Gender Voices in Hong Kong English Textbooks—Some Past and Current Practices

Abstract

This paper seeks to investigate whether recent improvements in the status of women in Hong Kong are reflected in patterns of gender representation in Hong Kong secondary English textbooks. A comparison of ten recently published books which are currently in use with ten published in the late 1980s/early 1990s and no longer in use revealed that women appeared more frequently in the former, and that greater use was made of gender-inclusive pronouns and the neutral address title Ms. Nevertheless some writers, it was found, continue to perpetuate the stereotyped image of women as weaker than men, and as operating primarily within domestic rather than social domains. The “male-first” phenomenon and the visual under-representation of women are still prevalent in recent textbooks.

Introduction

There is much evidence to suggest that schools serve as important agents in shaping children’s gender attitudes and behaviour (e.g., Jones et al. 1997; Sunderland et al. 2001), and furthermore that within schools textbooks play a significant role in the gender socialization of children (e.g., Britton and Lumpkin 1977). This paper examines changes in gender representation and the use of gender-neutral language in secondary English language textbooks published and used in Hong Kong over the past two decades. This period is one during which there has been, following the passing of the Sex Discrimination Ordinance in 1995, heightened awareness of sexist issues and practices in Hong Kong.

The Status of Women in Hong Kong

We have a very beautiful party leader, and the Democratic Party doesn’t.

(Cheng 2006)

Although this remark by Professor Yu-shek Cheng, the secretary-general of the newly formed Civic Party in Hong Kong, about the differences between the new party’s leader and the Democratic Party’s leader, was undoubtedly uttered in jest, it nevertheless reflects an attitude that a woman’s physical appearance is more important than her ability or talents. It is not difficult to find evidence that gender-stereotyping of this type is alive and well in Hong Kong. Witness the continuing popularity of such Chinese sayings as 醇酒美人 (fine wine and pretty women, which are regarded as every man’s wants), and 郎才女貌 (men’s talents and women’s beauty). Or witness the common belief in Hong Kong that families with only female and no male descendants are in danger of “dying out”. Hand in hand with such attitudes goes the traditionally unequal educational treatment of girls and boys in Hong Kong. Before 1978, when free and compulsory 9-year education was first introduced, young girls were more educationally disadvantaged than boys. According to the 2001 Population Census, 12% of Hong Kong women had never received schooling, while the corresponding figure for men was only 4.6% (Census and Statistics Department 2002). Consider finally the unequal treatment of women in the workplace: in the early 1960s women officers in the civil service were paid one quarter less than their male counterparts for doing the same job and pregnant women were dismissed by their employers (Chan 2000). The female–male earnings ratio in 1976 was .63 (Wong 1995, p. 61).

The status of women in Hong Kong has improved in the last two decades, with likely factors being the influence of the Western feminist movement, improvements in education, and economic development. In 2001 the school attendance rate of girls was higher than that of boys in the 12–18 age group and girls made up more than half the number of students (51.6%) in tertiary institutions, indicating that these days girls enjoy the same educational opportunities as boys. According to the Census and Statistics Department (2002), the labour force participation rates for females aged 25–34 rose from 66.8% in 1991 to 76.9% in 2001, from 55.5% to 61.1% for those aged 35–44, and from 48.4% to 53.7% for those aged 45–54. Likely explanations for these increases include improved educational opportunities for women and the increased likelihood of women remaining in the workforce after marriage.

In recent years the Hong Kong government has undertaken a number of initiatives based on gender equity principles. A Sex Discrimination Ordinance was passed in 1995 to make discrimination on the basis of sex unlawful, and to provide for the establishment of an Equal Opportunities Commission. The Commission has, since its establishment in 1996, worked towards the elimination of discrimination and the provision of equal opportunities for men and women. Since 2001, the Women’s Commission, a central mechanism to promote the well-being and interests of women in Hong Kong, has played a strategic role in advising the government on policy directions for women issues. Public education and promotional activities have been organized to raise public awareness of and sensitivity to gender-related issues, as well as to reduce gender stereotyping. Heightened public awareness is revealed in the guiding principles for quality textbooks recommended by the Curriculum Development Council Ad Hoc Committee on Textbook Quality (2003), with effect from 2001, which include the stipulation that: “The content and illustrations do not carry any form of discrimination on the grounds of gender, age, race, religion, culture and disability etc., nor do they suggest exclusion.” Nevertheless, there is as yet no formal code of practice in place for publishers to observe in the production of textbooks in Hong Kong. Different publishers merely interpret the stipulation in their own ways.

