? 
 ? ???????????? Heteronormativity in EFL textbooks A review of the current state of research on gender-bias and heterosexism in ELT reading material ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? Lisa Erlman Ämneslärarprogrammet 
 Examensarbete: 15 hp Kurs: LGEN1G Nivå: Grundnivå Termin/år: VT 2015 Handledare: Chloé Avril Examinator: Ronald Paul Kod: VT15-1160-005-LGEN1G Key words: EFL, (foreign) language learning, textbook(s), gender, gender-bias, sexuality, (hetero)sexism, heteronormativity ? ? ? Abstract ? Gender is a binary construction reinforced by language, and reinforcing heterosexuality. When learning a foreign language, one has the ability to reconstruct one’s norms and therefore acquire strategies for avoiding production of, and exposure to, heteronormative language. These two ideas permeate this literature review on heteronormativity in EFL textbooks. The use of ‘inclusive language’ (avoiding heteronorms) in textbooks has been extensively studied during the past three decades, but predominantly focusing on gender representation and sexism. In this review, research on gender-bias and heterosexism in ELT materials is acknowledged to see whether textbooks have become more inclusive or if they still contain heteronormative standards. Research from the past decade provides a discussion of different aspects of gender-bias and heterosexism in EFL textbooks. Most of the articles indicate a development in awareness, yet still provide results exposing bias especially in sexual inclusivity. The relationship between gender and sexuality is discussed on its own and in relation to research. It is concluded that heterosexuality is a gender norm and thus should be included in analyses of gender roles in textbooks. Finally, some suggestions for further research in this field are proposed. ??? Contents ? ?? 1. Introduction 1 ......................................................................................................................... 1.1. Gender and language 1 .................................................................................................... 1.2. Gender and foreign language learning 1 .......................................................................... 1.3. Heteronormativity and definitions 2 ................................................................................ 2. Textbooks 3 ............................................................................................................................. 2.1. The impact of textbooks 3 ............................................................................................... 2.2. The ‘real’ world 4 ........................................................................................................... 2.3. The present review 5 ........................................................................................................ 3. Gender in EFL textbooks 5 ................................................................................................... 3.1. Research and progress 5 .................................................................................................. 3.2. Dialogues and illustrations 6 ........................................................................................... 3.3. Gendered language 7 ....................................................................................................... 3.4. Gender roles 9 .................................................................................................................. 3.5. Female invisibility 11 ...................................................................................................... 4. Sexuality in EFL textbooks 12 ................................................................................................ 4.1. Sexuality in language classrooms 12 ............................................................................... 4.2. Sexuality in textbooks 13 ............................................................................................... 5. Discussion and concluding remarks 15 .................................................................................. 5.1. Methodological issues 15 ................................................................................................ 5.2. Sexuality is a gender norm 16 ......................................................................................... 5.3. Future research 17 ........................................................................................................... Reference list 19........................................................................................................................... 1. Introduction 1.1. Gender and language A FATHER AND A SON ARE IN A CAR ACCIDENT. THE FATHER DIES AT THE SCENE AND THE SON IS SEVERELY INJURED. THE BOY IS RUSHED TO HOSPITAL WHERE THE SURGEON LOOKS AT HIM AND SAYS ‘I CAN’T OPERATE ON THIS BOY, HE IS MY SON’. This famous riddle demonstrates gender-bias in language. Solutions to the riddle commonly involve it being the father of another son, or that the boy had homosexual parents thus two fathers. The simple answer that the surgeon is the boy’s mother often appears farfetched since the word surgeon is masculine and we therefore conceptualise a man. Hence, the riddle symbolises the social construction of gender through language. The difference between sex and gender is that sex is seen as a biological or physiological characterisation, and gender as a cultural or social characterisation. The idea that gender is culturally or socially constructed refers to an understanding or categorisation of roles applied to men and women. Butler (1990/1999) explores the relationship between the constructed gender and the physiological sex, and concludes that it is not fixed: If gender is the cultural meanings that the sexed body assumes, then a gender cannot be said to follow from a sex in any one way. Taken to its logical limit, the sex/gender distinction suggests a radical discontinuity between sexed bodies and culturally constructed genders. Assuming for the moment the stability of binary sex, it does not follow that the construction of “men” will accrue exclusively to the bodies of males or that “women” will interpret only female bodies. (Butler, 1990/1999, p. 10) Furthermore, this social construction is conducted and reinforced by language. My view of language corresponds with Fairclough’s (2001) claim that “[l]anguage is a part of society; linguistic phenomena are social phenomena of some sort, and social phenomena are (in part) linguistic phenomena” (p. 19). This then implies that social norms, such as the construction of gender, are (in part) linguistic phenomena. ? 