2.1 Norms and Normativity in Language, Gender, and Sexuality Research

It has been pointed out that early research on language and gender tended to focus on normative language, with special emphasis on women’s language. As Bucholtz (1999: 9) observed, “Much of the scholarship in language and gender has been what might be called ‘good-girl research’: studies of ‘good’ (that is, normatively female – white, straight, middle-class) women being ‘good’ (that is, normatively feminine).” Likewise, discussing early language and gender research, Goodwin (1999: 388) noted, “researchers assumed white middle-class American women’s speech to be the norm.” Such an approach parallels that of early Japanese language and gender research, which focused on dichotomous men’s and women’s language that represented behavioural norms of men and women based on middle and upper-middle class speakers in Tokyo (see Okamoto and Shibamoto Smith, 2004, 2016; Inoue, 2006, among others). In other words, early research on language and gender was itself ideologically biased.

Recognising the partiality and limitations of such early language and gender research, a large number of studies, especially since the 1990s, began investigating the language use of socially and situationally diverse “men” and “women.” Importantly, these studies are based on empirical data, which helped reveal many nonnormative uses. For example, studies documented many cases in which socially diverse women do not always conform to the normatively feminine speech style, or the stereotypical idea of women’s language as cooperative, supportive, and nonconfrontational (e.g. Goodwin, 1999; Mendoza-Denton, 1999). Japanese women also have been found to use linguistic forms, such as specific sentence-final forms (SFF s) and personal reference terms, that are considered men’s language (e.g. Okamoto and Sato, 1992; Miyazaki, 2004; Sunaoshi, 2004). Furthermore, this nonessentialist approach also meant increasing attention to issues concerning sexuality and language, including language use in LGBTQ+ communities (e.g. Land and Kitzinger, 2005; Levón, 2012; Dawson, 2019; for Japanese, see Abe, 2010; Maree, 2020) and the discursive reproduction of heteronormativity to reinforce hegemony (e.g. Gill, 2014; Jones, 2016; Kiesling, 2002; Mills, 2008; Talbot, 2014; Zhu et al., 2025; for Japanese, see Nakamura, 2004, 2007, 2013, 2025; Shibamoto Smith, 2004).

While these nonessentialist studies as a whole have greatly advanced language, gender, and sexuality research, there remain certain outstanding issues. One such issue, the focus of this study, concerns the notion of a norm and its role in actual language practice. The constructionist approach has drawn on Butler’s (1990) work regarding gender as performative, and has tended to emphasise the agency and creativity of language use; much less attention has been placed on “the highly rigid regulatory frame” (Ehrlich and Meyerhoff, 2014; McElhinny, 2014) that is another important aspect of Butler’s (1990: 33) performative theory. Hall, Levon, and Milani (2019: 482–483), in their introduction to a Language in Society special issue on normativity, called attention to a recent critique of queer theory for taking an antiessentialist and anti(hetero)normative approach to gender, which ironically resorts to the concept of normativity, a concept not clearly defined (see also Wiegman and Wilson, 2015). As Hall et al. underscored, the papers in the special issue asserted “the need to avoid overly simplistic dichotomies between what is ‘normative’ and what is ‘antinormative,’ and instead to illustrate the various strategies individuals adopt to negotiate multiple, dynamic, overlapping, and at times, contradictory normativities (Abu-Lughod 1990; Gal 1995)” (485).

The diversity and multiplicity of normativities seem to be closely related to the intersectionality of social categories as well as variability across time and locale. Gender by definition intersects with a variety of social and situational variables, which may be what gives rise to diverse ideas about speech norms. Hall et al. (2019: 484) contended that “we cannot speak of a singular ‘normativity’ organizing social practice. Instead, we must consider the multiple and competing normativities that are available for individuals to draw upon.” Blommaert (2007: 2–3), in a discussion of transnational communities, further explicated the nature of normativity:

[H]uman social environments [need] to be seen as polycentric and stratified, where people continuously need to observe “norms” – orders of indexicality – that are attached to a multitude of centers of authority, local as well as translocal, momentary as well as lasting: the family, the peer group and the immediate neighborhood networks; religion, the media, transnational networks, the State (both home and host), the labor market, and abstract ideals and role models (e.g., gender and social status roles). Such orders of indexicality [are] stratified complexes in which distinctions exist between “better” and “worse” forms of communication, and in diasporic neighborhoods, such orders of indexicality combine and compete: one can be a “good” user of language in the neighborhood network, but a “bad” one in the labor market or in the host State’s school system.