Previous Textbook Studies

The research effort devoted to textbook analysis indicates the importance of textbooks in the cognitive and behavioral development of young learners. Learners, who generally attach great credibility and authority to educational materials, tend to absorb and assimilate the materials in minute detail without comment, and to be susceptible to their influence. As previous researchers have observed, not only does gender portrayal in textbooks affect social values and behaviour (e.g., Briere and Lanktree 1983; Frasher and Walker 1972; Peterson and Lach 1990), but the gendered messages conveyed in textbooks undoubtedly have the potential to influence the development of students’ attitudes at an impressionable age. The insidious nature of gender bias in textbooks is noted by Britton and Lumpkin (1977, p. 41):

This subliminal repetitious implanting of bias influences their [children’s] lifelong aspirations; indeed it leaves a permanent distorted imprint upon our children’s future. It not only limits their choices in terms of life styles and career selections but it distorts their self-image and the images of the opposite sex

The likelihood of a negative impact on women’s learning when gender-biased language is used in teaching materials has also been noted (see, e.g., Crawford and English 1984). According to Macaulay and Brice (1997, p. 820), in their study of gender bias in example sentences in US syntax textbooks, “Seemingly trivial things like example sentences can contribute to a hostile environment…for women in the academic world”.

What are some of the ways in which writers convey sexist attitudes through the content and language of textbooks? In the following sections we review a number of problem areas.

Masculine Generic Constructions

A common manifestation of sexism in language is the “generic” use of man and man-compounds (e.g., man, fireman, salesman), and of masculine pronouns (e.g., him, he), when they refer to people in general or when the sex of the referent is not specified: If someone calls, ask him to wait; All men are selfish; Who is the spokesman of the organization? This practice has been objected to as reflecting an androcentric world-view, insofar as it can be unclear whether the forms include both men and women or whether they refer to men only and, as Briere and Lanktree (1983), Cole, Hill and Dayley (1983), Hamilton (1988), and others have shown, people rarely conceptualize women when masculine “generics” are used. A strategy adopted by writers to avoid masculine generic pronouns is the use of paired pronoun expressions such as he/she, he or she, him or her and his or her. This alternative, though common in formal genres and clearly gender-inclusive in form and meaning, has been criticized by some commentators (e.g., Fowler 1965; Jochnowitz 1982) on the grounds that it sounds cumbersome, especially in speech or when repeated. It also sounds unnatural when it is used in tag questions: Everybody is listening, isn’t he or she?

Another strategy for avoiding masculine generics is the use of they as an anaphoric pronoun with a singular, generic or indefinite antecedent. Generic they has been attacked by various commentators for its apparent violation of number agreement (e.g., Burchfield 1981; Fowler 1965; Partridge 1965), but is widely accepted by modern usage handbook writers and commentators (e.g., Balhorn 2004; Greenbaum 1996; Pauwels 2001a; Peters 2004).

Previous ESL textbook studies in the USA (e.g. Porreca 1984) and in Hong Kong (e.g., Equal Opportunities Commission 2000) have revealed that many writers attempt to avoid the masculine generic by using the paired pronouns. Selection of the masculine pronoun he, however, is regularly associated in English language textbooks in Hong Kong with such roles and professions as president, sportsperson, farmer and thief (Equal Opportunities Commission 2000).

Female and Male Social and Domestic Roles

Traditionally, men’s space tends to be public, in the workplace, and women’s space to be private, in the home. The prescriptive view is that while men should assume the role of breadwinner, women should be in charge of the family and be breadtakers. These culturally entrenched stereotypes are embodied in English proverbs: “A man’s home is his castle, and a woman’s place is in the home”, and Chinese sayings: 男主外, 女主內 “Men work outside and women stay at home”, and 入得廚房, 出得廳堂 “(a good wife) is a good cook in the kitchen and elegant enough to receive guests in the lounge”. Despite the fact that many women have entered the labor force these days, there is still a dichotomy between masculine and feminine roles, with such high-status occupations as lawyer, architect, astronaut, and professor being considered men’s jobs, and others such as teacher, nurse, and secretary as women’s jobs.