1.2. Gender and foreign language learning Sunderland (2000a) notes that in second and foreign language (SL/FL) education gender is “something not always apparent, but always present” (p. 203). This due to it being a social construction and thereby a linguistic phenomenon. When learning a new language, we learn ? 1 new norms, traditions and connotations incorporated into the FL because that language is an integral part of social phenomena apart from our initial culture, thus we adapt to new norms by re-conceptualising the social-linguistic phenomenon. Perhaps then, by excluding normative discrimination (e.g. gender-bias) from the language learning environment, the understanding of the language and its social conventions would abstain from common biased norms. Sunderland (1992) claims that the major issue with gender-bias in the second and foreign language classroom is that it leads to diminished learning opportunities for less included identities, i.e. women. Nonetheless, Sunderland (2000b) also claims that there is no evidence provided for this being the case, which in turn could legitimise ideas that teachers should use sexist material and language as it would not affect the learners’ learning. What seems to be disregarded in this theory is the idea of re-identifying oneself in the new language, i.e. to be identified and to define oneself (and one’s surroundings) using a new language with new norms and new possibilities. While it might be hard, if not impossible, to establish a relationship between bias in a classroom and a learner’s gender identity (Sunderland, 2000b), De Vincenti et al. (2007) assert that “the exploration of identities is an integral, albeit not always overtly acknowledged, part of the language classroom” (p. 60). ? 1.3. Heteronormativity and definitions Heteronormativity includes a binary system of gender where gender is in direct correlation to sex (Baker, 2008), i.e. a person is a woman because she is born with female genitalia. In a heteronormative society heterosexuality is the only normal sexuality (Baker, 2008) and other sexualities are either excluded or seen as divergent. Further, heteronorms imply an alignment of biological sex, gender identity, gender roles, and (hetero)sexuality. If a person is e.g. born with the male sex, identifies as a man, conforms to society’s view of how a man should act, and is sexually interested in the so-called ‘opposite sex’, then that person follows the heteronorm. The term sexual identity occurs, parallel to sexuality, throughout the review. This to suggest its correlation with other (parts of) identities, as opposed to terms such as sexual preference suggesting sexuality is a choice. When other sexualities than heterosexuality are mentioned the term non-heterosexuality may appear. This is in order not to exclude sexualities ? 2 by only mentioning bi- or homosexuality when mentioning other sexual identities than heterosexuality. Also, when some researchers have discussed a character’s ‘sex’, it is assumed that they are referring to gender, as it seems impossible to determine a fictive character’s biological or physiological attributes. Furthermore, although my standpoint is that (normative views on) gender and sexuality are interconnected and cannot be analysed on their own, when sexism and heterosexism are mentioned in this review they are not synonymous, but refer to gender and sexuality respectively. This is to stress which of the two has been analysed in the material reviewed. The combined discrimination will thus be referred to as heteronormativity. ? 2. Textbooks 2.1. The impact of textbooks In my experience, EFL (English as a foreign language) - or other language - textbooks are not as dominant in the educational setting as e.g. textbooks in mathematics. However, a survey of Swedish teachers’ use of textbooks shows that a majority of EFL teachers use textbooks (almost) every lesson (Skolverket, 2006b). This suggests that the question of their impact is relevant. I might concur with Sunderland’s view that a teacher’s ‘talk around the text’ or mediation of a text could lead to a completely new understanding of the text, and that how the teacher uses the text is therefore more important than the text itself (e.g. Pawelczyk et al., 2014; Sunderland, 2000a; 2000b; Sunderland et al., 2000). However, as it is quite common to use EFL textbook material as homework (Skolverket, 2006b) - a situation where a teacher cannot comment on the text, for researchers to look at the text itself may not be “a fruitless endeavour” (Sunderland, 2000b, p. 154). Mustapha (2013) recognises that “textbooks are often viewed by learners as authoritative” (p. 455), which would lead to students having great respect for the language - and perhaps also the norms - portrayed in their textbooks. It seems probable that learners view textbooks as authoritative, as they have been planned, written and revised with the learner in mind. While Sunderland (2000a) might be correct in believing that a text can be read and understood in a range of ways even by the same reader, the text could still be seen as reliable by that reader. To Mustedanagic (2010) this implies that a teacher must try to predict possible ? 3 interpretations of texts and how a text may affect students, hence we need to avoid using bad or biased texts: If we, for a moment, compare teaching language with building a house, the text represents our material. It can be dangerous to use bad materials for building, because there is always the chance that the house might collapse. The same thing can be applied to English textbooks. (Mustedanagic, 2010, p. 2) ? 2.2. The ‘real’ world An unresolved question concerning textbook material is whether gender representations in textbooks should reflect the current state of affairs “in their contemporary society or the desired/expected gender balanced society that is far from what learners experience in their day-to-day life” (Mustapha, 2013, p. 458). That is, as societies and cultures are full of norms, it seems impossible for textbooks to both portray these societies and cultures accurately and present a gender-balanced/neutral society with positive role-models. However, gender-bias such as occupational roles and invisibility is often worse for women in textbooks than it is in the society in which the textbook is used or on which it is based (Jones et al., 1997). Barton and Sakwa (2012) found that the occurrence of women in the investigated textbook was 35.7%, which does not reflect the depicted society with 51% women. Thus, it seems as if textbooks mirror the normative society more than they mirror the ‘real’ world. An argument in favour of textbooks mirroring the ‘real’ world is that if a textbook does not reflect the society or societies of the target language, cultural knowledge could disappear which could lead to miscommunications: The knowledge of discourses of gender and sexuality dominant in a particular culture could become crucial in deciding whether a particular utterance is a polite compliment, a light-hearted joke, or an attempt at sexual harassment. The lack of such knowledge may lead to miscommunication and negative attitudes toward the target language. (Pavlenko, 2004, p. 60) Another objection to the exclusion of discriminating language is that it would lead to worse literary quality, learner vocabulary and learner motivation, because it would restrict learners’ access to classic literature and social situations (Mustedanagic, 2010). Ergo, only using ‘inclusive language’ would mimic George Orwell’s novel Nineteen Eighty-Four in its exclusion of certain language. Nevertheless, it is most probable that teachers would ? 