The nature of norms thus understood requires us to examine the complexity of situated sociolinguistic practice, as illustrated by scores of empirical studies. Due to space limitations, I refer to only one such study here. In their study of gay Palestinians in Israel as represented in the documentary film Oriented, Milani and Levon (2019) found that the experience of these gay men cannot be accounted for in terms of a singular normativity, such as normative homonationalism, dominant among Israeli gay men. This is because their lived experience is simultaneously normative and nonnormative, as these men feel ambivalence between being “attached to Israel,” imagined as a homotopia, and “unmoored,” or “liberated and deeply constrained,” due to multiple normativities concerning gender and (homo)sexuality, nationality, and region, inter alia (625).

2.2 Styling and Multimodal and Multilevel Semiotic Resources

Another key theoretical issue for this study concerns the use of multilevel and multimodal semiotic resources for styling. This issue is closely related to the multiplicity of concepts of norms, discussed in the preceding section. As emphasised above, an individual’s sociolinguistic practice cannot be adequately understood in terms of binary distinctions, such as normative versus nonnormative, appropriate versus inappropriate, feminine versus masculine, or standard versus nonstandard. This is because every sociolinguistic interaction is likely to involve multiple normativities, and because any given use of a linguistic form may not clearly fit into an oppositional pattern (e.g. as either normative or nonnormative). The styling in which each speaker engages in each social context is thus a complex process of choosing among a variety of semiotic resources while negotiating the relevant multiple norms. The term “styling” is, therefore, more appropriate than “style” to refer to one’s sociolinguistic practice, as noted in Coupland’s (2007: 2) argument that “what matters for linguistic style is more to do with process than with product, more to do with use than with structure. Stylistic analysis is the analysis of how style resources are put to work creatively.”

But what are “style resources”? As discussed earlier, the “content” of speech norms, or normative speech styles, is usually quite vague or abstract. In the case of language and gender, normatively feminine and masculine styles are commonly characterised in terms of contrasting abstract characteristics pertaining to speakers’ qualities and stances, such as cooperative/conflictive, nonassertive/assertive, indirect/direct, submissive/dominant, polite/rough, and so forth (e.g. Eckert and McConnell-Ginet, 2013; Goodwin, 1999; Nakamura, 2007; Okamoto and Shibamoto Smith, 2008, 2016). Irvine (2001: 22) characterised styles as “part of a system of distinction, in which a style contrasts with other possible styles, and the social meaning signified by the style contrasts with other social meanings.” Such contrasting styles (e.g. feminine and masculine styles) are shaped by a set of style resources, or variants of variables, that are ideologically and hierarchically ordered, constituting part of the relevant “orders of indexicality” (Blommaert, 2007, 2010). That is, a particular linguistic form – for instance, one of the two pronunciations of -ing (i.e. [ŋ] and [n]) in English – may not straightforwardly express a particular social meaning, such as male/female or middle-class/working-class, but is considered to index such meanings by ideologically linking the pragmatic meanings, or stances and qualities (e.g. formal vs. informal), associated with the variant forms to particular social categories – a process termed “indirect indexing” (Ochs, 1992; Eckert, 2008).