Previous textbook studies found widespread gender-stereotyping, with males occupying a wider range of social and occupational roles and women depicted mainly in domestic and nurturing domains. This was evident in Cincotta’s (1978) early investigation of a widely used textbook for the teaching of French, and Hartman and Judd’s (1978) survey of ESL textbooks (mostly American, with a few British publications). In a more recent picture analysis of Chinese language textbooks used in Hong Kong primary schools, Law and Chan (2004) obtained similar results: females tended to be more visible in family and household settings and were engaged in a markedly restricted range of occupations, while males were more dominant in public settings. Meanwhile, Evans and Davies (2000) examined two major reading textbook basal series in the USA, finding that women were presented as more affectionate, emotional and passive, whereas men were shown to be aggressive, argumentative, and competitive.

Omission

A number of studies have revealed a quantitative imbalance in the appearance of women and men, with the former being less frequently mentioned and thus implicitly presented as being of lesser importance. For example, Hellinger’s (1980) investigation of sexism in English language textbooks used in German schools found that over 93% of the texts had male participants, but less than 30% had female. An American study of ESL textbooks by Porreca (1984) noted an asymmetry in the ratio of women to men (1:1.77) in their presentation in illustrations.

Titles

The titles Mrs and Miss have traditionally served to differentiate women in terms of their marital status, contrasting in this respect with the neutrality of the title for men, Mr. The attempt by US feminists in the late sixties to redress this asymmetrical situation by introducing the title Ms as a replacement for Mrs and Miss has had to confront some daunting obstacles. In a study by Lillian (1995) many university students were found to associate Ms with women who were divorced, feminist, lesbian, or older. Atkins-Sayre (2005), however, traces out the history of the successful introduction of the title Ms and reports that Ms has begun to shed its negative connotations. Holmes (2001) and Pauwels (2001b, 2003) find evidence of an increasing use of Ms among women, especially those who are younger and educated to tertiary level, in Australia, Canada, New Zealand and the USA.

Order of Mention

The conventionalized practice of putting male names first in paired expressions (e.g., boys and girls, Mr and Mrs Chan) reflects a widespread perception of male supremacy. As early as the sixteenth century, the practice was being defended with observations of the type: “let us kepe a natural order and set the man before the woman for manners Sake” (Wilson 1560, p. 189; cited in Eckert and McConnell-Ginet 2003, p. 34). In Porreca’s (1984) study of ESL textbooks the average ratio of female to male firstness was 1: 2.96.

Despite the potential influence of textbooks, some contemporary researchers argue that a focus on textbooks alone is not sufficient since the manner in which a text is interpreted depends, to some extent, on the readers’ responses to the text. Apple (1990, p. 30) stresses that students’ interpretations of texts are not totally predictable:

We simply cannot fully understand the power of the text, what it does ideologically and politically (or educationally, for that matter) unless we take very seriously the way students actually read them—not only as individuals but as members of social groups with their own particular cultures and histories.

More recent studies have investigated teachers’ treatment of and learners’ responses to gendered materials (e.g., Mannheim, 1994; Sunderland et al. 1997, 2001). Sunderland et al. (1997) argue that since reader response is not observable and is difficult to investigate, attention should focus on the ways in which teachers may ignore, extend or subvert traditional gender roles in their treatment of texts.

The Present Study

Earlier studies of the depiction of gender roles in Hong Kong textbooks include Yau-Lai and Luk’s (1988) study of Chinese history and social studies in junior secondary textbooks, Au’s (1993) study of social studies, Chinese language, and health education in primary textbooks, and the Equal Opportunities Commission’s (2000) report on the nature and extent of stereotyping in printed educational materials, including textbooks and examination papers. All these studies were synchronic, revealing gender bias and stereotyping presented in the educational materials used at the time of the study. We consider it important to conduct diachronic research on gender representation in Hong Kong textbooks in the new century in order to determine whether changes in the nature of gender representation in Hong Kong secondary English textbooks have kept pace with changes in the status of women. The introduction of the Sex Discrimination Ordinance in 1995 and the establishment of the Equal Opportunities Commission in 1996 were considered to be significant landmarks for the purposes of the present study, which involved a comparison of textbooks published in the decades before and after 1997. We chose to analyze English language textbooks in this study because they constitute an important part of the curriculum in Hong Kong, where English is a compulsory subject in secondary schools and students have to spend about 4–5 h on English lessons every week. Many of the textbooks in Hong Kong involve patterned structures and mechanical drills which, if based on gender-biased material, may well contribute to the development of sexist attitudes at a subconscious level. Following the most frequent features discussed in previous research, this study attempted to seek answers to the following questions:

• 1. What is the ratio of female to male characters?

• 2. To what extent are women/men portrayed in social settings?