4 commonly use normative language, and that the learners read literary works apart from their coursebook; to use a textbook with ‘inclusive language’ could therefore broaden the learners’ possibilities in language use, where they are able to use both exclusive and inclusive language. Moreover, Sunderland (2000b) mentions that as we continue to use the same textbooks, society moves on. Such that, since textbooks are often used for longer periods of time, if we use textbooks reflecting a desired society, society will progress until the point where the book mirrors it. Thus, most researchers seem to agree with the notion that textbooks should be more gender-balanced. ? 2.3. The present review Research on, and acknowledgement of, gender-bias in textbooks developed in the late 1970s. Since then, ample studies on the subject have been conducted, yet very few researchers analysing ELT (English language learning and teaching) textbooks have looked at heterosexism and heterosexuality’s connection to gender norms. Two reports made by the Swedish National Agency for Education on textbooks in several subjects, show that textbooks are used extensively (Skolverket, 2006b), and that they are both sexist and heterosexist (Skolverket, 2006a). In this review, research on gender and sexuality in ELT, textbooks, and EFL textbooks is examined to see 1) how well textbooks have progressed since the topic arose, and 2) how well researchers have considered all the dimensions of heteronorms when analysing gender and sexuality in textbooks. To see current global trends in heteronormativity in textbooks, most of the research and literature reviewed is from the past decade and from different geographical regions. In the following two sections, gender and sexuality in textbooks will be discussed separately because most researchers have discussed the topics separately. ? 3. Gender in EFL textbooks 3.1. Research and progress There seems to have been a great development of consciousness amongst the authors of ELT textbooks. When research flourished in the 1970s and the 1980s, the results showed poor representations of women in textbooks, more specifically in “visibility, stereotyping of personality traits and occupational roles (including illustrations), and derogatory ? 5 treatment” (Sunderland, 2000a, p. 212). The results nowadays do not always indicate as poor representations, and some even report equality (e.g. Jones et al., 1997), “indicating […] increased awareness of gender issues in recent years” (Litosseliti, 2006, p. 87). Notwithstanding, the great amount of research carried out has not led to complete equality between women and men in textbooks. Firstly, it has been noted that there is a lack of data from developing countries (Yagoubi-Notash & Kooshavar, 2013). Secondly, some research does not report improvement in textbooks; Ullah and Skelton (2013), for example, noted that their results were in line with results from a similar study carried out 10 years earlier. Recent studies on gender and sexism in ELT textbooks have looked at different aspects of gender representation: e.g. dialogues, illustrations, gendered language, gender roles and female invisibility (e.g. Bahman & Rahimi, 2010; Barton & Sakwa, 2012; Holmqvist & Gjörup, 2007; Jones et al., 1997; Ullah & Skelton, 2013). Research on these aspects will be discussed in detail below. ? 3.2. Dialogues and illustrations Researchers studying gender equality in dialogues have looked at how many female and male characters there are, and the frequency of spoken instances amongst these characters, as well as differences in speech between the characters. These have been considered important as it is likely to find a correlation between women’s and men’s language learning opportunities and the text (Jones et al., 1997), since a dialogue could be used for oral practice, where it is plausible for women to be handed the role of the female character and men the role of the male character (Jones et al., 1997; Sunderland, 2000b). The studies on dialogue equality have shown positive findings, with nearly no gender-bias in either direction (Hall, 2014; Jones et al., 1997; Yagoubi-Notash & Kooshavar, 2013). Jones et al. (1997) looked at three ELT textbooks to see the number of male and female characters; the number of appearances of the characters; which character initiated mixed-gender dialogues; the number of turns taken by female and male characters; and, how many words that were spoken by female and male characters. The three textbooks, chosen for their accessibility and multitude of dialogues, provide a difference in results. One shows equality in the number of characters, whilst the other two show fewer male characters; also, two of the textbooks have more dialogues initiated by women and have more dialogue words spoken by women. In general terms, the ? 6 three textbooks have little gender-bias in either direction, so, even if there are some minor inequalities in individual textbooks, the authors conclude that the textbook creators have addressed the issue. One study looking at the number of cross gender conversations and the number of words spoken by women and men in two textbook series (Yagoubi-Notash & Kooshavar, 2013), and one looking at the number of female/male oriented topics in dialogues in two textbooks (Hall, 2014), found similar results with close to no bias overall. However, it is not clear what Hall (2014) means by female/male oriented topics. It could be the number of female/male characters, as Hall comments the results by saying that it was the first time “it was observed that females outnumbered males”. Even so, by saying ‘topic’ it seems to refer to what is discussed, such that the analysis itself appears gender-biased in associating specific topics to women and men. Hall (2014) also looks at another important aspect of textbooks, namely illustrations, e.g. photographs and cartoons, and their frequency