Needless to say, numerous style resources are available to speakers, including multilevel and multimodal semiotic resources, the use of which is increasingly recognised as important to investigate for a better understanding of the process of styling (Goodwin, 1999; Maree, 2020; Starr, Wang, and Go, 2020; Okamoto, 2018; Okamoto and Morimoto, 2023). And those resources need to be used in a coordinated and coherent manner (e.g. Irvine, 2001: 22). In practice, however, this may not be a simple enterprise because, as discussed earlier, there are complexities arising from a variety of contextual contingencies, such as multiply relevant normativities, the nonbinary nature of a single normativity (e.g. femininity, politeness), and ideologically diverse views of a particular normativity, as illustrated in Section 4.

Examining metapragmatic comments on Japanese women’s speech, this study investigates one of the understudied issues in language and gender research, namely, native speakers’ diverse views of speech norms, which I believe will lead to better understanding of the highly ideological nature of speech norms, including the sexist expectations toward women’s language use.

4.1 Initial Post: Ideological Clash between a Teenage Girl and Her Mother

Example (1), the initial post of the thread, indicates that the writer, A, and her mother believe in different kinds of norms for girls’ speech. A’s mother criticises A for using expressions such as ~janē ka ‘~isn’t?’ and maji? ‘really?’ and tells A to use honorifics and not talk back. Thus, if A’s report is accurate, her mother appears to believe that girls/women should speak in a polite and submissive manner, indicating the influence of the hegemonic ideology of femininity.

A’s concept of speech norms, on the other hand, concerns age (youthfulness) and interpersonal relationship (intimacy). She regards expressions like ~janē ka and maji? as part of wakamono otoba ‘youth language’ and even though she admits that her language is kanari warui ‘pretty bad,’ she is bothered by her mother’s criticism because A thinks it is all right to use it among young people and among familiar people like family members – an ideology that prioritises age and relationship rather than gender. In other words, she seems to believe that one’s language use depends on the context, like the posters in Group 2. A’s post also indicates that she is upset about the sexism “hidden” in her mother’s (and society’s) expectations about women’s speech, which constrain women more than men.

4.2 Women Should Not Use Bad Language (Group 3 Responses)

Let us now consider the responses to A’s post, starting with Group 3 in Table 1, which constitutes 22% (12 posts) of the total responses. These responses agree with A’s mother that women should not use “rough” or “bad” language. Many responses in this group stress that women’s ways of speaking reveal their social backgrounds, such as class status, upbringing, refinement, and/or education – an ideology built on the intersection of gender and class. This is illustrated in Example (2).

(2) Language use is reflected in your face

If you keep using hidoi ‘terrible/bad’ and ranbōna ‘rough’ language, your face will look like that. If your language use is hin no aru ‘refined,’ your face also will look refined. (5 lines omitted) Let’s stop using men’s language and shift to women’s language. This is also important in getting a job. Women who insist on using men’s language do not have a good life. So, if you can rectify it, you should. (Hatsugen Komachi, June 10, 2023)

This writer’s comments convey the beliefs that there are distinct women’s and men’s languages, that women should use the former, that their speech is a sign of women’s class status, and that women’s use of women’s language serves as an asset, or linguistic capital (Bourdieu, 1999), which will help them obtain an attractive face and a good life.

A few respondents in this group admit gender discrimination in the norms for women’s speech. One of them, while acknowledging the importance of gender equality, concedes to A’s mother’s view, as follows.

(3) I tend to be sensitive to the issue of gender equality, but

When I brought up my daughter, I used the phrase “because you are a girl” a number of times. To be honest, I didn’t like it, but in Japan there still remains gender discrimination. (6 lines omitted) And if you cannot be taken seriously by the man you love because of your bad language, you won’t be able to change his mind no matter how much you scream about gender equality. (8 lines omitted). (Hatsugen Komachi, June 11, 2023)

The writer of (3) supports gender equality but succumbs to what she believes to be the real world in which men (still) expect women not to use bad language.

The writer of (4) is very critical of A’s language use, referring to it as sonna kotoba ‘such language’ and ranbōna kotoba ‘rough language.’