• 3. To what extent are men/women depicted in domestic roles?

• 4. What is the visual representation of men and women?

• 5. What generic pronouns are used?

• 6. How are women addressed?

• 7. What is the frequency with which women precede men when both are referred to in a single phrase, and vice versa?

These research questions led to the following hypotheses:

• 1. The ratio of female to male characters will be higher in recent textbooks.

• 2. Women will be portrayed more often in social settings in recent textbooks.

• 3. Men will be depicted more often in domestic settings in recent textbooks.

• 4. Women will have more visual representation in recent textbooks.

• 5. More gender-neutral generic pronouns will be used by recent textbook writers.

• 6. Women will be more likely to be addressed by the title “Ms” in recent textbooks.

• 7. There will be differences between earlier and recent textbooks in the order of mention of men and women when they are referred to in a single phrase.

Method

Twenty English language textbooks were selected from the lists of textbooks recommended for use by the Education and Manpower Bureau (which was renamed the “Education Bureau” in July 2007) and the former Education Department in Hong Kong (see the Appendix), their contents having been deemed acceptable in terms of coverage, content, sequence, exercises, language, illustration and format. All the books were broadly representative of the English textbooks published and used in most local Hong Kong secondary schools in the 1980s–2000s. Ten of the books, referred to as “recent textbooks”, were published after 1997, following the introduction of the Sex Discrimination Ordinance in 1995 and the establishment of the Equal Opportunities Commission in 1996, and all were in use at the time of the study. The other ten, referred to as “earlier textbooks”, were published in the late 1980s or early 1990s, before the Sex Discrimination Ordinance was passed. Half of the books were for senior forms and the other half for junior forms. Three chapters from each book, including all the written texts and pictures, were randomly selected for detailed content and linguistic analyses.

A systematic recording and tabulation was made of the characters and mentions of men and women in each chapter selected. The researchers counted: (1) male and female characters (the character being checked once, regardless of the number of times he/she appeared in the chapter), (2) male and female mentions, taking into account every time the character appeared in the chapter, (3) female and male social roles (e.g., tourist, teacher, athlete), (4) female and male domestic roles (e.g., brother, sister, mother, father), (5) occurrences of generic they, generic he, generic she, and paired pronouns he/she, (6) address titles Miss, Mrs and Ms (the title being checked once for each character, regardless of the number of times mentioned), and (7) order of mention of female and male characters when mentioned in a single phrase (e.g., Jane and Frank vs Frank and Jane). The assignment of a social role to a character did not preclude assignment of a domestic role as well. Hence an individual who was both a father and engineer was entered in both the “domestic role” and “social role” categories.

Illustrations were analyzed in this study because there appear to be a large number of pictures in Hong Kong English language textbooks that are designed to enhance students’ understanding and learning interests. The focus of the picture investigation was on: (1) the number and proportion of men and women in the pictures, and (2) the kind of activity undertaken by the main character in each picture. For the sake of simplicity the pictures which had shared main characters were excluded in the activity analysis.

At the beginning of the study one investigator completed the analysis of two textbooks and the results were cross-checked by the other investigator. Discrepancies detected were resolved by reviewing the items jointly. A research assistant was then trained to conduct the quantitative analysis. To ensure reliability, a random 10% sample of all the textbooks was also coded by the investigators. Overall, the second coder agreed with the first coder over 90% of the time, indicating that the coding system was reliable. The inter-coder reliability for each category was: (1) male and female characters: 90%; (2) male and female mentions: 84%; (3) female and male social roles: 87%; (4) female and male domestic roles: 90%; (5) occurrences of generic pronouns: 84%; (6) address titles: 96%; (7) order of mention: 93%; (8) number and proportion of men and women in pictures: 94%; (9) the kind of activity undertaken by the main character: 91%. Chi-square analysis (χ ²) was applied to the results, with a level of at least five per cent (p ≤ .05) being considered significant, and a degree of freedom of one. One limitation of the present study is that as copyright clearances were not obtained to store the textbooks in an electronic form, electronic corpus searches for the use of particular occupational terms (e.g., police officer vs policeman) were not feasible. Frequency counts of gender-neutral and gender-biased occupational terms were not within the scope of the present study.

Results and Discussion

Female and Male Characters

According to census and by-census statistics, the number of men in Hong Kong consistently surpassed that of women prior to 1997. Since 1997, however, the demographic pattern has changed: women now outnumber men (by a ratio of 1,000:911 in 2006, according to the Census and Statistics Department, 2007, p. 27). Is this demographic change reflected in recent textbooks? To determine the answer to this question we counted the number of male and female characters, and the number of their mentions in the texts. The findings are presented in Table 1.

