Gendered representations in swearing: bitch and bastard in the spoken BNC2014

Abstract

This study examines how bitch and bastard construct gendered identities in contemporary British English conversation. Using corpus-assisted critical discourse analysis of the Spoken BNC2014, it examines collocational patterns and “be + bitch/bastard” constructions to trace how gendered meanings are enacted across speaker and target sexes. bastard predominantly targets men, representing masculinity through moral evaluation, functioning as a discursive resource for policing fairness and integrity. BITCH constructs more variable representations: it is frequently used by and about women to regulate interpersonal and emotional conduct, yet can also mark assertive femininity or position men outside socially recognised norms of masculinity. These patterns highlight how moral and relational discourses intersect in the linguistic representations of gender, sustaining long-standing associations of masculinity with public morality and femininity with emotional virtue. The findings show that derogatory language remains a critical discursive site where gendered identities and hierarchies are reproduced, contested, and occasionally re-signified in everyday interaction.

Gender and swearing

Patterns of language are often discussed as characteristic of particular social groups, such as social classes (Guy, 1988) and genders (Newman et al., 2008). Research on gender and language has explored how speakers’ linguistic choices both reflect and reproduce gendered identities. This includes differences in thematic focus, lexical choice and structural complexity (Jespersen, 2013), and gender-specific characteristics reflected in language use (Coates, 2015). In such research men’s and women’s language use is often compared. For example, Newman et al. (2008) associates men’s language with references to external events and objects, and women’s with interpersonal and psychological processes. It has also been claimed that women’s language is concerned more with relational maintenance (Rayson et al., 1997) and is often interpreted as avoiding open conflicts through mitigating stance-taking (Eckert and McConnell-Ginet, 2013).

In such a context, it is hardly surprising that swearing has been treated as a locus of variation in male and female speech, though the social context in which gendered swearing has been studied has varied over time. Historically swearing was perceived through a gendered lens. At least from the late seventeenth century (McEnery, 2006), swearing was socially coded as masculine, while women were seen as custodians of morality and propriety, roles that ostensibly precluded the use of coarse language. As a result, we would argue early, intuition based studies of swearing in English assumed that women swore less and used weaker bad language than men (e.g. Lakoff, 1973). With the emergence of large corpora and ethnographic data, such categorical claims have been questioned. McEnery (2006) finds that men and women may use generally the same amount of what McEnery terms “bad language” but overuse different sets of words, respectively. There are also studies showing the two genders’ swearing differences in terms of functions of swearing (e.g. McEnery and Xiao, 2004), tolerance for the use of swearwords (e.g. Jay and Janschewitz, 2008), and variation by the gender of the interlocutor (e.g. Beers Fägersten, 2012). For example, men have been reported to show greater tolerance of profanity (Jay and Janschewitz, 2008). Comparative contrasts have also been identified in written discourse: McEnery and Xiao (2004) claim that texts targeted at female readers generally use fuck less than those aimed at men.

Given that language use both reflects and shapes social identity (De Fina, 2011), swearing provides a valuable lens for examining gendered identity. The use of swearwords not only reaffirms social positioning but also participates in the (re)production of gender stereotypes. Thelwall’s (2008) analysis of MySpace, a youth-oriented social networking site, found that young women in the United Kingdom used strong swearwords at rates similar to young men. In contrast, their counterparts in the United States showed a clear gender gap, a pattern taken to indicate that gendered expectations surrounding swearing may be shifting in online settings in the UK. McDowell and Schaffner (2011) observe that in a reality TV programme about male and female athletes in American football, the women’s use of insults, typically claimed to be masculine linguistic practices, served to challenge the masculinity stereotype linked with sports competitions. Likewise, Lucchetti (2021) demonstrates that women’s use of Russian mat (i.e. obscene language based on sexual taboos) in recent years signals the renegotiation of gender roles, challenging long-standing associations between profanity and masculinity in Russian society.

This article focuses on two apparently gendered swearwords, bitch and bastard, to explore how claims about gender and swearing manifest in present-day British English. These terms have been described as gendered counterparts (see Greenbaum, 2012; Montagu, 2001) so represent a good pair of target words for our study. These words also fall in what Love (2021: 739) terms “pure swearwords” and, similarly, “core swearwords” described in Beers Fägersten (2012: 18–19), both referring to items widely accepted as swearwords across contexts. Focusing on these canonical examples thus allows us to avoid marginal cases and to investigate terms whose status as swearwords is well established. ¹ This focus also allows us to explore words for which the literature reports contradictory findings. Previous studies present mixed evidence on the gendered usage of these terms. Gauthier and Guille (2017) found that female users (ages 12–30) in the UK showed a preference for bitch and males for bastard in tweets, while their gender assignments were inferred indirectly using profile-name matching to historical baby-name lists. By contrast, Drummond (2020) observed an opposing trend in spontaneous speech, yet his findings derived from a highly specific and socially bounded cohort of 26 individuals (ages 14–16) in a single Pupil Referral Unit in Manchester, limiting its generalisability. What we can achieve in this paper is to expand the view of the usage of the words in present day casual conversation and provide a broader context to interpret the results of Drummond in particular. However, the main advantage of studying these words is that by examining how these words are used in everyday conversation, and in particular by considering who is targeted with these terms in conversations, we can gain insights into a society’s expectations and perceptions of role and gender.

