The Application of Functional Linguistics in Exposing Gender Bias in Iranian High School English Textbooks

The purpose of the present study is to find out whether Iranian high school English textbooks are sexist textbooks which discriminate between male and female students based on their sex. Halliday's systemic-functional linguistic theory was used as an innovative framework to language gender bias studies. To this end, four English textbooks currently taught in the Iranian high schools were chosen as a sample of this study. The linguistics analysis revealed that there were more males in the Theme and Rheme position. In other word, males had dominated the communicative positions of the clauses in the textbooks. Also the outcome of the participant role analysis showed that the frequencies of males for different participant’s roles were more than that of females. Thirty males played the role of actors where as only twenty females were in the position of actor throughout the text of these textbooks. Eleven males were found in the role of senser in all clauses of the textbooks. The finding may be attributable to the writer `s attitude, culture, and traditional patriarchal system of Iranian society. Yet another reason why men are more visible than women may be related to the power of the influential policy makers (mostly men) who make curriculum decisions that favor boys and men.


Introduction
Educationalists are concerned about sexism and writers' attitude in textbooks because some textbooks may have destructive effects on students' personality.For example, preponderance of males in the math textbooks may suggest to female students that mathematics is not really for them.A biased representation of female and male can lead to students' sense of what is normal for women and men in our society.In other words, the content of the textbooks helps reinforce gender as social division and perpetuate inequalities between men and women.Experts and authorities in education also believe that the health and mental, social, cultural and scientific growth of every learner depends on a balanced and appropriate system of education.Books are part of this appropriate curriculum (Kemp, 1977, p. 17).The content of the books contains the most essential executive policy for achieving the targets of every educational system.That is why it is essential to develop accurate and scientific-based textbooks.Textbooks play important roles in determining the educational content and policy.Thus, they have been the focus of education authorities, textbook designers, researchers and teachers' attention.
Education authorities have suggested diverse criteria for selecting appropriate and efficient content of the textbooks.One of these is that the content of the textbooks should be consistent with and relevant to the sex of the intended learners (Eisner, 1985, p. 201-202).The content of textbooks is useful and efficient when it is consistent with the learner's age and sex.According to Graci (1989, p. 79), lack of equality in gender presentation is one of the frequent problems of the published books.Some researchers of ESL textbook (Porreca, 1984;Lesikin, 2001) have shown them to contain gender bias.
The problem of gender bias language has been around us for centuries and still does exist in the dot com era.Human seems to find it easier to travel through cyberspace than to eradicate the problems of gender discrimination.Gender ISSN 1925-4768 E-ISSN 1925-4776 86 biased language can have adverse and undesirable effects on the learners.Learners' identities are shaped by what they read in textbooks.Therefore, it is of utmost importance to scrutinize the textbooks given to students (Norton, 1997;Penny, 199o.).The purpose of this study is then to scrutinize Iranian Enlish textbooks for sex bias and discrimination.To be more precise, the present study seeks to investigate the relative social prominence of males and females, based on soci-linguistic theory of language: namely systemic functional linguistics.
Definition of the key terms of Halliday's theory will be given from the very outset, since this gives the readers a lead-in to the study.The terms that have to be explained here are: Theme, Rheme, last stressed element, and transitivity (processes and participants).
Theme is the most important part of the clause from the point of view of its representation of a message in a sequence (Quirk & Greenbaum, 1973, p. 412;Brown & Yule, 1983, p. 135).Theme is the 'psychological subject' or the 'subject of discourse' (Brown & Miller, 1991, p. 343).It serves as point of departure of the message, and it is that with which the clause is concerned (Halliday, 1994, p. 37).Rheme is the remaining part of the message, which develops the Theme.The last stressed element (end-focus) of a clause is also important.The person or persons occurring as the last stressed element bears or bear the information focus of clause and, like the person or persons in the theme position, has or have communicative prominence (Quirk & Greenbaum, 1973, p. 412;Brown & Miller, 1991, p. 348-349).It is important to note that there are other terms used instead of theme and rheme.These are: topic-comment, background-focus, given and new information (Brown & Miller, 1991, p. 343;Verschueren, 1999, p. 138).It is important to mention that the notion of Theme and Rheme can lead into the study of structure of text and it is not the intention here to pursue this view.'Theme', 'Rheme' and 'focus' here refer to structural position within sentence.In other words, this study is concerned with those features of word order within sentence that are related to the communicative function of sentence within texts.
The way language organizes experience for us is not restricted to the meaning of individual words.Significant orderings of experience are also carried out by the way in which we put the words together into sentences.Not only vocabulary but also grammatical construction helps to represent social realities in determinate ways.(Hopper & Thompson, 1980, p. 251;Goatly, 2000, p. 227).One important kind of grammatical construction is transitivity Transitivity is a way of describing the relationship between participants and processes in the construction of clauses; who (or what) does what to whom (or what).Transitivity relations and the roles of the participants depend upon the kind of process encoded by the main verb in clause.For English, there are four fundamental types of process: material, mental, verbal, and relational processes (Halliday, 1994, p. 109).
Material process verbs are those which describe an action or event, and answer the question what happened?The person responsible for causing the action is called actor and the one who receives the action is called the goal (Halliday, 1994, p. 109).Other functional approaches use the term agent instead of actor and patient instead of goal, although they are not always synonymous (Finch, 2000:94).The main reason for analyzing material process is to uncover whether men or women are presented as the most powerful or most frequent and active participants in the text.
Mental processes (realized by verbs such as know, feel, think and believe) involve roles such as senser (the one who performs the act of 'knowing', 'thinking', or 'feeling').The main purpose for analyzing mental processes is to uncover who is represented as feeling, thinking, or seeing individuals.Another purpose for analyzing mental process is to see whether men or women are represented to feel the same experiences or there are mental processes that women or men have never experienced.
Verbal processes are processes of saying or writing (signaled by terms such as suggest, promise, enquire, tell, inform, demand, nominate, cry, ask, pray, say, etc.).Typical participant roles are sayers, message and recipient: the person doing the saying or writing is called the sayer (Halliday, 1994, p. 138).There are two object participant roles: first, what is said, and secondly, the person that is said to.These are, respectively, the target, and the receiver.Sometimes the target and receiver are called message and recipient respectively.
There are some reasons for analyzing verbal processes.One is to see who gets to hold the floor, to have their words reported.Another is to see what kinds of effects the sayer might have on those listening, whether they come over as dominant, for example, or what speech acts they perform.Analysis of the target will also tell us the main concerns of the sayer.Therefore, by analysing verbal processes we can discover who controls who through speaking.
Relational process in its simplest form involves entities which are identified by reference to one of its attributes.The process may be realized by verbs such as become, seem, be, have, remain, stay, equal, include, contain, etc. Relational processes relate to participant roles called: token and value.Other important roles in relational processes are those of possessor and possessed.The main purpose of analyzing relational processes is to find out which group of men or women is represented more often as possessor.