Table 1 Characters and frequency.

A comparison of the earlier and the recent textbooks indicates a significant reduction in the numerical dominance of male characters, both in terms of character types (χ ² = 9.927, p < .005) and frequency (χ ² = 175.002, p < .001). The ratio for the total number of female characters to male characters is 1:1.48 for the earlier textbooks and 1:1.14 for the recent textbooks. A comparison of the average ratio of female to male mentions for the earlier and recent books indicates a more even spread nowadays, with the ratio for the former being 1:1.69, and for the latter being 1:0.96. Four of the ten recent textbooks examined had a higher number of female mentions than male. This confirms the first research hypothesis that the ratio of female to male characters would be higher in recent textbooks. At first blush, this might suggest that some textbook writers have become more aware of sexist issues. However, when the content of each unit was closely examined, it was found that the ratio of men to women was not evenly distributed in the texts. For example, in Unit 7 of R1, which is concerned with crimes, there are 104 male tokens but only 38 female. While the men act as robbers, police officers and Junior Police Call members who help catch the robbers, women play minor roles in fighting crimes (e.g., calling for help). By contrast, in Unit 9 of the same book, whose theme is friendship, a total of 134 female occurrences and 43 male occurrences were recorded. Likewise, R2, which has 70.7% of female mentions, has two units which are dominated by women, and the topics are about friendship (224 women vs 113 men) and appearance (132 women vs 18 men). Similarly, R7 has 33 mentions of men but 143 occurrences of women in the unit “I love Hong Kong”, which is about taking friends around for sightseeing and shopping. This suggests that gender stereotyping is still deeply rooted in some textbook writers’ minds: it is men who assume the more active roles involved in committing and fighting crimes, while women are allocated quieter, more experiential roles to play, and they are more concerned about friendship and their appearance.

Female and Male Social and Domestic Roles

We also examined the different roles, both social and domestic, served by women and men in the written texts. Following Law and Chan (2004), we divided the social roles into five major categories (see Table 2): male-monopolized, male-dominated, female-monopolized, female-dominated, and gender-shared. If the type of social role was served by men only, it was classified under “male-monopolized”. An example found in recent textbooks is the role criminal: there were 20 tokens for male and none for female. Likewise, if a social role was served by women only, it was categorized under “female-monopolized”. An example found in recent textbooks is receptionist: two tokens for female and none for male were found. Female-dominated roles refer to those which were largely taken up by women rather than men, while male-dominated roles are those in which men were portrayed more often. Gender-shared roles refer to those roles performed by men and women equally. An example in recent textbooks is the role employer: one token each was found for men and women.

Table 2 Social roles.

Table 2 indicates that men occupy a wider range of social roles in the male-monopolized category in both the earlier and the recent textbooks: 30 and 36 respectively. The corresponding figures for the female-monopolized category are 16 and 8. Similarly, there are many social roles which are dominated by men in both the earlier and recent textbooks. The figures are 16 and 17 respectively. Meanwhile, the corresponding figures for female-dominated social roles are only 5 and 12. A comparison of the male categories (both male-monopolized and male-dominated) with the corresponding female categories (both female-monopolized and female-dominated) found no significant differences between the earlier textbooks and the recent books (χ ² = .263, p > .05). The second research hypothesis, that women would be portrayed more often in social settings in recent textbooks, was therefore rejected. These findings suggest that women have consistently played a more limited range of social roles than men in Hong Kong English language textbooks over the past two decades. A further examination suggests a perpetuation of the traditional stereotypes associated with women and men. For example, women occupy such positions as secretary, receptionist, fashion designer, counselor, maid, and typist, although there are occasional portrayals of women as doctor, dentist, police officer, and manager. Likewise, men tend to occupy traditionally “male” roles, ranging from the lower ranking roles of criminal, driver, farmer, construction worker, security guard and postman, to the higher ranking roles of scientist, inventor and film director. In both kinds of textbooks men tend to be more involved in physically-demanding jobs and activities, including member of disciplinary forces and soccer player.

Women are often portrayed as housewives engaged in domestic duties in both earlier and recent textbooks. Witness the contrasting men and women’s roles in the following exercise on articles in the earlier textbook E1 (p. 7):

• 1. Mr Lam is _____ assistant manager.