Gendered social actor representations

Gender differs from sex in that it is socially constituted rather than biologically assigned (Coates, 2015). It is therefore commonly understood as a situated performance: an identity continually negotiated through interaction. This distinction is crucial in the study of linguistic gender representation, which examines not only “who” produces discourse but also “what” social identities are represented and “how” they are constructed (Sunderland, 2004).

The “how” concerns the linguistic strategies used in discourse, while the “what” includes not only the representation of social actors, which in our case are women and men, but also the underlying ideologies conveyed through language use. These ideologies are constructed by social, cultural and historical contexts, which may not be expressed explicitly but may instead be conveyed through specific linguistic connotations. For instance, Pearce (2008) examines collocational and grammatical constructions of man and woman in the 1994 British National Corpus (BNC), revealing that representations largely reproduce conventionalised gender stereotypes in British society. Caldas-Coulthard and Moon (2010) compare lexical choices that British tabloids and broadsheets make to label male and female social actors. By comparing the adjectival pre-modification of man, woman, girl and boy in the newspapers, they argue that the media’s use of narration strategies for gender stereotypes is implicit for the average public reader.

Yet in these studies of gender representations in everyday language use, only a handful of studies have focused on more hostile contexts of production. For instance, Krendel (2020) explores how gendered lemmas (woman, girl, man and guy) are represented in three anti-feminist sub-communities of the discussion forum Reddit. She concludes that while female social actors were morally belittled and sexually constructed, male social actors were portrayed in inferior positions, revealing an internally conflicted masculinity.

However, there has been limited research into how gender identities are represented through the use of ostensibly gendered swearwords. While mentions of bastard and/or bitch in swearing studies typically focus on their usage by different genders (e.g. Beers Fägersten, 2012; Drummond, 2020; Gauthier, 2017; Love, 2021), they do not systematically analyse how those words represent gender, or critically connect these linguistic representations to the social, cultural and historical ideologies they reproduce. To our knowledge, no corpus-based study of everyday conversation has systematically explored the gender identities that these two terms express, let alone if and how these would vary across speakers and targets of different genders.

To address these gaps, the present study focuses on bitch and bastard, ostensibly paired representations of females and males in swearwords. The research is primarily concerned with the following questions:

1. What gendered representations of bitch and bastard emerge in British-English conversation?

2. How do the sexes of speakers and targets affect these representations?

In line with the BNC2014 metadata, which records speakers as male or female, this study employs sex as a categorical variable. However, in interpreting the results, we treat these patterns as reflections of socially constructed gender norms. This allows us to examine how the language of swearing both represents biological categorisation and negotiates broader gendered expectations in discourse.

Methodology

Data collection

The analysis draws on the Spoken British National Corpus (BNC) 2014 ² (Love et al., 2017), a corpus containing 11.5 million words of orthographically transcribed spontaneous conversations among 672 British English speakers, collected between 2012 and 2016. ³

Within the Spoken BNC2014, bitch occurs 318 times (27.84 instances per million words), and bastard appears 185 times (16.20 instances per million words). A comparison with the conversation subcorpus of the BNC1994 indicates that bitch has remained relatively stable in frequency (from 26.12 to 27.84 instances per million words), whereas bastard has declined markedly (from 40.07 to 16.20 instances per million words). This suggests a shift in their relative prominence in everyday British English over the past two decades.

All analyses were conducted via the online corpus tool CQPweb (Hardie, 2012), which provides both technical flexibility and access to sociolinguistic metadata. CQPweb allows searches to be performed using both simple and CQP syntax queries, facilitates filtering by speaker sex and other metadata fields, and displays dialogues in a speaker-focused format. These affordances supported a discourse-oriented analysis of how bitch and bastard represent and evaluate gendered social actors in present day spoken English.

Corpus-assisted critical discourse analysis

Both qualitative and quantitative analytical methods have been employed in studies of gender and language. Some researchers adopt qualitative methods to analyse discourse in detail (e.g. Stapleton, 2003), while others use large datasets to identify broader patterns and generalisations (e.g. Newman et al., 2008).

Recent years have seen increasing use of methods from corpus linguistics (CL) in the study of swearing (Love, 2021; Lutzky and Kehoe, 2016). It is important to note that CL is not a purely quantitative approach: it typically combines quantitative and qualitative analyses. Quantitative techniques help to identify the overall aboutness of the data and reveal patterns that may not be immediately visible to researchers (Baker and Egbert, 2016) while qualitative interpretations allows for theoretical refinement and the development of new insights (Baker et al., 2008).