(4) Oh, my

~janē ka. Where did you learn such language? I’ve never heard or used it. Even the young people on TV are not using it, although I’ve heard it in dramas, like “Irun janē ka, detekoi, kora. ‘Aren’t you there? Come out, hey.’” I think it’s a pretty ranbōna ‘rough’ language. Is it ossan kotoba ‘older men’s language’? (Sorry, older men.) Or is it inaka kotoba ‘rural language’ (Sorry, rural people). (Hatsugen Komachi, June 11, 2023)

By wondering whether A’s language is ossan kotoba or inaka kotoba, this writer also suggests that girls should not use either one.

4.3 Regardless of Gender, One Should Not Use Bad Language (Group 1 Responses)

Group 1 in Table 1 constitutes the largest number of responses, at 43% (23 posts). The writers in this group express the belief that the issue is not relevant only to women, but to everyone, as clearly stated in Example (5).

(5) That’s right, isn’t it?

Your mother is wrong, right? It’s not because “you are a girl,” but danjo kankei naku ‘regardless of gender,’ it is kitanai ‘bad’ language use. Even if a man uses it, it displeases the addressee. (4 lines omitted) (Hatsugen Komachi, June 9, 2023)

Many writers who share this view also emphasise that using kitanai ‘bad’ language has to do with one’s social background, especially poor upbringing, as exemplified by (6).

(6) Oh, that’s not it

It’s not youth language. But isn’t your sodachi ‘upbringing’ bad? We have a teenage son, but he doesn’t use such language even to family members. We don’t use honorifics at home, but we don’t use kitanai ‘bad’ language. (8 lines omitted) (Hatsugen Komachi, June 10, 2023)

This writer explicitly asserts that the use of “bad” language is neither a matter of youth nor a matter of gender, but rather has to do with the user’s upbringing.

Several writers in this group also point out positive values of using good language, as seen in (7).

(7) Buki ‘Weapon’

If you seriously think you want to convey your thoughts to adults, you must speak calmly using tadashī ‘correct’ language. If your language is ranbō ‘rough’ as you wrote in your post, rigid (in thinking) adults will not listen to you even if it is a valid opinion. You should study polite and correct language seriously, thinking it is a weapon for asserting your opinions. (3 lines omitted) (Hatsugen Komachi, June 11, 2023)

This writer suggests that one should use tadashī ‘correct’ and teineina ‘polite’ language because it can be a buki ‘weapon’ that compels others to listen to one’s opinions. The ability to use correct and polite language is thus ideologically regarded as a form of linguistic capital (Bourdieu, 1999).

The writers of (8) and (9) also deem using bad language to be a sign of class status, as they relate it to poor education and vulgarity. In addition, the writer of (8) comments on youth language, and that of (9) on the issue of gender inequality.

(8) May language be beautiful

A young person like you may think that rough language is somewhat trendy and cool, but yononaka no daitasū ‘most people in society’ do not think so. They regard such ranbōna ‘rough’ language use as gehin ‘vulgar’ and [demonstrating] chiteki reberu ga hikui ‘a low level of intelligence.’ (Hatsugen Komachi, June 11, 2023)

(9) It’s a life-long problem

When told to correct your language use because you are a girl, you want to rebut, saying, “If it’s a boy, is it OK?” It’s kind of understandable. But if you use that kind of language, you are of course, seen as kyōyō ga naku ‘uneducated’ and gehin ‘vulgar’ not as a girl but as a (decent) person. (7 lines omitted) (Hatsugen Komachi, June 10, 2023)

The writer of (8) argues that although young people may find rough language to have positive meanings, such as being trendy and cool, the majority of people disagree, regarding its users as vulgar and unintelligent, regardless of gender. By resorting to what “most people in society” think, the writer indicates that it is common sense – something that everyone should know. Like the writer of Example (3), the writer of (9) first concedes that it is understandable that A is upset about the sexism underlying women’s speech norms. Yet, she maintains that it is important to not use bad language if A does not wish to be seen as an “uneducated” and “vulgar” woman.

Three posts in Group 1 also referred to regional dialects. Example (10) is one of them.

(10) That’s right, isn’t it?