• 2. Mr Brown works for _____ import firm.

• 3. Mrs Lee is _____ housewife. She spends ____ hour _____ day in the health club to keep fit.

Both the earlier and recent textbooks thus reinforce the stereotyped view that, while both women and men are depicted as playing such traditional roles as mother and father, wife and husband, sister and brother, and daughter and son (see Table 3), it is only women who serve as homemakers, engaging in domestic chores such as cleaning and cooking. The third research hypothesis, that men would be depicted more often in domestic settings in recent textbooks, was thus rejected.

Table 3 Domestic roles.

Visual Representation of Men and Women

The unbalanced portrayal of males and females in the textbooks extends, as Table 4 shows, to their pictorial representation. Of the 533 pictures depicting humans in the earlier textbooks, the number representing male characters only was 232 (43.5%), while that for female characters only was 149 (28%). The number of pictures with more male characters (45 or 8.4%) also outnumbered that with more female characters (18 or 3.4%). Likewise, in the recent textbook sample the number of male-only pictures was 389 (43.3%) and that of female-only pictures was 247 (27.5%). The number of pictures with more male characters (63 or 7%) also outnumbered that with more female characters (28 or 3.1%). A comparison of the pictures showing males only and more male characters with those showing females only and more female characters recorded no significant differences between the earlier and recent textbooks (χ ² = .005, p > .05). The fourth research hypothesis, that female visual representation would be higher in recent textbooks, was rejected. The conclusion is inescapable that women have been under-represented in visual images in the secondary English language textbooks in Hong Kong over the last 20 years.

Table 4 Visual representation (pictorial representation of men and women).

Table 5 shows the kinds of activities in which the main characters engaged (shared main characters were excluded). The pictures reinforce traditional gender stereotypes: there are more women than men engaging in various kinds of household chores (11 vs 4 for earlier books, and 7 vs 2 for recent books). No significant differences between the two sets of textbooks were found (p > .05). As for the findings for written texts, the third research hypothesis, that men would be depicted more often in domestic settings, was rejected. Nevertheless, some departures from the traditional gender stereotypes were noted. In the earlier texts, two of the four pictures showing men doing housework reflect gender stereotypes—one shows a man making a mess while cooking, and the other a man washing his car. In the more recent texts, one picture shows a man cooking and thinking of asking his son to help him wash the dishes.

Table 5 Visual representation (activities shown in pictures).

On the other hand, there are more male than female characters performing non-household activities, such as engaging in play and work, with the males being portrayed as more active and sporty than the females. The number of pictures depicting male characters at play is about twice that for females in both the earlier and recent books. As for written texts, no significant differences between the two sets of books were found (χ ² = .946, p > .05), disconfirming the second research hypothesis that women would be portrayed more often in social settings in recent textbooks.

As noted above, the female labour force in Hong Kong has grown sharply since the early 1990s. Appropriately, then, the ratio of pictures showing women at work was higher in the recent textbooks than in the earlier ones. The average ratios of women at work to men at work were 1:2.88 for the earlier books and 1:1.91 for the recent books. Nevertheless, the differences found were not significant (χ ² = 1.920, p > .05), indicating that the workplace depicted in both the earlier and more recent textbooks is one that is dominated—as in reality in Hong Kong—by men.

How representative is the distribution of gender roles in the workplace as depicted in the textbooks? The findings of the 2001 Population Census suggest that the depiction is quite realistic. In 2001 in Hong Kong a significantly higher proportion of women were working as clerks and as workers in elementary occupations than men, who were more likely to be found in such occupations as managers, administrators, craft and related workers, and plant and machine operators and assemblers. The census statistics also indicate that in the period 1991–2001 a much higher proportion of economically inactive persons was recorded in the prime working age bracket 25–54 among women than men mainly because of the predominance of homemakers at these ages among women (about 37%) over men (0.9%). What these findings suggest is that it would be unrealistic to expect writers to distribute male and female characters equally across the range of activities and occupations presented, in view of the apparent domain-specificity of men’s and women’s lives in Hong Kong, with many men and women continuing to have quite different lifestyles. Whether it is the task of textbook writers to reflect reality by representing the sexes in ratios as they appear in the real world or whether they should fight gender stereotyping, which would often entail not representing the sexes in the same ratios as they appear in the real world, is an issue worth discussing.