Review of the Literature
The issue of sex treatment in textbooks has received considerable attention (Porreca, 1984;Peterson & Kroner, 1992;Reese, 1994;Ansary & Babii, 2003).Porreca (1984) investigated how sexism is manifested in ESL texbooks.She analysed the content of 15 widely-used ESL textbooks, focusing on the categories of (1) omission ratio of females, compared to males, in text and illustration, (2) firstness, (3) occupations, (4) frequency of male nouns to female nouns, (5) generic masculine nouns and pronouns (man/he) and ( 6) types and frequency adjectives for men and women.She found that 'there is evidence that sexism continues to flourish in ESL textbooks (Porreca, 1984, p. 718).She reported that "although females comprise slightly over half the population of the United States, they are represented only half as often as males in both text and illustration".Poulou (1997, p. 70) found that the discourse roles of men and women in dialogues of textbooks were different.In another study, Ansary & Babii (2003, pp. 67-68), demonstrated that of the total pictures of an EFL textbook 59.48% were males and 26.72% females.They believe that this 'an indication of strong prejudice against women'.More recently, using systemic-functional model as a tool for her analysis, Lesikin (2001) uncovered widespread gender bias in a published textbook, namely: Grammar in Use.She found that out of 55 gender-specific nouns and pronouns in clause, 44 were Themes.Of these, the ratio of females to males was 34% to 66%.Eleven gender-specific nouns and pronouns in clauses were Themes.Of these, the ratio of females to males was 27% to 73%.In addition, seven gender-specific nouns were last stressed element (end-focus).Of these, the ratio of female to males was 29% to 71%.
Most recently, in our country, Ansary & Babaii (2003) explored the status of sexism in current ESL/EFL textbooks.They performed two types of analyses (quantitative and qualitative) to examine the manifestations of sexist attitudes and values in two textbooks (Right Path to English I and II) which were locally designed to cater for and respond to the English language needs of Iranian students at guidance schools.They found that "women suffered most obviously from low visibility" (p.69).The ratio of females to males was 1 to 1.4 in text and 1 to 1.6 in illustrations.Of the total 40 topics in dialogues, 27 topics were found to be male-dominated.
Considering what was found in the literature and the researchers' experience, the following research questions were formulated: 1. Is the number of males (in the position of Theme, Rheme, and last stressed element) larger than that of females? 2. Do male nouns and pronouns (functioning as participants in clauses) outnumber female counterparts in all participants' roles?