of representation of women and men. Contrary to the dialogue findings, illustrations were dominated by male characters (Hall, 2014); these results conform to most research on occurrences of female and male characters in illustrations (Barton & Sakwa, 2012; Holmqvist & Gjörup, 2007; Lee & Collins, 2009; Mustedanagic, 2010). Moreover, some researchers have also looked at what characters of different genders look like in the illustrations or what they are depicted doing (e.g. Holmqvist & Gjörup, 2007; Lee & Collins, 2009; Mustedanagic, 2010). The results indicated tendencies for illustrations to depict what could be seen as stereotypical gender roles, such as women being shown wearing skirts, dresses and make-up, and men being shown wearing trousers, shirts and ties (Mustedanagic, 2010). Also, male characters dominated illustrations of activities such as working, playing and personal activities, whereas female characters were more prominent in illustrations depicting housework (Lee & Collins, 2009). In sum, textbook illustrations portray traditional binary gender roles whilst dialogues appear unbiased. ? 3.3. Gendered language Gendered language refers to linguistic features such as male-biased terms e.g. bachelor/ spinster, male-generics e.g. man instead of human, and titles e.g. Mr, Mrs (Bahman & Rahimi, 2010; Lee, 2014). In gendered language, women are given words with negative ? 7 connotations (e.g. spinster), and men are seen as the norm where masculine words are used in whole (e.g. male-generics) or as part of (e.g. Mr/s) the female word construction. The problem with these linguistic features is therefore that men are portrayed as dominant and women as subservient (Porreca, 1984). This could be the case even when one chooses a modern version of the bachelor-spinster word pair and says bachelorette - where the male word is used but with a suffix to demonstrate that it refers to women. The reason for these male-biased terms, or rather their suffixed counterparts, is wanting not to use male-generic terminology as it would lead to invisibility for women (Mills, 2008). In Mustedanagic’s (2010) study of male-biased terms in textbooks, it is clear that the frequency of gendered word-pairs has decreased in later years. The research presents results from two EFL textbooks, one from 1984 and one from 2000, where the former has a ratio of 65:33 in gendered:gender-neutral words, and the latter a ratio of 33:38, hence showing great improvement. Lee and Collins (2009) demonstrate similar improvements and say that, in several of the textbooks they reviewed, gender-inclusive terms were used when the character’s gender was not specified. Thus, gender-neutral terms such as police officer (cf. policeman), flight attendant (cf. steward/ess), and principal (cf. headmaster) should be, and are increasingly, considered (Lee & Collins, 2009; Sunderland, 2000a), as they would include all gender identities. Use of male-generics such as man instead of human was rare even three decades ago (Porreca, 1984). It should, however, be something that one considers when writing and using a textbook, as there are still occurrences according to recent studies (e.g. Barton & Sakwa, 2012). When looking at male generic forms in textbooks, most researchers have counted the number of appearances of ‘he’ as a generic construction. The research shows diverse results, where one study found one single occurrence of the generic ‘he’ in ten textbooks (Lee & Collins, 2009), another study found a “frequent use of the male pronoun ‘he’ when the [character’s gender] is unstated” (Barton & Sakwa, 2012, p. 182). Even so, the trend in textbooks is to try to avoid the male-generic forms (e.g. Barton & Sakwa, 2012; Lee & Collins, 2009; Porreca, 1984) by using a generic they or mentioning both males and females saying she or he or s/he. Despite the grammatical incorrectness of using the generic they, it has become more acceptable in both speech and textbooks (Lee & Collins, 2009), as it could feel unnatural to use she or he or her or his etcetera (Barton & Sakwa, 2012). Especially in ? 8 spoken language, e.g. in textbook dialogues, it could feel clumsy to say: ‘he or she went to the market, didn't he or she’. It has been argued, however, that a character’s gender is clear from the context, and thus it is clear when man is used as a gender-specific term (e.g. man for male person) and when it is used as a generic term (e.g. man for human being) (Porreca, 1984). Nonetheless, this has been criticised as one rarely conceptualises women when hearing the male-generic term (Porreca, 1984), which is clarified by using the female equivalent as a generic term (i.e. woman instead of human being) as it enhances how gender-specific the male generic form is (Mills, 2008). Occurrences of titles have not been extensively researched. However, Lee (2014) found that there was a greater occurrence of the male title Mr than any female titles, and that Ms and Miss occurred in textbooks, whilst Mrs did not. In toto, gendered and gender-biased language in textbooks has decreased in later years. ? 3.4. Gender roles Gender roles, presumably the most common way to identify gender-bias in everyday life, also occur in textbooks. These roles are occupations, activities, and characteristics such as emotions and looks which we tend to associate with a specific gender. In general, textbooks seem to reflect a rather traditional society, where men have powerful and prestigious occupations such as judge, doctor, and military officer, whilst women have lower status occupations (if any) such as teacher, nurse, and receptionist (Hall, 2014; Holmqvist & Gjörup, 2006; Litosseliti, 2006; Ullah & Skelton, 2013). A recent study carried out a qualitative analysis of gender roles in two (at the time) currently used textbooks in Iranian secondary schools (Hall, 2014). The results show how one of the textbooks present men in a wider range of jobs and overall higher status jobs, whilst the other textbook analysed presented “an equal number of job possibilities [for men and women] with almost the same ranking as each other” (p. 258). Most of the jobs Hall (2014) qualifies as having the same status seem valid, as some examples are English teacher and teacher and farm worker and farmer for women and men respectively, however, there is also the example of nurse and doctor, which could seem to indicate a lower status for the woman and more importantly stereotypical gender roles. As the later results were gathered from the more gender-balanced textbook, it seems that even in ? 