture, many people associate certainty as a sign of strength and deontic stance taking as a sign of weakness, probably because certainty is related to assertiveness and selfconfidence.English male authors seem to be affected by this cultural belief.To strengthen their authoritative voice in the arguments, they take benefit of employing a higher proportion of "epistemic" stance markers in their academic writings.
The analysis of English articles also reveal some similarities between male and female English writers.Both male and female writers used "textual" stance markers more often than others.One explanation for this is that writers feel responsible to engage readers, and guide them through the persuasive manner of argumentation. .This finding is also in line with Pishghadam and Norouz Kermanshahi's (2012) study.
In addition, both male and female writers used "deontic" markers in the second place in their articles, tracing its roots in "relativistic" view of western people to the world (Hofstede, 1980).
The second research question probes whether there is any statistically significant difference between male and female writers' use of interpersonal metadiscourse subcategories of stance markers in Persian articles.In order to investigate this research question, table 3 and figure 2 show the difference between Iranian male and female writers' use of different types of stance markers.

Male
Textual > attitudinal>epistemic>deontic > causation Female Textual> attitudinal> deontic> epistemic> causation Figure 2.The distribution of stance markers between Persian male and female writers As table 3 and figure 2 show, there are small differences in the performance of Iranian male and female writers in the utilization of stance markers.
However, to gain a more informative picture of the differences, a chi-square test was conducted.The results are summarized in Table 4.As the table shows, the differences are not significant except in in the case of "deontic" markers(x2= 7.6, p< 0.05), indicating that the frequency of "deontic" markers is statically different in female written articles compared with those written by males.
This can be justified along some lines of reasoning: first, consistent with Lakoff's (1975) view, women tend to use more hedges (= deontic markers) in their discourse showing their hesitation, while men tend to avoid any ambiguity and uncertainty.Since men have proved to have high ambiguity intolerance, they use more "epistemic" markers to enhance the degree of certainty in their writings or speeches.Another justification is that Iranian female writers employ "deontic" stance markers greatly in their arguments to show their humidity and respect to readers.Accordingly, it can enhance the persuasiveness of their arguments, since "deontic" markers are tactful means of gaining community acceptance and solidarity with the audience (Abdollahzadeh, 2011).Furthermore, it seems that Iranian female writers are more willing to show one of the most important features of their gender identity-constituting solidarity with others.Through utilization of higher proportions of "deontic" markers, they convey their claims and simultaneously leave a room for alternative interpretations in the community, highlighting their respect for readers' ideas.
With regard to similarities, both English and Persian writers utilized "textual" stance markers in the first place.What distinguishes them is that "attitudinal" stance markers had the second highest frequency in Iranian writers' articles.It can be explained that Iranian people-both male and female-are affected by their culture in which absolute words are not of high value and emotions and gestures are assigned the major roles (Hofstede, 1980).Another justification is that Iranian people are emotional and collective, to the extent that as Hofstede (1980, cited in Pishghadam & NorouzKermanshahi, 2012) puts it; a collective type of culture is dominant in Iran where the concepts of "we", dependence and emotionality are brought into focus.The interconnection between culture and stance taking is also emphasized in Haddington's (2005) statement that 'sociality' or 'culture' are highly enacted in human interaction.This finding is not consistent with Abdollahzadeh's (2011) claim that Iranian people favor more impersonality and less dependence on subjective presentation of their attitudes, opinions and feelings, considering it can enhance the objectivity and acceptability of their claims.
It is worth mentioning that the above finding is not consistent with Thompson and Hopper (2001) view that western speakers do not talk much about events, but rather show their identities and express their feelings and attitudes.
The third research question probes whether there is any statistically significant difference between male and female writers' use of interpersonal metadiscourse subcategories of stance markers in English articles written by Persian speakers.In order to investigate this research question, Table 5 shows the general pattern of utilization of such stance markers in English articles written by Persian speakers.Figure 3 summarizes the frequency of stance markers used by male and female participants in English articles written by Persian speakers.
To analyze whether the above differences are statistically significant, a chi-square test was conducted (see Table 6).Regarding the third research question, it was found that similar to English writers, English articles written by female Persian speakers followed the same pattern of stance markers in their articles.It indicates that they are affected by English to a great extent and try to make their writings closer to native ones.Although English articles written by male Persian speakers used "textual" markers with the highest frequency in their articles, they did not follow the same pattern of markers as their English counterparts; they used "deontic" stance markers in the second place.This can be justified in the light of the fact that they may be affected more by their own (Persian) culture rather than English one.
In general, what was common among Persian and English articles, and English articles written by male and female Persian speakers were the utilization of "textual" markers in the first place.One explanation can be that according to Reilly, Zamora, and Mcgivern (2005), argumentative texts tend to present a theme or argument and are shaped to show the logical organization of such argument."Textual" markers refer to any rhetorical strategy or textual organization used in the argumentations which represent the writer's logical thinking and the kind of position or stance he/ she is taking.

CONCLUSION
The present study investigated male and female writers' pattern of stance markers utilization.The result of data analysis revealed that, regarding English articles, both sexes used the same pattern of stance markers except for epistemic ones.Another finding of this study was that both male and female authors followed the same pattern of stance taking in Persian articles except for the deontic markers.In English articles written by Persian speakers, female writers used the same pattern as their native counterparts, while male ones were affected mostly by their native language.Moreover, the finding of this study revealed that metadiscourse subcategory of stance markers is used in all writing which indicates the universality of the concept of metadiscourse; however, the number and the type of stance markers are different based on the writers' cultures.In fact, native English and native Persian writers think in a different way since they use different interpersonal metadiscourse stance marking in their articles.
The research findings also have important pedagogical implications.First, based on Hyland's (2010) idea about interpersonality reflected in academic writings, writers try to make use of different linguistic markers to enter their 'voices' into the texts to be heard by their expected readers.To reach this goal, writers should be completely aware of the norms and common linguistic patterns prevalent in the target community for which they write.Hence, one implication that may be drawn from the findings of this study is that students' knowledge of common patterns of stance markers in target language should be emphasized and enhanced in writing lessons.For students whose dominant language is not English, there is a need to conduct instruction adjusted according to the stance taking patterns of the target language to familiarize students with frequent linguistic markers for the realization of such patterns.Furthermore and more specifically, the findings of this study may be of some interest to the writing course designers and assessment professionals to take stance markers into consideration and incorporate them in writing courses and assessment checklists.
Several methodological limitations of this study should be also mentioned.The current study was limited to an examination of a limited number of articles in TEFL.It could have been optimized if a larger sample size across different fields and various genres had been investigated.No doubt, accordingly, the generalisability of the results will enhance to a great extent.
In sum, although there might be similarities and/ or differences regarding gender-related issues, they cannot be justified and explained solely and as Tannan (1982) suggests, gender-related patterns should be investigated in relation to other dynamics like age, ethnic background, class, etc.Therefore, further investigations are recommended to enrich the understanding of gender-related issues.

Table 2 .
Chi-Square test for English articles

Table 4 .
Chi-Square Test Results for Persian Articles

Table 5 .
General pattern of utilization of stance markers in Persian articles The distribution of Stance Markers between Persian Male and Female Writers

Table 6 .
Chi-Square test results for English articles written by Persian speakers