In this respect, critical discourse analysis (CDA) provides a complementary perspective. As a cross-disciplinary approach, CDA examines how socio-power abuse and inequality are reproduced, legitimised, and challenged through language uses (i.e. discourse) (van Dijk, 2015). Given that swearing carries context-sensitive interpersonal and evaluative meanings, the frequency comparison in Section “Data collection” offers only an initial picture. To understand how bitch and bastard contribute to gendered representations in conversation, it is necessary to connect statistically salient patterns with close analysis of specific contexts. Corpus-assisted critical discourse analysis supports this integration, and it is in this spirit that we use this method to investigate swearwords in this paper.

Analytical procedures

We designed two complementary analytical perspectives to examine how bitch and bastard function as discursive resources for representing gendered social actors.

The first stage involved a collocation analysis to provide an overall picture of the meanings and associations (Baker, 2023) of bitch and bastard. A span of three words on either side of each node was used to capture the immediate lexico-grammatical environment, including modification and predication patterns that shape social actor representations (van Leeuwen, 2008; Wodak, 2001). Log Ratio (filtered) was selected to measure the strength of association between node and collocate, with Log-Likelihood (LL) test automatically applied in CQPweb (15.0 for bitch and 14.4 for bastard) to ensure statistical reliability (Hardie, 2012). This ensured that low-frequency collocates were retained only if their association with the node (i.e. bitch or bastard) passed the LL filter, indicating that the co-occurrences were not due to chance and offered analytically meaningful evidence for understanding the node words. This process produced 29 collocates for bitch and 28 for bastard. Extended concordances for these collocates were then qualitatively examined to identify recurring evaluative and representational patterns, which are presented in Section “Collocational comparisons.”

The second stage involved a close analysis of the “be + bitch/bastard” construction to examine how explicit characterisation constructs across speaker and target sexes. Copular constructions were chosen because they predicate characteristics or specify identities to subjects (Declerck, 2011) and they represent the most frequent syntactic pattern for both terms in the corpus, accounting for approximately 30% of all concordances. Prioritising this construction enabled detailed interpretation within a recurrent discursive pattern, providing depth over broader but less frequent forms. Related expressions (e.g. bitchy and bastardised) were excluded due to low frequency and indirect relevance; they are therefore mentioned only where relevant to the analysis.

CQP queries used were [lemma=“be”] []{0,3} [word=“bitch” | word=“bitches”] and [lemma=“be”] []{0,3} [word=“bastard” | word=“bastards”], returning 116 cases for bitch and 48 for bastard. Speaker sex was identified from corpus metadata, while target sex was inferred from contextual cues such as pronouns, titles, kinship terms, and other explicitly gendered references. Ambiguous cases were discussed jointly by the first and second authors. After excluding false positives (for example, cases where be and bitch/bastard appeared in separate sentences rather than forming the “be + bitch/bastard” construction) and cases lacking clear referents, 82 cases of bitch and 33 of bastard were retained for qualitative interpretation. We coded each concordance for the primary evaluated aspects in the use of bitch or bastard. The coding scheme was developed though data-driven, iterative reading of the extended concordances to identify recurrent evaluative foci. The resulting patterns are reported in Section “Concordance analysis.”

Findings

Collocational comparisons

Collocates of bitch and bastard were categorised according to their meaning in the concordances, allowing comparison of representations across categories in Table 1. Some overlap occurred among categories (e.g. gender indicators [female and male] and sexual connotation). Likewise, individual collocates may perform different semantic or pragmatic roles across concordances, leading to their inclusion in multiple classifications.

Table 1.

Collocation categories of bitch and bastard.

No.	Categories	bitch	Categories	bastard
1	Female	whore, bitch, she, slag	Male	bastard(s)
2	Sexual connotation	whore, lick, fuck, slag	Sexual connotation	dirty
3	Physical characteristics	cheek-bones, hottest, boob, fat	Physical characteristics	fat, dirty
4	Behavioural characteristics	criticises, inconsiderate, queen, messy	Behavioural characteristics	sly, sneaky, selfish, stingy, lazy, sniffling, spoilt, dominator, jammy, lucky, commie, smug, evil
5	Emotional characteristics	miserablest	Emotional characteristics
6	Confrontation and intensifier	fuck, fucking, you, face, cut, utter, such	Confrontation and intensifier	you, fucking, frigging, absolute, bloody
7	Miscellaneous category	a, refer, bug, being, breathe, check, kuh, hence	Miscellaneous category	bas-, Geordie, those, huh, oh, wait

In Table 1, collocates listed in corresponding rows are compared. The first five categories (shaded in white) capture semantic dimensions of representation, while the remaining two (shaded in grey) describe interactional or functional roles such as intensifier and out-of-scope lexical items. As the present study primarily focuses on gendered representations, the discussion below centres on the semantic categories, referring to the non-semantic ones only where they shape gender-related interpretation.