9 recent studies of current textbooks there are some stereotypical careers applicable to each gender. Furthermore, Hall (2014) found that work was not the only stereotypical gender role appearing in textbooks. Domestic chores were also divided between the genders in a stereotypical manner, where women dominated the roles in household responsibilities such as cooking and cleaning. Several studies have shown similar results when analysing the division of domestic labour (e.g. Barton & Sakwa, 2012; Shardakova & Pavlenko, 2004; Ullah & Skelton, 2013). Some researchers have also analysed women’s roles in textbooks and seen that women are portrayed through their relationships to men (Shardakova & Pavlenko, 2004; Ullah & Skelton, 2013). Shardakova and Pavlenko (2004) looked at gender-bias in two Russian textbooks which had previously scored positively in studies on gender-bias in textbooks. The researchers reported that women were portrayed as a man’s wife, a professor’s neighbour, and a son’s mother; if this is the only way in which female characters are defined, the textbooks surely should not be seen as unbiased. Moreover, a Pakistani study of, inter alia, EFL textbooks, also reported that female characters were portrayed in relation to men, whilst men were portrayed as the head of the family (Ullah & Skelton, 2012). A study of Australian English textbooks, however, saw that domestic roles were distributed to both women and men equally, and that not only women were mentioned in relation to their role in the family (Lee & Collins, 2009). However, in general, men’s leisure activities were portrayed as more active (e.g. Lee, 2014) where they e.g. played more sports than women (Ullah & Skelton, 2012). Characteristics associated with male and female characters have been analysed by looking at what adjectives were used to describe them (e.g. Barton & Sakwa, 2012; Lee, 2014). Researchers have found that descriptions of physical appearance in textbooks portray men as muscular and strong, whilst women are described by their physical shape and attractiveness (Lee, 2014; Ullah & Skelton, 2013). Women are also portrayed as more emotional and weaker than men (Barton & Sakwa, 2012; Lee, 2014); a recent study looking at a Ugandan EFL textbook recommended by the Ministry of Education found that nine out of eleven adjectives used to describe women were emotive (Barton & Sakwa, 2012). In sum, gender roles appear to still reflect a conservative and heteronormative society. ? ? 10 3.5. Female invisibility The most common gender-bias in textbooks and reading material, seems to be female invisibility, i.e. the dominant occurrence of male characters and lack of female characters. This could include male-generic terms, as they lead to less visibility for women, but also the actual frequency with which characters of different genders occur. When examining gender visibility, researchers have counted the occurrences of gendered nouns (e.g. man, girl, Ms, father) and found that male nouns are dominant in textbooks (Bahman & Rahimi, 2010; Lee, 2014; Ullah & Skelton, 2013). Several researchers also chose to count names and pronouns (e.g. Bahman & Rahimi, 2010; Ullah & Skelton, 2013), which could be problematic since names do not need to be gendered and because (male) pronouns can be used as generic constructions. Holmqvist and Gjörup (2007) analysed the presence of men and women in illustrations of (at the time) current and out-dated Swedish textbooks to see if there had been any progress. They found that in a textbook from 2004 the ratio of male to female dominated illustrations was 17:9, which shows female invisibility but still an improvement from a textbook from 1975 which had the ratio 24:3. Hence, male characters still dominate textbooks in texts and illustrations (Bahman & Rahimi, 2010; Barton & Sakwa, 2012; Holmqvist & Gjörup, 2007; Lee, 2014; Ullah & Skelton, 2013). When women are omitted “the implicit message is that women’s accomplishments, or that they themselves as human beings, are not important enough to be included” (Porreca, 1984, p. 706). Even if visibility, or lack thereof, is still an issue, it seems to be improving (Holmqvist & Gjörup, 2007; Mustedanagic, 2010), and appears not to be an issue in textbook dialogues (Jones et al., 1997; Hall, 2014; Yagoubi- Notash & Kooshavar, 2013). There are also less obvious versions of female invisibility, such as male-generics and firstness. Firstness implies that the one gender is stated first: Romeo and Juliet, Husband and Wife, He or She, and Ross and Rachel. Researchers looking at firstness have seen a large dominance in male first-place occurrences (Bahman & Rahimi, 2010; Lee, 2014; Lee & Collins, 2009), the only exception mentioned is where women are mentioned as family members, e.g. mum and dad, suggesting that the premium place for women is in the home (Lee, 2014). ? ? 11 4. Sexuality in EFL textbooks 4.1. Sexuality in language classrooms Kehily (2001) claims that “[w]ithin pupil cultures sexuality takes on particular significance and is frequently linked to gender identity and issues of gender appropriate behaviour” (p. 119). She exemplifies this through an interview with two sixth-form students (age 15-16) where the students discuss name-calling. They agree that young men who work hard and have an interest in academic achievement get called ‘gay’, which shows how some sexual identities (such as being gay) are often seen as refraining from social norms, and that they are also strongly connected to a person’s gender as the ones who get called ‘gay’ are young men, not women. Similarly, Nelson (1999) observed an ESL class discussion with the aim to see and understand how sexual identities were mediated and negotiated in classroom discussions. When the class discussed if two men could hold hands walking down the street, students quickly began talking about these men’s sexuality. Meaning that a social phenomenon seems to regulate (the comprehension of) a person’s sexuality. Therefore, Nelson claims that the cultural framework produces acceptable sexual identities, i.e. heteronormativity. In another study, Nelson (2008) observed an ESL class during discussions of homosexuality. Smaller group discussions were followed by whole-class discussions of experiences of gay and lesbian people; the students answered or discussed questions provided by the teacher. Nelson acknowledged that students are “routinely coded as straight - or as having no sexual identity” (p. 136) in educational practices. She saw how this was reinforced by a teacher during the discussions, as the students were asked to talk ‘about’ gay or lesbian people, thus seeing non-heterosexualities as the other or different. Nelson further observed how the teacher continuously mentioned only homo- or heterosexuality, excluding other sexual identities. One could assume that these word choices were not aforethought, as one does not always consider implications in advance of what one says. Textbooks could therefore be helpful as they are revised, thus: To make the classroom a safe and caring environment for all, both homophobia (a prejudice) and heterosexism (systematic discrimination) need to be addressed. The first step in this process might be to ensure that materials evaluations include homophobia and ? 