(2 lines omitted) If I heard someone, a man or woman, says janē ka yo, I would react like ee – ‘what?’ I wonder which regional dialect it is. I’m about your mother’s age and have a daughter and a son. If either one of them used such language, I’d be flabbergasted. I’m used to kitanai ‘bad’ language because I live in a region whose dialect is considered kitsui ‘harsh’ by people in other regions. Yet, I don’t want my children to use such language. (6 lines omitted) (Hatsugen Komachi, June 10, 2023)

According to this writer, regional dialects are not good language for anybody, women or men – and certainly not for herself or her children, even though they live in an area where a regional dialect is spoken. (See Section 4.4 for further discussion on this matter.)

4.4 The Acceptability of Language Use Depends on the Context (Group 2 Responses)

Group 2 in Table 1 constitutes 26% (14 posts) of all responses. Although details differ, the responses in this group all share the idea that language use depends on the context and that it is all right for young people to use the kind of language A is talking about among friends, but not elsewhere. This is illustrated in Example (11).

(11) TPO

Youth language has existed in all times. There are, of course, people who do not use it, but I think it is all right to use it, but only among friends. Even among family members, it should not be used toward parents because parents and children are not equal. (2 lines omitted) Do you by any chance use it to people other than your friends, like your teachers? If so, you should make an effort to correct it now because you will be in trouble when you start working. (Hatsugen Komachi, June 11, 2023)

As demonstrated in the heading of the post in (11), the writers in Group 2 emphasise the “TPO” – time, place, occasion – of language use. Many of them consider the language A is talking about inappropriate to use toward adults or people who are not friends. They also underscore that using good language can be an asset in adult life. These writers thus assume that different kinds of norms should apply to different social contexts.

The writer of Example (12) also subscribes to the idea of TPO. She also brings up an issue concerning regional dialects.

(12) Language use is TPO

If you understand TPO and differentiate language use according to the situation, I don’t think it’s a problem. I (obasan ‘older woman’) am close to 60 years old and have been zūtto kuchi ga warukatta ‘using bad language all these years.’

The language of the place in which I was born and grew up was kitanakute ‘bad’ and when I used my dialect, my mother warned me against it. So, by the time I went to elementary school, I was hyōjungoppoi hanashikata o shiteta ‘speaking kind of standard Japanese.’ But still that was like the so-called otoko kotoba ‘men’s language.’ The friends I had as I grew older were all alike using kuchi ga warukatta ‘bad language.’

Even now my language is bad. I talk with my daughter using muchakucha na ‘topsy-turvy’ language. I speak somewhat carefully toward my husband, but it is basically arappoi ‘rough.’ But it doesn’t matter because I’m seen as a person who differentiates language use according to the situation. My friends from student times usually used ossoroshī ‘awful’ language, but they changed it when they talked with their teachers, parents, workers, and strangers. (7 lines omitted) (Hatsugen Komachi, June 12, 2023)

The writer of this post shares the negative view of regional dialects seen in Examples (4) and (10) but expresses it more extensively, suggesting the powerful influence of the standard language ideology (Milroy, 2001) underlying the Japanese government policy enforced since the beginning of modern Japan. According to this ideology, standard Japanese is correct and good language, and regional dialects are incorrect and bad (Lee, 1996; Okamoto and Shibamoto Smith, 2016). The writer of (12) characterises the dialect spoken in her hometown with quite negative attributes, such as warui ‘bad,’ kitanai ‘bad,’ mechakucha ‘topsy-turvy,’ arappoi ‘rough,’ and ossoroshī ‘awful,’ and explains that she uses it daily but not exclusively, because she (as well as her friends) alters her language use to adapt to the situation.

Lastly, many writers in all three groups expressed their conviction that language use can be an asset that helps one to appear educated, refined, taken seriously, and the like – a constructionist’s view that language does not simply reflect the context, but rather can be used creatively to construct it.