Based on the collocational results and their extended concordances, we interpret the representations of bitch and bastard from two complementary perspectives: (1) gender and sexualisation, based on collocates in the first three categories; and (2) interpersonal relations, derived from collocates within the subsequent two categories.

Gender and sexualisation

This section examines the collocates in “female/male”, “sexual connotation”, and “physical characteristics” categories in Table 1. While distinct, “physical characteristics” are included here as they contribute to the sexualised evaluations of targets.

Collocates in the “female” and “male” categories confirm that bitch and bastard are primarily gender-specific in usage. This suggests, prima facie, that bitch is inclined to be associated with female targets and bastard with male targets. However, both terms occasionally refer to the opposite sex, as will be discussed in the relevant examples below (Examples 1, 16, and 17).

Both bitch and bastard carry elements of sexualisation directed at their targets, though this tendency is much stronger for bitch. bastard only significantly collocates with dirty in the “sexual connotation” category (Log Ratio = 6.606), which in two-thirds (66.67%) of its occurrences it denotes sexual impropriety. In Example 1, a female speaker (S0246) is teased by a male interlocutor (S0249) as a “dirty bastard”. This teasing appears to arise from her tendency to interpret otherwise innocent comments in a sexualised way, a reading supported by another participant’s (S0357) remark that she goes “always a step too far”.

Example 1

S0246: and everybody says I’m weird like

S0249: >>you are though

S0246: yeah but –UNCLEARWORD

S0249: that’s you rubbing off on him

S0246: I don’t rub anything on him

S0249: dirty bastard

S0357: always a step too far – ANONnameF

. . .

S0249: she’s never happy even if the conversation ended she has to add to it something dirty at the end

For bitch, four collocates (whore, slag, lick, fuck) demonstrate explicit sexual meanings in their concordances. Whore collocates show a particularly strong association (Log Ratio = 9.847), appearing as “bitch whore” in 87.5% of the concordances in which whore appears as a collocate of bitch. Slag likewise carries a strong sexual connotation (Log Ratio = 8.325). Further evidence of sexualisation arises from lick and fuck collocates, as in Example 2, where “bitch” is embedded within a sexual banter (“lick my bits”).

Example 2

S0400: oh --UNCLEARWORD bitch lick my bits

Sexualisation associated with bitch is also evident in the “physical appearance” category. Here, bitch occurs significantly with collocates referring to female physical features, including cheek-bones (Log Ratio = 13.547) and boob (Log Ratio = 8.055), as well as evaluative adjectives such as hottest (Log Ratio = 8.503) and fat (Log Ratio = 5.150). The concordances suggest that women’s bodies are recurring sites of judgement. Example 3 illustrates this connection: speakers in this example are discussing a female acquaintance. They initially praise her fitness (“a fucking fit body” and “six-pack”) before turning to criticism of her head as “a bit of a shambles”, a “bit of a letdown”, and “head transplant girl”. They then compare her to a “drag queen” or “sister”, and explicitly address her as “a man”. This excerpt shifts from admiration to defeminisation, policing what counts as (socially) acceptable feminine appearances.

Example 3

S0041: but she has a fucking fit body (.) like she has a si- six-pack and everything

S0084: wow

S0041: her head just is a bit of a shambles

S0084: bit of a let down

S0041: yeah

S0084: you can’t do anything about your head (.) that’s the thing

S0041: yeah (.) oh there’s more (.) that’s head transplant girl again

S0084: urgh she’s a man she’s a man

S0041: yeah

S0084: she’s a drag queen

S0041: sister

S0084: bitch

S0041: bitch (.) hence boob reduction

S0084: oh yeah (.) yeah

Among the aforementioned bitch collocates, hottest appears to be the only potentially complimentary one. Both of its concordances occur in a discussion between a female speaker (S0246) and a male speaker (S0245) about Blurred Lines by Robin Thicke, a song the male speaker describes as “about the rape” (see Example 4). While hottest might appear evaluative in a positive sense within the lyric concordance, S0246 treats its presence in a workplace as socially unsuitable, particularly in the presence of “loads of old people”.

Example 4

S0245: oh (.) er erm that Robin Thicke song’s on the playlist at work

S0246: what Robin Thicke song?

S0245: you know the one about the rape

S0246: Blurred Lines

S0245: yeah

S0246: and doesn’t it say you’re the hottest bitch in this place?

S0245: mm

S0246: I just realised the other day that it’s not the like censored version

S0245: oh is it not?