Generic Pronouns

The significant differences found between the two sets of textbooks confirm the fifth research hypothesis that more gender-neutral pronouns would be used by textbook writers. As Table 6 shows, masculine generic pronouns are far less frequent in the recent textbooks (29 tokens) than in the earlier textbooks (53 tokens). One strategy adopted by writers to replace masculine pronouns involves the use of paired pronoun expressions such as he/she, s/he, and his or her. A total of 154 instances of paired pronouns were found in the recent textbooks, as opposed to only 24 tokens in the earlier texts (χ ² = 70.465, p < .001). The following are some examples illustrating their use:

• 4. Ask your partner if s/he has ever done the following things (E3, p. 74)

• 5. My best friend’s name is…He/she is…(R1, p. 87)

• 6. A computer “hacker” is a person who spends his or her free time…(R6, p. 8)

Table 6 Generic pronouns.

Another strategy for avoiding masculine generics is the use of they. The recent textbooks yielded a total of 41 instances of generic they, as against just four in the earlier textbooks. The use of generic they significantly outstripped that of the masculine pronoun (χ ² = 36.498, p <. 001). Some examples of generic they from the textbooks follow:

• 7. When we persuade, we give reasons and arguments to someone in an attempt to change their mind about a decision they have made (E6, p. 157)

• 8. We would like everybody to bring their own food and drinks (R1, p. 32)

• 9. No one has found my CD, have they? (R3, p. 96)

The authors of one textbook, R3, openly endorse generic they, instructing students to use it in question tags with an indefinite noun as the antecedent (p. 96):

When the subject in the main clause is the pronoun anyone/anybody, everybody/everyone, no one/nobody, somebody/someone, none or neither, we use the pronoun they in the question tag.

Another pronoun strategy for avoiding gender bias in recent textbooks is the use of she as a generic alternately throughout a text with he. A textbook example follows:

Your friend failed a big exam. You…

a. try to cheer her up

b. feel bad that you did better than her

Your best friend wants you to write his English assignment. You …

a. tell him that you won’t do it all but you’ll help him

b. …(R2, p. 49)

There was only one instance of generic she in the earlier textbooks, as against 26 in the recent textbooks. The increase in use of generic she as compared with the decline in generic he was found to be significant (χ ² = 30.166, p < .001).

Titles

There are some notable differences between the recent and earlier books in the frequencies for Mrs, Miss and Ms. As Table 7 indicates, while the earlier texts have 37 characters being addressed by Mrs and 21 by Miss, the corresponding figures for the recent texts, 23 and 8, are considerably smaller. Only two tokens of Ms were found in the earlier textbooks (an unsurprising finding in view of the unpopularity of this title in Hong Kong in the late 1980s and early 1990s), but the number rises to eight in the recent textbooks. A comparison of the titles Mrs and Miss added together with the new title Ms recorded significant differences between the two sets of textbooks (p < .05), a finding confirming Research Hypothesis 6, that women would be more likely to be addressed by the title Ms in recent textbooks.

Table 7 Titles.

Order of Mention

In the present analysis, given two nouns paired for sex, women were rarely mentioned first, except in the fixed phrase ladies and gentlemen, and in cases where the female was older and/or more senior (e.g., Maria and little Albert, a picture of Winnie and Paul, with Winnie being the elder sister). As indicated in Table 8, all the textbooks reviewed, whether earlier or recent, evidenced a much higher tendency for men to be mentioned first (e.g., Ben and Mary, Mr and Mrs So’s bedroom, successful sportsmen and sportswomen, brothers and sisters). The male-first phenomenon was also found in form-filling contexts in the textbooks—for titles Mr always came before Mrs, Miss and Ms; for sex male always came before female. The average ratio of female- to male-firstness for the earlier books was 1:2.16, while that for the recent books was 1:16.55. The difference between early and recent texts was significant (χ ² = 58.281, p < .001). The significantly higher frequency of male-firstness in the recent books is largely attributable to the frequency of alternative pronoun expressions such as he or she, he/she, his or her, his/her, etc. The last research hypothesis, that there would be differences between earlier and recent textbooks in the order of mention of men and women when they are referred to in a single phrase, was confirmed.

Table 8 Order of mention.

Conclusion

A recent interview with English language publishers conducted by members of the Equal Opportunities Commission (2000) found that there was a high level of awareness of the importance of avoiding stereotyping in textbook materials, especially in relation to matters of gender representation. The present diachronic study suggests that such awareness is beginning to translate into practice. Firstly, the findings confirmed the first research hypothesis that the ratio of female to male characters would be higher in recent textbooks. The hypothesis that more gender-neutral generic pronouns would be used by recent textbook writers was also confirmed. Finally, the study confirmed the sixth hypothesis that women would be more likely to be addressed by the title “Ms” in recent textbooks.