Methodology
As the review of this study indicated, many gender-biased and gender stereotyping studies in textbooks carried out in 1970s and 1980s used content analysis as the most common methodology.Fewer studies used linguistics analysis, however.This study intends to use linguistics analysis as an innovative method in gender bias studies.
Since not only lexical but also linguistics choice of the writer contributes to sexist language (Cameron1990, pp.17-18), the methodology chosen for this study focuses on both vocabulary and agreement of grammatical items in sentences.This methodology is to examine the linguistics choice of the writer and is based on Halliday`s systematic -functional grammar.The methodology has been already used by Lesikin (2001).One main advantage of this methodology is that it can be applied with both extended discourse and content -reduced sentence level item.According to Halliday (1994, p. 37) language is structured to make three kinds of meaning simultaneously: experiential, interpersonal and textual meaning.One purpose of this study is to explore what the text means textually and experientially .Another purpose to examine the ideas and attitudes that the writer conveys through language.Eggins (1994, p. 78) stated that experiential and textual meaning are realized respectively through Theme and transitivity patterns of grammar.Accordingly, Halliday`s linguistics framework seemed most suitable to our linguistics analysis.

Materials
Four current EFL/ELT textbooks, i.e.English Book 1, 2, 3, (Birjandi, Noruzi & Mahmoodi, 2006) and Learning to Read English for Pre-university Students (Birjandi & Anabi, .2006)that are locally designed to cater for and respond to the English language needs of the Iranian students at high school were selected to serve as the corpus of the present study.

Data Collection Procedures
To gather the data for this study, some steps must be taken in order.First, the textbooks are selected.Then, these textbooks are examined and studied carefully in depth.This step of content analysis is called mass observation the result of which will be newly categorized hypotheses (Cohen & Manion, 1992, p. 49).In content analysis tradition, the researcher has to do two mass observations: the first one is done after choosing the documents (textbooks).The purpose of the first phase is to have a cursory look at the textbooks in order to make sure that the documents contain the data that the researcher is looking for.
Having this in mind all the clauses containing at least one gender-specific noun or pronoun (e.g.Maryam has an English book) were collected.Then, these clauses are categorized by gender and by Theme/Rheme distinction.Simultaneously, each clause is examined to see if it contains a last stressed gender-specific noun or pronoun in the same clause signaling a competitive focus of new information.If a clause contains a last stressed gender-specific noun or pronoun, the gender-specific focus is noted.Themes and rhymes which have both males and females noun and pronoun are eliminated (e.g.Ali and Maryam went to school) since co-occurring forms offset each other.After counting the clauses, the number of themes, rhymes and last stressed elements are tabulated according to gender and then number of themes and last stressed are compared to the number of rhymes.Afterwards, the roles of those gender-specific nouns or pronouns labeled as Themes and last stressed elements are determined, and they are examined in terms of participant functions.
Finally, the Themes and last stressed elements by participant function and gender are tabulated.For that matter, the quantity of themes and last stressed elements are incorporated into the results and are compared with the quantity of rhymes.

Data Analysis Procedures
To analyze the quantitative data the following descriptive statistics are used: a) frequency and percentage and b) chi-square to examine whether there are statistically significant differences between sex and investigated categories of the study.All the hypotheses are set at the 0.05 level of probability.

Results
In order to present the quantative data easily, first the frequencies and percentages were collected and summarized, and then to help explain differences in the number of males and females, chi-square test was used to ascertain if there were significant differences between the number of males and females in different categories of this research.