S0246: so it was like really quiet and there was like loads of old people sat there and he just went you’re the hottest bitch in this place and that’s like yep changing that

For BASTARD, the only “physical characteristics” collocate is fat (Log Ratio = 5.608), yet its concordances suggest that it is rarely used as a literal judgement of body size. It is also poorly dispersed – two thirds of fat’s occurrences appear in one conversation (text ID: SKB5; see Example 5). The expression is prompted by someone sitting on the speaker’s (S0208) “backy”, and “Mr fat bastard” functions as light-hearted, situational teasing rather than an appearance-based judgement or explicit insult.

Example 5

S0208: yeah where’s your backy?

S0202: I don’t know

S0215: –ANONnameM’s sat on it

S0202: oh

S0214: huh

S0202: –UNCLEARWORD fat bastard

S0214: Mr fat bastard

Interestingly, no comparable collocational patterns emerge linking fat with non-derogatory gendered references. woman, girl, and man did not appear as significant collocates. The much lower strength of boy (Log Ratio = 3.667) and guy (Log Ratio = 2.520) co-occurring with fat compared to bitch (Log Ratio = 5.150), together with the playful use of fat bastard (e.g. the male case in Example 5), suggests that references to men’s weight can be read as casual description or banter. In contrast, fat’s more frequent association with bitch (for female targets) in derogatory contexts (e.g. Example 6, where S0046’s response confirms fat is used literally), alongside bitch’s unique collocation linked to physical evaluation (e.g. Example 3), indicates that women’s bodies are subject to greater evaluation and insult than men’s in the conversations examined.

Example 6

S0046: –UNCLEARWORD by the way (.) you know –ANONnameF you’ve met – ANONnameF (.) the bird in here that you you saw (.) she’s not pregnant

S0041: I knew it

S0046: I knew it and all

[unknown male]: fat bitch

S0046: to be fair (.) she’s not fat you know she’s not fat (.) you know everything (.) no you never met you never met her

Interpersonal relationship

Collocates in the “behavioural characteristics” and “emotional characteristics” categories reveal the interpersonal dynamics associated with bitch and bastard. Both terms co-occur with lexemes of dominance, notably queen (Log Ratio = 5.792 for bitch) and dominator (Log Ratio = 14.329 for bastard). The concordances sometimes reveal playful or ironic tones, expressing a positive, admiring stance towards the referent. In Example 7, a male speaker (S0599) refers to his female flatmate as a “queen bitch”, while praising her confidence and sociability. The expression “queen bitch” thus conveys admiration with humour, constructing assertive femininity as simultaneously desirable and excessive.

Example 7

S0599: she’s called –ANONnameF and I feel like I’ve known her for like ages now because she’s so cool and

S0598: yeah

S0599: like we just so comfortable with each other we just hit it off

S0598: >>yeah (.) mm

S0599: >>it was so weird (.) she’s the main she’s the queen bitch (.) . . .

Beyond these shared associations of dominance, the collocational patterns reveal distinct evaluative emphases. While both bitch and bastard comment on behaviours impacting upon others, bitch collocates tend to target emotion, whereas bastard collocates target moral integrity.

For bitch, concordances of miserablest and criticises point to expectations that women should display empathy and emotional attentiveness. In Example 8, S0188 describes his girlfriend’s friends as “the miserablest bitches”. His complaint that “she never made any effort to talk to me either” extends this evaluation to his girlfriend, framing both the friends and the partner as failing to provide expected emotional labour or social rapport. In Example 9, the female target is labelled a “bitch” for “criticis(ing)” the female speaker and making her feel “really upset” (see Example 9). In such contexts, bitch sanctions perceived breaches of care, sensitivity, or interpersonal warmth.

Example 8

S0188: yeah two of them were were the miserablest bitches I ever met in my entire life

S0200: what –ANONnameF’s friends? chatty –ANONnameF’s friends?

S0187: yeah mate

S0188: she never made any effort to talk talk to me either

Example 9

S0439: erm (.) slagging her off saying I can’t believe you’ve gone on holiday with somebody who upset me I can’t believe you’re being so disloyal (.) erm you obviously don’t care about me she’s --ANONnameF’s such a bitch she criticises me and really upset me and if you’re friends with her then I’m really (.)

By contrast, bastard collocates such as sly, jammy, stingy, sneaky, selfish, evil, and lazy construct a moral rather than emotional evaluation. They describe self-interest and moral weakness: cunning (sly, sneaky), self-centeredness (stingy, selfish), and lack of virtue or diligence (lazy, evil). Some collocates, including jammy, lucky and spoilt, imply undeserved success or unfair advantage. In Example 10, the speaker refers to homeowners as “lucky bastards”, attributing their financial achievement to luck rather than legitimate effort.

Example 10

S0058: . . . did I tell you she’s bought a house? Or her and –ANONnameM have

S0120: yeah (.) yeah

S0058: yeah (.) next to her parents (.) awkward (.) but I don’t know (.) I’m sure it‘ll be good and it doesn’t (.) mm?