5.1 Intersectionality and Diverging Views of Norms

We saw in the preceding section diverging ideas that native speakers of Japanese have about how one (in this case, a teenage girl) should talk. What gave rise to this divergence? The analysis indicates that even though the heading of the initial post (i.e. ‘I’m reprimanded for my bad language use because I’m a girl’) concerns only girls/women, it is not only gender that the respondents to the initial post took into consideration in evaluating A’s language use. This is because A is not just a girl but can be described in terms of many other social and contextual variables. Accordingly, in evaluating a person’s speech, some people may focus on one variable and others on other variables. Importantly, however, social and contextual factors do not directly govern the evaluation of language use because such evaluation is by definition an ideological process, as discussed in Sections 2 and 5.2.

Table 2 is a modified version of Table 1. The third column of Table 2 shows the criteria used in evaluating A’s language use that are common to all posts in each group, and the fourth column, criteria mentioned in only some of the posts.

View Full Size

Views of speech norms observed in responses to the initial post in (1)

Citation: Contrastive Pragmatics 7, 1 (2026) ; 10.1163/26660393-bja10150

• Download Figure
• Download figure as PowerPoint slide

As evident in Table 2, different people focused on different social and contextual aspects of A as a person, as she represents herself in her post. Group 1 writers focus on class status. Group 2 considers more than one aspect (i.e. age, intimacy), assuming different norms for different social situations. Group 3 focuses on gender. Some respondents in all three groups (five writers altogether) also used another criterion, namely standard Japanese as opposed to rural dialects, which they consider kitanai ‘bad’ and inappropriate for an educated/decent person or girl to use.

Note that while Tables 1 and 2 illustrate the diversity in the views of speech norms, the majority of posters on Hatsugen Komachi are presumed to be adult women (note 2). An analysis of metapragmatic commentary from different social groups, such as teenage girls or men/boys, might well have results that exhibit different views of norms – a topic for future study.

5.2 Ideologies, Speech Norms, and Indexical Meanings

Let us look at Table 3. The first column lists the linguistic forms mentioned in the discussion thread in question divided into two levels: “good” forms (first row) and “bad” forms (second row); the second column shows the attributes (qualities and stances) of those linguistic forms noted by posters; and the third column corresponds to social groups of speakers deemed to use the forms in the first column.

View Full Size

Stratified indexicality indicated by the responses to the initial post in (1)

Citation: Contrastive Pragmatics 7, 1 (2026) ; 10.1163/26660393-bja10150

• Download Figure
• Download figure as PowerPoint slide

As evident in Table 3, the same linguistic forms are perceived quite differently by different people. Eckert (2008: 455) argued:

[T]he very fact that the same linguistic variables may stratify regularly with multiple categories – e.g. gender, ethnicity, and class – indicates that their meanings are not directly related to these categories but to something that is related to all of them. In other words, variables index demographic categories not directly but indirectly (Silverstein 1985), through their association with qualities and stances that enter into the construction of categories.

The link between a particular linguistic form/variable and the kind of speakers assumed to use that variable is made via the pragmatic meanings (or qualities and stances) associated with the variable (see Ochs, 1992 for a discussion of indirect indexicality). This link is undoubtedly ideological because not all women or educated people use “good” language. Rather, these are stereotypical images of persons of a particular social group (e.g. women/men, people in higher/lower social classes, urban/rural residents) interacting in particular social situations (e.g. speaking with a stranger or a familiar person, or on formal/informal occasions). That is, Table 3 illustrates hierarchically ordered indexical signs that comprise part of what Blommaert (2010: 38) called “orders of indexicality,” that is, indexicalities “that operate within large stratified complexes.” The view of women’s speech norms that A’s mother maintains treats men and women unequally as part of stratified orders of indexicality. Moreover, the normative view of women’s speech does not concern only gender inequality but also inequality related to class status, regions (i.e. centre vs. periphery), and standings of varieties of Japanese, as these are all part of the “large stratified complexes” of the orders of indexicality associated with the use of the Japanese language.