12 heterosexism in their criteria, because sexual identity is already an integral part of ESL. (Dumas, 2008, pp. 6f) ? 4.2. Sexuality in textbooks When looking at sexuality in textbooks, most researchers look at the occurrences of sexual diversity and what types of occurrences there are. Gray (2013) looked at 10 different global EFL textbooks with two main questions in mind: “Are there any representations of clearly identified LGBT characters in these textbooks?” (p. 47) and “Is there any treatment of a topic related to sexual diversity […] or the teaching of lexis related to sexual diversity […]?” (p. 48). Sadly, he found that no mention of non-heterosexualities was present in the textbooks. There were “short texts about gay figures” (p. 49), but no mention of the figures’ homosexuality; the non-heterosexual figures (e.g. Elton John) were only mentioned because of their celebrity, not in reference to their sexual identity. Moreover, in research on gender in EFL textbooks, it is described how different domestic roles are positioned e.g. mother and father, and husband and wife (Lee & Collins, 2009). Even if these are not analysed with sexual identities in mind, it comes across as heteronormative when no other constellations (e.g. daddy and dad) are mentioned. Research on sexuality in FL textbooks other than English, namely French, Italian and Japanese, was conducted by De Vincenti, Giovanangeli and Ward (2007). They looked at queer pedagogies in language classrooms in an Australian university, and therefore also at how sexual identities were portrayed in textbooks used in universities. The authors identify French as a language “used in a number of countries where the cultural contexts differ” (p. 62). Since English in the same manner is spread throughout the world, it would only seem natural to look at the analysis of reading material in French to compare to EFL textbooks. The problems the authors found interesting in French textbooks were: visibility of sexual identities; how sexual identities are portrayed; and, whether they reflect a “wide cross-section of cultural contexts” (p. 62). The research shows how textbooks aim to be sexually inclusive by not mentioning traditional families or by including both hetero- and homosexual references. Hence, the authors recognise that “there have been conscious changes made to the way sexual identity is presented” (p. 64) in textbooks, but also conclude that “simply including representations and information on non-heterosexuals in materials does not ? 13 necessarily provide affirmation of diversity, but can serve to reinforce the isolation of a social group by portraying them in a negative light” (pp. 64f). Similar studies have been conducted on textbooks used in other subjects than EFL, e.g. social sciences, biology, and religion (e.g. Røthing & Svendsen, 2011; Temple, 2005). Temple (2005) looked at “20 texts currently approved for French Québec secondary schools in five subjects” (p. 279), even if none of these subjects are language learning, one could assume that the norms are similar. Meaning that, the way of expressing sexuality in biology and religion textbooks would appear different from EFL textbooks, but the underlying norms of the textbooks should be the same as they are based on the same society and its values. Furthermore, as these texts have been approved by the Ministry of Education, texts for language learning would supposedly follow the same norms. Temple found that 95% of the pages covering sexuality made no reference to other sexualities than heterosexuality, thus the issue of visibility is very much present in sexuality as well as gender. Moreover, she found that heterosexism expressed itself in four manners: through fixed dichotomy between hetero- and homosexuality; through heterosexuality as the only ‘normal’ sexuality; through portraying other sexualities than heterosexuality as abnormal; and, through a fixed distinction between male/masculine on the one hand and female/feminine on the other. A Norwegian study (Røthing & Svendsen, 2011), however, found that “sexuality is referred to as free-spirited, liberal and voluntary” (p. 1959) in the 8 social science textbooks analysed. Through examples of ways in which textbooks portray sexuality they show that there seem to be mentions of positive views on non-heterosexualities. Even so, the authors do not mention to what extent these views were apparent in the textbooks, merely that they were addressed. This could imply that the non-heterosexualities mentioned were only present on a few pages, which would lead to the same problem of visibility Temple (2005) saw in textbooks. Nevertheless, the authors recognise the mentioning of non-heterosexualities as an expression of sexual liberalism (Røthing & Svendsen, 2011). It seems that heterosexism is present in current textbooks, even in more sexually liberal cultures. Even if there are findings of non-heterosexual mentions, the results do not show extensive spread. Additionally, the only research found on sexuality in EFL textbooks shows no mention of non-heterosexualities, thus, whilst there are tendencies of sexual liberalism in textbooks, EFL textbooks still appear heterosexist. ? 14 ? 5. Discussion and concluding remarks 5.1. Methodological issues There are several methodological issues throughout the material analysed, where three problems are reoccurring: firstly, dismissal of sexuality in studies on gender-bias in textbooks; secondly, a global generalisation of geographically specific research; and thirdly, how the researchers have revealed or assigned a character’s gender. The first obstacle is something I have shed light upon throughout this review, where it has been evident that one cannot look at sexuality without also looking at gender (e.g. De Vincenti et al., 2007; Røthing & Svendsen, 2011; Temple, 2005), but when gender is the sole thing analysed, heteronormative reasoning protrudes from the research since the heterosexual domestic roles mentioned (in e.g. Lee 2014; Lee & Collins, 2009; Ullah & Skelton, 2013) are not problematised in reference to sexual identity but simply to the woman’s role in the relationship. The second obstacle is the issue of generalising geographically specific studies. Since different nations have different norms, starting points and legal restrictions, it seems difficult to apply findings or analyses of findings to a global scale. What is seen as a gender norm in one