S0120: lucky bastards (.) lucky bastards

S0058: yeah (.) go and live in Sunderland it doesn’t cost anything

Concordance analysis

This section examines the representations of bitch and bastard in the “be + bitch/bastard” constructions, and how they vary according to speaker and target sex (Tables 2 and 3).

Table 2.

bastard concordances.

Target category	Female speaker (n = 16)	Male speaker (n = 17)
Targeting same sex	0 (0.00%)	10 (58.82%)
Targeting different sex	8 (50.00%)	0 (0.00%)
Targeting unspecified human referent(s)	7 (43.75%)	5 (29.41%)
Targeting objects	1 (6.25%)	2 (11.76%)

Table 3.

bitch concordances.

Target category	Female speaker (n = 58)	Male speaker (n = 24)
Targeting same sex	40 (68.97%)	1 (4.17%)
Targeting different sex	6 (10.34%)	11 (45.83%)
Targeting unspecified human referent(s)	3 (5.17%)	2 (8.33%)
Targeting objects	9 (15.52%)	10 (41.67%)

Both female and male speakers use the “be + bastard” construction with comparable frequency, but it exclusively targets men. Among male speakers, “be + bastard” tends to convey moral judgement (60.00% of human target concordances), especially perceived unfair or exploitative behaviour (33.33%). Explicitly playful uses account for only about one quarter (26.67%). The moral dimension is demonstrated in Example 11, where the speaker (S0453) labels someone a “selfish bastard” after perceiving ingratitude for their help.

Example 11

S0453: >>he is he’s a selfish bastard and fundamentally after everything I’ve done to support him when his grandma died who do you think was the first person he called?

S0454: >>yeah I know

The dominant characteristics identified in Section “Interpersonal relationship” are also evident here. In Example 12, bastard is used to characterise a school bully, emphasising his misuse of power and dominance among peers.

Example 12

S0570: >>–ANONplace he’s a bastard

S0571: is he?

. . .

S0572[??]: >>and he’s like one of the main people in year ten

S0570[??]: >>yeah yeah –UNCLEARWORD bad men

Female speakers also use bastard to convey moral judgement (46.67% of human target concordances), particularly in relation to perceived misuse of power (26.67%), but additional characteristics emerge. These include interpersonal inconsideration (13.33%) and inappropriate sexual behaviour (20.00%), subtly aligning with the interpersonal concerns observed in the bitch collocates (Section “Interpersonal relationship”). In Example 13, a female speaker (S0202) calls taxi drivers “bastards” for blocking the entrance to a train station – an act viewed as inconsiderate. In Example 14, a schoolboy is called a “bastard” for inappropriate sexual behaviour (“wanking under the table” and “having sex on the stage”). This suggests that female speakers extend the evaluative scope of bastard to include social and sexual impropriety alongside moral judgement.

Example 13

S0202: fucking taxis I hate them they always do this block up the whole of the fucking train station entrance it’s like fucking move stop being cunts

S0218: (. . .) my car does not like this hill (. . .) why don’t they fucking pull over down there?

S0202: they’re just fucking being arseholes fucking taxi bastards I hate taxi drivers they’re such cunts

Example 14

S0440: but he was an absolute bastard at school

S0439: >>isn’t he the one that got told off for wanking under the table?

S0440: yeah

. . .

S0440: cos it’s not it’s not it’s not just wanking and

S0439: >>and he didn’t he get told off for having sex on the stage?

S0440: something like that yeah

The “be + bitch” concordances reveal more complex representations, differentiated by both sex of the speaker and sex of the target. Consistent with earlier findings (Section “Gender and sexualisation”), bitch predominantly targets females, with only seven cases (11.11% of human target concordances) referring to males. Overall, “be + bitch” is also used twice as frequently by female speakers as by male speakers in the corpus.

When female speakers describe female targets as bitch, the focus often lies on negative interpersonal attitudes and dominant characteristics that can make interaction difficult (70.00%) (Example 9). However, the tone is not always critical; bitch can also be self-ascribed in some female in-group contexts as a humorous or mock-self-deprecating label, functioning as a playful and affiliative stance-taking (Example 15).

Example 15

S0441: ah there’s a bug go away get out of my spinach

S0439: that’s what happens when it’s home grown

S0441: I killed it I had to I’m sorry sorry god

S0439: you just killed a bug you bitch

S0441: I’m an actual bitch

When directed by female speakers at male targets, bitch often marks deviations from traditional masculine norms (66.67%), such as portraying men as sneaky or untrustworthy. Example 16 demonstrates this use: the male target is labelled a “bitch” for being “sneak(y)”, “dodgy”, and willing to “throw you under a bus”. In this case, bitch partly overlaps with bastard, as both can be used to police male behaviour that diverges from expected masculine conduct.