Thematic Structures (Theme, Rheme and end-focus)
Table 1 demonstrates the frequency and percentage of gender-specific nouns and pronouns in the positions of Theme, Rheme and last stressed element in all the clauses of the investigated textbooks.As can be seen from table 4.1, a larger percentage of males is in the position of Theme, Rheme, and last stressed element.
Out of 324 gender-specific nouns and pronouns in the clauses, 265 are Themes, like "he" in He spent the day working (English Book 1, 2006, p. 92).Of these the ratio of male to female was found to be 80% to 20%.Forty-four gender-specific nouns and pronouns in clauses are Rhemes like "Her" in The man gives her a reward (English Book 2, 1985/ 2006:4).Of these the ratio of males to females is 71% to 29%.In addition, 15 gender-specific nouns as Rhemes are the last stressed elements like "Oliver Twist" in The boy's name was Oliver Twist (English Book 2, 2006, p. 45).Of these the ratio of males to females is 74% to 26% (see Table 1).
An analysis of end-focus (last stressed element) was also conducted.The result of this revealed that out of 15 gender-specific nouns and pronouns functioning as the last-stressed element, 11 were males and only 4 were females.To be more precise, as Theme, Rheme and last-stressed element the ratio of male to female was 79% to 21% (Table 1).
Finally, Table 2 displays the frequency of participant roles of gender-specific nouns and pronouns in the clause of all the four textbooks.As shown in Table 1, 265 gender-specific nouns and pronouns in the position of Theme were found to function in five participant roles, as shown in Table 2.That is, they were actors, sensers, tokens, sayers and possessors.Males outnumbered females in all roles except that of senser.Females as senser were twice as often as males.However, 35 males were actors, more than two times that of females.In addition, there was more than twice regarding to possessor.Sixty eight percent of males functioned as possessor whereas 32% of females were possessors.Actors and sayers have the greatest number of nouns and pronouns.Twenty five males and 23 females are most evenly matched in the roles of token, the third largest role.

Participant Roles
Table 4 shows that males outnumbered their female counterparts in all participant roles.In other words, males played more participant roles than those of females.One noticeable finding in Table 4 is the fact that females outnumbered males in the role of senser, the reasons of which will be discussed in section 5.2.

Thematic Structure (Theme, Rheme and end-focus)
It was originally assumed that the number of male thematic roles (those who are in the position of Theme, Rheme, and last stressed element of clause) is larger than those of females.In order to corroborate the hypothesis, a Chi-square was applied to the data in Table 1.The outcome of this is shown in Table 3.
On the inspection of the third column of Table 3, one can see that the observed frequencies of males and females are different.The implication is that the observed frequencies of males, as it was already envisaged, were more than those of females.However, in order to generalize this conclusion, the difference between the observed frequencies of the two sexes should be statistically significant.This condition, however, was not met as the Chi-square obtained (2.54) was below the critical Chi-square (5.99).Thus, although there were more males than females in the positions of Theme, Rheme and last stressed element, the null hypothesis is sustained.Therefore, on the basis of the results of the study there does not seem to be significant difference between the observed frequencies of males and females.
The findings of the present study are entirely consistent with previous studies (Macaulay & Brice, 1997;Lesikin, 2001).Lesikin relates this thematic structure to communicative structures.The EFL/ESL writers' probable motivation was to bring men into prominence.It is important to note that a person in the position of Theme or the last stressed element (end-focus) in clause is the person in the position of communicative prominence (Quirk & Greenbaum, 1973, p. 412;Brown & Miller, 1991, pp. 348-349).It is the person who is the center of conversation, the topic in writing, or the information focus until another person is introduced or focused on or the communication or text ends.People who are the center of conversations, topics of writing, or the information focus would seem to have more social prominence than people who are not.We communicate about people who in some way interest us or are important to us; those who are not of interest are not the focus.Thus people can be perceived as having different degrees of social prominence by the degree of interest shown to them (Lesikin, 2001, p. 227).
Similarly, in a written text such as textbooks, if characters of one gender (represented by gender-specific nouns or pronouns) occur more frequently in the position of communicative prominence (as Theme or last stressed element) the underlying message to the readers is that one gender has higher social prominence than the other.They most likely present stronger, more alluring role model than those characters in Rheme position.
The dominance of males in the communicative positions can also be explained by linguistic relativity theory.According to linguistic relativity, language predisposes speakers/writers to think in particular way about themselves, other members of society and the world around them in general.So, we can claim that Persian language may be a sexist language which affects our thinking about men and women.Persian language makes it difficult for the writers to think about women in the same way as speakers of non-sexist languages.The native speakers of Persian language are led to construct a biased representation of the sex, rather than simply reflect a definite reality of the world of men and women (Goatly, 2000, p. 21).
The preponderance of males in Theme and Rheme positions can also be explained by the writers' ideology which may be based on the premise that women are the second sex and men are the first and foremost sex in society.