culture, may not be a gender norm in another culture. To analyse heteronormativity in one region’s textbooks might therefore require a researcher familiar with said region’s norms. To make global generalisations of heteronormativity in textbooks, there is also a need for a wider spread of research; many recent studies have been conducted in Middle Eastern countries (e.g. Bahman & Rahimi, 2010; Hall, 2014; Ullah & Skelton, 2013), whilst research from e.g. Sweden is missing apart from student theses (e.g. Holmqvist & Gjörup, 2007; Mustedanagic, 2010). The third obstacle is a question I have asked myself when reading a lot of the articles: How do the researchers determine the gender of a fictive character? To see gender-bias in textbooks, you have to establish which characters are men and which are women. If characters do not exclaim ‘I am a woman’, or if they are not referred to in terms of him, she, boy, woman, etc., how is their gender possibly determined? There are instances in the research when some characters are seen as gender neutral or referred to as having an ‘unknown gender’ (e.g. Holmqvist & Gjörup, 2007; Mustedanagic, 2010), but these instances are far from many, and none of the reviewed material mentions how they actually determine the gender of the ? 15 characters. It seems that one way that gender is determined is through the characters’ names, for instance when determining the gender of dialogue participants, where a character’s name is often stated next to its utterance. Jones et al (1997), who looked at bias in dialogues, do, however, note that if a person’s gender “was unclear from the text or from visual clues” (p. 485) they did not include these instances when discussing the bias and that they did not assume that “an unnamed ‘mechanic’ would be male – unless there was a picture of a man” (p. 486). They did not assume a mechanic was male because, similar to the introductory riddle about the surgeon, the word mechanic is not gender-specific even if it is seen as masculine. Still, the authors do not clarify if they would categorise names as male or female. Also, Bahman and Rahimi (2010) give examples of males as main characters saying that “a toy dog was given a males [sic] name and called Joey” (p. 275). Furthermore, when looking at occurrence and visibility, the reviewed material has considered names to be one of the instances showing maleness or femaleness of characters. Since names are not necessarily, but perhaps traditionally, representative of only one gender, it seems nonchalant to base research on them. I have a friend who was assigned the female sex at birth, she identifies as a woman, and is identified by others as a woman, she is unquestionably a woman; even so, her name is considered a male name: Nicholaus. This personal experience may not be generalisable as of yet, but still shows how gendered names could misguide the interpreter. One could argue that when the researchers assume a character’s gender based on their name, so will teachers and students. Yet, when researchers do not contemplate different interpretations of a name, they assign themselves an interpretative prerogative. Additionally, I occasionally find it prejudiced to assume gender based on visual clues; however, most of the researchers mentioning illustrations analyse their gender-bias, there is therefore a purpose in assuming the illustrated character’s gender. When the names are seen to demonstrate a character’s gender, the names themselves are not analysed. It might thus be of importance to either ignore names when looking at gender-bias, or mention the ambiguity of the analysis and results. ? 5.2. Sexuality is a gender norm In this review, I have looked at research on gender and sexuality in textbooks. What has become clear is how no researcher studying sexuality can avoid talking about gender, whilst ? 16 almost no researcher studying gender-bias considers sexuality as an aspect of their analysis. I argue that sexuality, or rather heterosexuality, is closely connected to gender norms. In a (heteronormative) society, when one is seen as a woman one is expected to be interested in men and vice versa. The current understanding of sexuality would not exist if genders, gender norms, or ‘opposite genders’, did not exist. By analysing gender in textbooks in terms of female/male characters, we adhere to a binary system of gender which reinforces heteronormativity and thereby also heterosexuality. Butler (1990/1999) argues that the “institution of a compulsory and naturalised heterosexuality requires and regulates gender as a binary relation in which the masculine term is differentiated from a feminine term, and this differentiation is accomplished through the practices of heterosexual desire” (p. 30). I.e. ‘institutionalised heterosexuality’ requires a binary gender system as it demands opposing genders; and, by reinforcing heterosexuality, gender norms are reinforced. Or, simply, as Cameron and Kulick (2003) phrase it: “Whenever sexuality is at issue, gender is also at issue - and, importantly, vice versa” (p. 142). It is also visible from the research reviewed that both gender-bias and heterosexism most commonly express themselves through invisibility, i.e. the lack of non-male and non- heterosexual characters in textbooks. Perhaps then the optimal way of portraying characters would be as gender neutral, not gender balanced, as it would allow for every learner to identify with every character and exclude issues of invisibility and heteronormativity. However, gender neutrality would most probably lead to the use of male-generics, which again leads to female invisibility. Nevertheless, researchers should problematise, if not conceptualise, these intertwined norms when analysing (hetero)sexism in textbooks to illuminate heteronormativity. ? 5.3. Future research In conclusion, the research on gender and sexuality in EFL textbooks is informative and generally reliable. The research, as a whole, indicates a development in awareness and inclusivity amongst textbook writers. Nonetheless, further research which combines gender and sexuality is required. Future research on gender-bias and heterosexism in EFL textbooks is needed to shed light on the heteronormativity present there, so that authors and publishers of materials can try to avoid it, and so that teachers dare to discuss it. Also, a lack of research ? 