Example 16

S0328: he’s got he’s got a sweet side but he is actually a bitch

S0383: not but he’s like

S0328: he’s would he would do dodgy shit if he could get away with it

S0383: but I feel like he’s very conscious of it like it’s totally constructed

S0328: he has moments of remorse but he he’ll fucking throw you under a bus though don’t don’t don’t mistake that I’ve seen him do the sneakiest shit to his friends but like you know he’ll still consider them friends and really love them but like he’s definitely a bitch I don’t know man he smokes too much weed or something

In some cases, bitch is also associated with subjugation or a perceived lack of agency among male targets. In Example 17, the term is used jokingly yet still suggests diminished power or control, implicitly contrasting with the ideal of masculine strength.

Example 17

[unknown female]: do you drive?

S0202: no but my boyfriend does so

S0216: oh right that’s handy

S0202: he’s my little bitch

Male speakers’ use of bitch towards female targets is highly variable, ranging from outright insult to uses that carry a positive evaluation. In derogatory contexts, the term polices perceived deviations from normative femininity, echoing patterns found in female-to-female usage. In Example 18, the insult “ugly bitch” targets the woman’s physical appearance and treats her perceived lack of attractiveness as grounds for derogation. Here, bitch intensifies the negative evaluation, presenting physical attractiveness as an expected aspect of normative femininity.

Example 18

S0245: >>your sister’s a fucking ugly bitch

S0249: >>like the ugliest effing person I’ve ever seen in my life

S0357: >>plus her sister as well used to wear wigs and that didn’t she?

S0249: yeah but she was still horrible

Conversely, two instances of “be + bitch” feature bitch used positively or affectionately by male speakers. In Example 19, the speaker refers to his grandma as a “bad bitch”, a phrase expressing admiration for her independence and for being “cool”.

Example 19

S0246[??]: your grandma’s well cool (.) I love that you had an early Christmas –UNCLEARWORD

S0246[??]: she’s like fuck yous I is gone away

S0400: basically she’s a bad bitch

S0245: fuck you all

S0246[??]: I’m old I‘ll do what I want

Only one “be + bitch” instance involves a male speaker using bitch to describe another male (Example 20). This, explicitly derogatory, usage critiques a perceived lack of resilience or toughness: again invoking the gendered expectation that men should display physical and emotional strengths.

Example 20

S0602: he’s a little bitch isn’t he? (.) like he didn’t he just keep on running away with that fight with that one guy?

Discussion and conclusions

The moral policing of masculinity

The consistent representation of bastard in the corpus highlights its function as a linguistic resource for policing the boundaries of acceptable, hegemonic masculinity in the UK in the early 21st century. It sanctions men who are perceived to violate the implicit ethical code underpinning dominant masculine norms (Baker and Brookes, 2025), that is, assertiveness, competitiveness, success, and rational self-control, yet with the expectation that these should be achieved through legitimate effort. bastard thus functions as a discursive mechanism maintaining the moral dimension of masculinity. Its collocates further highlight this pattern: sly and sneaky suggest cunning and manipulation, stingy and selfish suggest self-interest, while lazy and evil suggest moral weakness (see Example 16).

This moral regulation aligns with broader neoliberal discourses that invoke hard work and merit as tools of legitimation (Littler, 2013). Collocates like jammy, lucky, and spoilt reveal resentment towards men seen as having benefited from unearned advantages, echoing neoliberal discourses linking hard work to moral worth. Expressions such as “lucky bastard” (Example 10) question the legitimacy of success, distinguishing between effort-based achievement and undeserved fortune. The discursive representations thereby highlight a neoliberal moral economy that simultaneously celebrates individual success and condemns perceived unearned privilege.

Although this moral policing is consistent across the corpus, subtle differences emerge by speaker sex. Male speakers more often apply bastard to condemn breaches of authority or moral failings (see Example 11). Female speakers expand the evaluative scope to include interpersonal and sexual impropriety, criticising inconsiderate male behaviour (see Examples 13 and 14). These uses suggest that women re-signify the term not only to challenge unfairness, but also to assert moral accountability in interpersonal and sexual contexts, blurring boundaries with the discursive constructions of bitch discussed below.

The double-bind of femininity

Representations of bitch in the corpus reveal a contemporary double-bind in expectations of femininity (Kleinman et al., 2009). In the corpus, women are positioned as expected to be confident and assertive, yet simultaneously assessed for emotional care and bodily propriety.

Firstly, bitch frequently polices interpersonal conduct. Collocates such as miserablest and criticises sanction failures of empathy or sociability (see Examples 8 and 9). According to appraisal theory (Martin and White, 2005), these evaluations fall under the Judgement category, policing the targets’ behaviour against expectations of normality and propriety in social interaction. The women are positioned as neglecting expected emotional support or as being unduly critical. Physical-appearance collocates (e.g. boob, cheek-bones, fat) extend this regulation to the body, sustaining long-standing scrutiny of women’s looks and demeanour (see Examples 3, 4, and 6).