5.3 Implications for Styling in Situated Practice

Evidently, native speakers are highly aware of what they believe to be the norms for each social situation, but this does not mean that they always conform to such norms in situated language practice, as demonstrated by previous studies (e.g. articles in Okamoto and Shibamoto Smith, 2004). How then do speakers as social actors choose the linguistic (and other semiotic) forms that shape their speech style? It seems that speakers do negotiate speech norms in a given situation, but how? While this question calls for serious future research, I present in what follows a couple of examples from previous studies that may offer some clues.

Okamoto and Morimoto (2023) examined eight dyadic conversations of male and female college students with regard to the use of seven normatively gendered multilevel variables. They found that in contrast to the common perception that gender difference among young speakers has decreased considerably, the participants’ use of these variables was largely gendered except for the use of SFF s, which was hardly gendered. This suggests that linguistic gender norms are highly but not always relevant to actual practice. Furthermore, their language use was much more complex in its details. Two women, for example, used normatively feminine forms in mixed gender conversations more frequently than in same-sex conversations, suggesting the effects of heteronormative expectations depending on the interlocutor. However, they largely avoided feminine SFF  s, which are often associated with older or mature women (e.g. Inoue, 2006). Thus, one of the conclusions Okamoto and Morimoto drew was that the female participants used masculine or neutral SFF  s to construct their youthfulness, while at the same time they attempted to construct femininity by using other variables considered feminine, although the degree of their use differed depending on the context, including the interlocutor’s gender.

In another example, Okamoto (2021b) included a comparison of two conversations: one between two standard-Japanese-speaking female college students from Gunma and the other between two elderly women who spoke Yamagata dialect. In both cases, the participants were close friends. The study focused on the use of two kinds of normatively strongly masculine forms:⁶ SFF s and the phonological form [ē], a contraction of [ai] and [oi] (e.g. janē instead of ja nai ‘is not’). Here, I only discuss the results of the latter because it corresponds to one of the forms criticised by A’s mother (i.e. ~janē ka). Table 4 shows the results.⁷

View Full Size

The use of the normatively strongly masculine form [ē]

Citation: Contrastive Pragmatics 7, 1 (2026) ; 10.1163/26660393-bja10150

Source: Based on Table 1 in Okamoto (2021b)

• Download Figure
• Download figure as PowerPoint slide

As shown in Table 4, all four women used this strongly masculine form, but they differed in the quantity of its use – and in the quality. The two Gunma women used the contracted [ē] only 10% and 23% of the time and when they used it, they accompanied it with some kind of hedge, including laughter and tte kanji ‘like’ (e.g. while laughing: atama hakaraka n̲ē̲ shi ‘My mind is not working’). Such use of hedges suggests that they were aware that they should not use a ranbōna ‘rough’ form like [ē]. In contrast, the two Yamagata women used it 67% and 87% of the time, and without any hedges (e.g. shoppagu n̲ē̲ na ‘it’s not salty, right?’); they laughed, but only when the talk involved funny content. This suggests that these women considered the use of [ē] as part of their normal speech and not strongly masculine, regardless of its perception as rough, bad, and unfeminine by those who uphold the hegemonic, standard Japanese, ideology of women’s speech. Moreover, unlike the Gunma women, the Yamagata women also used masculine SFF s and did not use feminine SFF s. As argued by Blommaert (2010: 80):

“[T]he margin,” so to speak, is not necessarily a space in which people fail to meet norms, but it can as well be seen as a space in which different but related norms are produced, responding – “ecologically,” so to speak – to the local possibilities and limitations.

Importantly, however, according to Okamoto (2021b), even though the Yamagata women used normatively masculine [ē] and SFF s extensively, they did not totally ignore (hegemonic) norms of women’s speech. In fact, they constructed a polite and cooperative, that is, “feminine,” speech style by using forms such as humble expressions and backchannels to indicate listenership. Thus, neither the Gunma nor the Yamagata women’s language use can be accounted for in terms of a simple binary criterion such as normative versus nonnormative.