Despite the fact that textbook writers are evidently making increased efforts to avoid biased or stereotypical treatment of males and females, there is still much room for improvement. There was no significant change, textual or visual, in the representation of women in social and domestic settings from the earlier to the recent textbooks. Women continued to be associated with a limited and stereotyped set of activities and careers, and with activities in which they play weaker, more passive roles than men. The disconfirmation of the fourth research hypothesis, that women would have more visual representation in recent textbooks, followed from the visual under-representation of women over the past 20 years. The confirmation of the last research hypothesis, that there would be differences between earlier and recent textbooks in the order of mention of men and women when they are referred in a single phrase, indicates that the male-first phenomenon and women’s secondary status have been prevalent over the last two decades, and that the problem is getting more serious these days with the frequent use of paired pronouns such as he or she, and him/her.

In our view, it would be desirable if a formal code of practice for editors in relation to issues of gender equity were to be developed for Hong Kong, with guidelines for the use of inclusive language and balanced treatment of the sexes. These might include recommendations for accurate descriptions of contemporary practices and equal visibility for female and male characters, associated with a wider range of occupational roles and personal traits. Furthermore, we believe, teacher education programs should provide pre-service and in-service teachers with strategies for handling the shortcomings and gender biases found in teaching materials. Teachers should learn to ask themselves how men and women are represented, and what types of issues this representation raises.

One topic that we have not touched on in this study, but one which may well be worth investigating as a complement to the study of gender bias in textbooks, is the concept of “teacher talk around the text” (as propounded by Sunderland et al. 1997, 2001). To the extent that teachers are catalysts—rather than merely passive entities—in the classroom, it would be of interest to know in what ways Hong Kong teachers handle the gendered content of textbooks, and to what extent in their teaching they “ignore”, “endorse” or “subvert” stereotypical ways of thinking.

Appendix

Earlier Textbooks

E1: Christie, H., Yuen, K. S., & Nancarrow, C. (1994). English: A modern course 1. Hong Kong: Aristo.

E2: Christie, H., & Yuen, K. S. (1994). English: A modern course 5. Hong Kong: Aristo.

E3: Methold, K., & Tadman, J. (1990). New integrated English 1. Hong Kong: Longman.

E4: Methold, K., Tadman, J., & Lam, J. (1990). New integrated English 5. Hong Kong: Longman.

E5: Howe, D. H., Kirkpatrick, T. A., & D. L. (1986). Access today 1. Hong Kong: Oxford University Press.

E6: Howe, D. H., Kirkpatrick, T. A., & D. L. (1986). Access today 5. Hong Kong: Oxford University Press.

E7: Mackay, R., & Byron, S. (1994). Impact 1. Hong Kong: Macmillan.

E8: Smallwood, I. M., & Walsh, S. (1994). Impact 5. Hong Kong: Macmillan.

E9: Etherton, A. R. B. (1986). Creative English for secondary schools 1. Hong Kong: Ling Kee.

E10: Etherton, A. R. B. (1988). Creative English for secondary schools 5. Hong Kong: Ling Kee.

Recent Textbooks

R1: Williams, A., & Dawson, C. (2004). New English treasure 1B. Hong Kong: Oxford University Press.

R2: Dixon, M., Kent, J. C., Norberg, M., & Williams, A. (2006). New progress 4A. Hong Kong: Oxford University Press.

R3: Kent, J. C., & Hodson, R. (2003). Progress 5. Hong Kong: Oxford University Press.

R4: Sampson, N. (2001). New way to English 2A. Hong Kong: Macmillan.

R5: Li, P. L., Leetch, P., & Burns, G. J. (2001). Real English 1B. Hong Kong: Macmillan.

R6: Sampson, N. (1997). English 2000, 5. Hong Kong: Macmillan.

R7: Nancarrow, C., Thomas, G., & Yuen, K. S. (2004). Living English 1A. Hong Kong: Aristo.

R8: Nancarrow, C., Hsing, B. M., & Yuen, K. S. (2004). Living English for the Certificate Exam 4A. Hong Kong: Aristo.

R9: Nelson, J. A., Chan, K., & Swan, A. (2004). Longman express 1B. Hong Kong: Longman.

R10: Vickers, E., Wheeler, J., & Lee, I. (2003). Longman express 5. Hong Kong: Longman.