Participant Roles
It was anticipated that males outnumber their counterparts in all participant roles.Looking at Table 2, you can see that this is true except for senser.These findings provide evidence that males functioned more than females in different roles.However, in order to validate this hypothesis, a Chi-square test was conducted.The result of this is presented in Table 4.
The obtained Chi-square was far much greater than the critical value of Chi-square, thus, although females outnumbered males in the role of senser, the differences were statistically significant.Therefore, the last hypothesis was accepted at the 0.05 level of significance.
The greater quantity of males in the position of communicative prominence in the textbooks surveyed suggests that males have greater social prominance.The greater number of male nouns and pronouns in clauses in general reinforces these findings of dominance.The findings also suggest that fairly traditional, stereotypical roles are presented for each gender (the males as actors and sayers; the females as sensers).
Why males outnumber females in almost all participant roles can be explained by the writers' motivation and ideology.The language people use to achieve their goals is affected by the ideology of the speaker or the writer who use the language.In plain terms, ideology has a significant impact on language use.The writer's ideology and viewpoints influence the writer's style, linguistic behavior, lexical and grammatical patterns, and most importantly, the representation of different kinds of people such as political groups, women and men, the old and the young and so on (Eggins,1994, p. 10;Jaworski & Coupland,1999, p. 496).
To encode our ideology, we employ linguistic structures like transitivity, passivity, nominalisation, and sometimes sequencing (Fowler, Hodge, Kress & Trew, 1979, p. 7).For example, politicians use metaphor, metonymy, analogy, and euphemism in order to manipulate language the way they desire.In other words, they use these linguistic elements to encode their ideological biases, attitudes and values.
The writers' of the textbooks may want to convey the idea that those (men) who function as actors are strong, active and hard-working whereas those who are not actors of the sentences are lazy, incompetent and passive.For example, Gordon (1997) argues that men often consciously or unconsciously are represented as competent and active because this ideology has been entrenched in some communities and cultures (pp.51-52).
Women function more than men as sensers.This may be related to widely shared stereotypes about women.The writers of these textbooks view women as more emotional and less actors than men.Since women are responsible for caring about their children, they might be stereotyped as kind, emotional and sensitive rather than sensible.This implies that women generally cannot be sensible when the circumstances require them to be sensible.They are emotional and sensitive to take a good decision in situations where it is incumbent upon them to take sensible decisions.
Those who disagree with the proposal that women are more emotional rather than rational assert that women are not just emotional human beings but they, just like men, can grow intellectuality on the condition that they are provided with the needed conditions and facilities just like men who are not immune from the usual faults and intellectual weakness, women suffer from the same faults and weakness.As with men, women too can develop and go through the various stages of intellectual development and thus attenuate their faults and reduce the number of their mistakes if the needed educational and training facilities become available to them.
The preponderance of females as sensers may be explained by the fact that the writers treat the opposition between emotion and logic as the opposition between men and women.In this opposition men are symbols of wisdom and women are symbols of emotion.The direct result of this thinking is women's dependence on men for wisedom and logic.
Men are more than women as sayers of the clauses.This implies that women must be passive listeners and they must listen to what men say.One probable reason why men are more than women as sayers is the patriarchal morality of the wirters.The writers may believe that girls must remain quiet listeners and these are men who have the right to take the floor and are legitimate to have their words reported.Other probable reasons are the dominance of men over women, culture, and ideology of the writers.
Men outnumbered women as possessors, too.The writers may be a real fan of patriarchal system.In patriarchal system, property and possession is in the control of men.According to this system, property must pass from father to son.Women cannot often own property and they are thought to be the legal property of men.Unequal division of wealth and possession between men and women which has been institutionalized in our society for many years is forced on the writers unconsciously.Another factor attributable to this problem is the current economic and financial status of women in the country which is reflected in the language as it is.
The token has been divided evenly between males and females.This can be explained by the fact that readers, especially young students at high school, are often unaware of the participant roles that are assigned to males and females.Participant roles like actors, sensers, possessers and sayers may be more latent and hidden than token (a noun or pronoun which is described or related to another one).Thus, readers do not object to these participant roles that are assigned to men and women and they see the division of participant roles between men and women as natural and fair division even though this is not the case.However, if a thing is related to the token or an attribute is assigned to the token (value) this may be resisted by the readers because an attribute which is assigned to the ' token' is on the surface of the text and every normal reader may notice this more easily than other participant roles (Example: Maryam is lazy).Thus the writer may want to avoid the confrontation of the readers (males and females) and they divide the tokens evenly between males and females