17 “in this area contributes to the maintenance of the status quo of heteronorm- ativity” (Pawelczyk et al., 2014, p. 61), as the learners would then redefine themselves in terms of a heteronormative view of society portrayed in the preserved textbooks. Some suggestions for future research are: (a) First and foremost, to research textbooks with heteronormativity in mind. Research where sexuality is seen as a gender norm, so that the gap between gender norms and sexuality diminishes. (b) To study if, and how, teachers approach the issues of heteronormativity they encounter in textbooks. Thereby, looking at the connection between sexual identities and binary gender in e.g. class discussions or ‘teacher talk around the text’. (c) To look at students’ reactions to heteronormative texts in contrast to their reactions to inclusive texts. How do students understand a text differently, and what effects are there on their language learning and understanding of identities, when the text is either heteronormative or inclusive? (d) To research the expression of other sexual identities than what belongs to the heterosexual/homosexual binary in textbooks. Are there other sexualities present in the material? Of course, this would also include looking at gender in some respect. (e) To analyse which culture(s) textbooks portray. Do they represent the society in which the textbook is used, the culture(s) where the language is spoken as a first language, or a desired society? ? ? 18 Reference list ? Bahman, M., & Rahimi, A. (2010). Gender representation in EFL materials: An analysis of English textbooks of Iranian high schools. Procedia - Social and Behavioral Sciences, 9, 273–277. doi:10.1016/j.sbspro.2010.12.149 Baker, P. (2008). Sexed texts: Language, gender and sexuality. Oakville, CT: Equinox Pub. Barton, A., & Sakwa, L. N. (2012). The representation of gender in English textbooks in Uganda. Pedagogy, Culture & Society, 20(2), 173–190. doi: 10.1080/14681366.2012.669394 Butler, J. (1999). Gender trouble feminism and the subversion of identity. New York: Routledge. (Original work published 1990) Cameron, D., & Kulick, D. (2003). Language and sexuality. Cambridge: Cambridge University Press. De Vincenti, G., Giovanangeli, A., & Ward, R. (2007). The queer stopover: How queer travels in the language classroom. Electronic Journal of Foreign Language Teaching, 4(1), 58–72. Dumas, J. (2008). The ESL classroom and the queerly shifting sands of learner identity. TESL Canada Journal, 26(1), 1–10. Gray, J. (2013). LGBT Invisibility and heteronormativity in ELT materials. In J. Gray (Ed.), Critical Perspectives on Language Teaching Materials (pp. 40–62). Palgrave Macmillan. Hall, M. (2014). Gender representation in current EFL textbooks in Iranian secondary schools. Journal of Language Teaching & Research, 5(2), 253–261. doi:10.4304/jltr. 5.2.253-261 Holmqvist, P., & Gjörup, L. (2007). The representation of gender and gender roles in English textbooks (Lärarutbildningen: Examensarbete). Malmö Högskola. Retrieved 150210 from http://dspace.mah.se/bitstream/handle/2043/3426/representation%20of %20gender.pdf?sequence=1&isAllowed=y ? 19 Jones, M. A., Kitetu, C., & Sunderland, J. (1997). Discourse roles, gender and language textbook dialogues: Who learns what from John and Sally? Gender and Education, 9(4), 469–490. doi:10.1080/09540259721204 Kehily, M. J. (2001). Issues of gender and sexuality in schools. In B. Francis & C. Skelton (Eds.), Investigating gender: Contemporary perspectives in education (pp. 116–125). Buckingham: Open University Press. Lee, J. F. K. (2014). A hidden curriculum in Japanese EFL textbooks: Gender representation. Linguistics & Education, 27, 39–53. doi:10.1016/j.linged.2014.07.002 Lee, J. F. K., & Collins, P. (2009). Australian English-language textbooks: The gender issues. Gender and Education, 21(4), 353–370. doi:10.1080/09540250802392257 Litosseliti, L. (2006). Gender and language: theory and practice. London: Hodder Arnold. Mills, S. (2008). Language and sexism. Cambridge  ; New York: Cambridge University Press. Mustapha, A. S. (2013). Gender and language education research: A review. Journal of Language Teaching & Research, 4(3), 454–463. doi:10.4304/jltr.4.3.454-463 Mustedanagic, A. (2010). Gender in English language and EFL-textbooks (Lärarutbildningen: Term paper). Halmstad University. Retrieved 150212 from http://www.diva- portal.org/smash/record.jsf?pid=diva2:346993 Nelson, C. (1999). Sexual identities in ESL: Queer theory and classroom inquiry. TESOL Quarterly, 33(3), 371–391. Nelson, C. D. (2008). Sexual identities in English language education: Classroom conversations (1st ed.). Hoboken: Taylor and Francis. Pavlenko, A. (2004). Gender and sexuality in foreign language education: Critical and feminist approaches. In B. Norton & K. Toohey (Eds.), Critical pedagogies and language learning (pp. 53–71). Cambridge: Cambridge University Press. Pawelczyk, J., Pakuła, Ł., & Sunderland, J. (2014). Issues of power in relation to gender and sexuality in the EFL classroom - An overview. Journal of Gender and Power, 1(1), 49–66. Porreca, K. L. (1984). Sexism in current ESL textbooks. TESOL Quarterly, 18(4), 705–724. ? 20 Røthing, Å., & Svendsen, S. H. B. (2011). Sexuality in Norwegian textbooks: Constructing and controlling ethnic borders? Ethnic and Racial Studies, 34(11), 1953–1973. doi: 10.1080/01419870.2011.560275 Shardakova, M., & Pavlenko, A. (2004). Identity options in Russian textbooks. Journal of Language, Identity, and Education, 3(1), 25–46. Skolverket. (2006a). I enlighet med skolans värdegrund? (No. 285). Retrieved 150206 from http://www.skolverket.se/publikationer?id=1659 Skolverket. (2006b). Läromedlens roll i undervisningen: Grundskollärares val, användning och bedömning av läromedel i bild, engelska och samhällskunskap (No. 284). Retrieved 150206 from http://www.skolverket.se/publikationer?id=1640 Sunderland, J. (1992). Gender in the EFL classroom. ELT Journal, 46(1), 81–91. doi:10.1093/ elt/46.1.81 Sunderland, J. (2000a). Issues of language and gender in second and foreign language education. Language Teaching, 33(04), 203–223. doi:10.1017/S0261444800015688 Sunderland, J. (2000b). New understandings of gender and language classroom research: texts, teacher talk and student talk. Language Teaching Research, 4(2), 149–173. doi: 10.1177/136216880000400204 Sunderland, J., Cowley, M., Abdul Rahim, F., Leontzakou, C., & Shattuck, J. (2000). From bias “in the text” to “teacher talk around the text.” Linguistics and Education, 11(3), 251–286. doi:10.1016/S0898-5898(00)00034-6 Temple, J. R. (2005). “People who are different from you”: Heterosexism in Quebec high school textbooks. Canadian Journal of Education, 28(3), 271–294. Ullah, H., & Skelton, C. (2013). Gender representation in the public sector schools textbooks of Pakistan. Educational Studies, 39(2), 183–194. doi: 10.1080/03055698.2012.702892 Yaghoubi-Notash, M., & Kooshavar, E. (2013). Adopted vs. adapted EFL textbooks in Iranian context: Focusing on gender representation and teacher beliefs. Journal of Language Teaching and Research, 4(4), 791–800. doi:10.4304/jltr.4.4.791-800 ? 21