Secondly, bitch is sometimes semantically re-appropriated in positive evaluations. In a small number of cases, male speakers use expressions such as “bad bitch” to admire female resilience, independence, or “cool(ness)” (see Example 19). Originating from African American Vernacular English and later popularised through hip-hop (Koponen, 2021), its re-appropriation in British English suggests admiration for female confidence and autonomy, while detached from the misogynist contexts in which it first emerged. Yet the positive interpretation often aligns with a masculinised ideal of agency, authority and toughness (Baker and Brookes, 2025), meaning that the re-appropriation still sits within gendered norms.

Thirdly, female in-group use sometimes employs humour and mock impoliteness (i.e. surfance-form impoliteness not intended to cause offence; Culpeper, 1996) (Example 15). Here, bitch can signal closeness while still regulating behaviour, showing how women manage solidarity and social discipline simultaneously. When directed at men, the term usually points to perceived deviation from conventional masculinity (e.g. manipulation, submissiveness, and weakness) (see Examples 17 and 20). In these cases, bitch partially overlaps with bastard, but places more emphasis on perceived loss of power.

Historical resonances

While situated in present day British English discourse in this paper, the patterns observed for bitch and bastard resonate with enduring gender ideologies often traced to Victorian moral culture (Digby, 1992). These continuities can be viewed as discursive reproductions of the symbolic divide between the public and private spheres (Tosh, 2005): the former historically coded as masculine and associated with rationality, productivity, and moral duty; the latter as feminine, linked to emotional virtue and domestic care.

Our findings suggest that these moral grammars extend beyond non-derogatory language (e.g. Baker, 2014; Caldas-Coulthard and Moon, 2010; Pearce, 2008) to inform everyday evaluative discourse, including swearwords. The expectation that men should demonstrate integrity, rational self-control and success in public or professional life is reproduced when bastard is used to criticise moral failure or unearned success. Similarly, the association of femininity with caregiving and emotional support surfaces when bitch sanctions women perceived as tactless or self-centred. Through such ordinary acts of evaluation, contemporary British English speakers re-circulate historical constructions of masculinity as public morality and femininity as relational virtue, embedding these cultural scripts within the linguistic routines of everyday communication.

Conclusions

This study investigated how bitch and bastard construct gendered identities in present day British English conversation through a corpus-assisted critical discourse analysis of the Spoken BNC2014. By combining collocational and concordance analyses, it demonstrated how the two terms cover distinct yet interrelated dimensions of gendered evaluation and representation in everyday conversation.

bastard is used by speakers of both sexes, but almost exclusively targets men. It mainly functions as a moral critique, condemning selfishness, deceit, and unearned privilege, and thereby reaffirms expectations of masculine integrity and fairness. bitch, by contrast, occurs more frequently, particularly among female speakers, and primarily targets women. It operates as a form of interpersonal regulation, sanctioning perceived failures of empathy or social tact, and it also extends this regulation to the body through swearing that targets women’s physical appearance. At times, however, it is re-signified as praise for assertiveness or applied to men to critique deviation from normative masculinity.

These findings extend prior corpus-based research on gender representation (Baker, 2014; Caldas-Coulthard and Moon, 2010; Pearce, 2008) by showing that even overtly derogatory language participates in the same ideological processes observed in non-derogatory discourses. Widely and frequently used swearwords thus provide a revealing perspective on the moral and relational structures through which gendered identities are sustained, negotiated, and occasionally challenged in contemporary British society.

Methodologically, this study demonstrates the value of corpus-assisted critical discourse analysis for research on gender and language. Unlike Drummond’s (2020) ethnographically rich but localised study of teenage swearing in a specific institutional setting, the present analysis draws on a large and socially diverse record of spontaneous conversation, the Spoken BNC2014 (Love et al., 2017), enabling patterns of gendered swearwords to be examined beyond a single community of practice. This allows not only for more generalisable observations regarding the gendered use of swearwords in contemporary British society, but also for a more comprehensive analysis of how they discursively construct gender norms in everyday conversation. It traces patterns in gendered, moral, and interpersonal characterisation across collocation and concordance evidence, thereby combining empirical breadth with interpretive depth.

At the same time, several limitations should be acknowledged. Focusing on copular constructions offers analytical depth, but it necessarily excludes the implicit and less frequent syntactic environments. Moreover, the orthographic transcripts of Spoken BNC2014 omit prosodic and other multimodal cues, such as intonation and facial expressions, which may influence how swearwords are understood in interaction. Finally, this study is inherently synchronic, relying exclusively on Spoken BNC2014; tracing the historical shifts of gender representations behind these terms would require a comparative analysis with the Spoken BNC1994. Future research could extend the analysis to additional syntactic environments and incorporate multimodal corpora to capture a wider range of such gendered representations.