Pedagogical Implications
The result of this research will be of great interest to syllabus designers who are interested in improving the quality of sex equality in the content and pictures of the textbooks.It is essential for syllabus designers to take into account the learners' sex equality in all forms.Indeed, the instructional objectives of an educational program are not achieved unless syllabus designers take into account the sex equality, human right and social justice.
In this regard, syllabus designers can take advantage of this study as it reveals the points that the grammatical arrangement of the sentences and thematic structure of the clauses can discriminate against females, and the pictures are gender biased pictures.Thus, a considerate syllabus designer should pay close attention to the issues of gender and equality when it comes to designing of textbooks as part of the curriculum.In brief, syllabus designers are advised to keep a balance between male and female pictures and the frequency of males and females' mentions in the textbooks.

Conclusion
In conclusion, some recommendations to this linguistic inequity and imbalance should be suggested.To solve the problems of linguistic inequality, teachers around the country have an important role.They should attempt to bring these linguistic imbalances to the textbooks writers' attention who are responsible for designing and writing the textbooks.Another solution is that the writers should attempt to let more females to function as actors, sayers and possessors.
Other possible solution to these problems is to let both men and women authors to have roles in writing and designing textbooks.Still another solution is to require the writers to present a balanced number of male and female pictures.And yet another possible solution is to establish special educational councils at the Ministry of Education which is responsible for removing sex discrimination and stereotypes from the textbooks.
Teachers' responsibility is to discuss what he or she perceives as gender bias.This discussion might lead to reflective writing or to students interviewing friends, teachers and family members on specific topics raised about gender roles, behaviors, stereotypes, or expectations.Students may also create alternative texts as language learning activities.Students, for example, might take a published text containing generic masculine forms (e.g., he mankind) and rewrite the text using more inclusive language.Another activity could have studens applying this new knowledge to their own writing.We can also counter the gender bias by supplementing the material with more evenly represented text in instances where we are comfortable with other aspects of the textbooks or cannot change the textbook for programmatic reasons.Still another solution to the problem of gender bias is to write to the publishers or speak to their representatives at professional conferences to make them aware of our discoveries of gender bias in their materials and distaste for these biases.A final, but not very practical, way to resolve this problem is to require the writers to avoid using gender stereotypes that are imposed on the writer consciously or unconsciously.
Sexism today may be viewed as no longer relevant issue or it may be considered as a minor problem which would rapidly disappear as people become aware of the problem.As this study has shown, however, this assumption has no basis in fact.We can claim that since the first study of sexism in EFL/ESL materials in the 70s, little has changed over the past four decades.This suggests that EFL/ESL teachers still, in the age of dot-coms, use materials which are loaded with sex discrimination and are fraught with unfair and inexcusable language of male-dominated societies.
It should be noted that in this study, we examined the problem of sexism in only four textbooks.Perhaps, further research on a wide range of current widely-used ESL/EFL textbooks may give theoretically sound indications of how sexism works and how it affects students' performance.

Table 1 .
Frequency of gender specific nouns and pronouns as Therme and Rheme in all the four textbooks

Table 2 .
Frequency of participant roles of gender specific nouns and pronouns in all four textbooks

Table 3 .
Chi-square test for sex category and participant roles

Table 4 .
Chi-square test for sex category and participant roles