Introduction

Swear words in daily communication function as a speech act that is either intentionally or unintentionally imposed by the speaker upon the addressee, and serve as an externalised manifestation of language to achieve the speaker’s intended communicative effect. Their formal expressions, encompassing lexical items, phrases, and even full sentences, exist in both Eastern and Western societies. Research indicates that swear words account for 0.3–0.5% of daily language usage (McEnery, 2006), whereas their frequency exceeds 5% in social media communications (Jay, 2018). Swearing, the emotional use of taboo language (e.g., insults, epithets, and expletives), is intended primarily to express the speaker’s strong feelings rather than literal meaning. As a form of cultural taboo, its core function is to wound the listener or discredit a third party through stigmatising expressions (Allan and Burridge, 2006, 2009). The related English expressions include “swear words”, “bad language” or “taboo words” (Andersson and Trudgill, 1992), among which “swear words” is the most commonly used (O’Driscoll, 2020). Ci Hai, an encyclopaedic dictionary in China, defines it as “using abusive language against others” (Xia and Chen, 2009, p. 1518), while the Oxford Advanced Learner’s Dictionary characterises it as “to use rude or offensive language” (Hornby, 2012, p. 1563).

Definitions of swear words in the Chinese and English contexts are largely analogous, as such words are understood in both contexts to be malicious or vulgar expressions employed to inflict humiliation. However, Chinese and English belong to distinct language families: the former emphasises parataxis (a linguistic structure characterised by a heavy reliance on implicit mechanisms, such as context, shared meaning, and cultural understandings, to convey logical-semantic relationships between constituents within sentences), whereas the latter prioritises hypotaxis (a linguistic structure characterised by the predominant use of explicit grammatical markers, such as inflections, word order, and function words, to explicitly encode the syntactic and semantic relationships between constituents) (Nida, 1982). Furthermore, the profound cultural disparities between Eastern and Western traditions contribute to a degree of untranslatability in cross-linguistic equivalences. For instance, the Chinese swear word “杀千刀的!” (one who deserves to be killed a thousand cuts) employs an extremely violent image—derived from the historical punishment of lingchi (death by a thousand cuts)—to convey the speaker’s rage. This exemplifies semantic bleaching through hyperbolic parataxis, where the literal meaning becomes functionally void and serves to intensify the emotional expression. When such culture-bound swear words are translated into English, the hypotactic nature of English necessitates explicit syntactic supplementation. The proposed translation—“You damned bastard!”—adheres to the grammatical convention of English curses (adjective + noun structure) while diluting the original violent imagery through the religious lexicalisation of “damned”. Crucially, this renders the punitive cultural connotations embedded in the Chinese phrase—rooted in imperial penal history—entirely absent in the target language.

As Pedro (2018) stated that translation is inherently challenging, rendering swear words proves particularly thorny because of the translator’s embedded social attributes. In addition to grappling with the conventional difficulties attendant to any textual translation, the translator must negotiate additional sociolinguistic constraints (e.g., social context, cultural norms, and ideological frameworks) that uniquely govern the expression of swear words. It can be concluded that the examination of translators’ behaviours in the realm of swearing translation is not only a linguistic but also a sociocultural endeavour. Therefore, an interdisciplinary approach that integrates corpus analysis, sociolinguistics, and cultural studies is needed to provide a holistic understanding of this complex phenomenon. The following research questions are proposed to address the boundaries of translation, with the goal of showing how swear words can be translated into the target language without neglecting the cultural identities and sensitivities they carry.

  1. (1)

    What patterns and traces of translator behaviour can be identified in the translation of swear words?

  2. (2)

    What are the underlying causes for the differential manifestations of translator behaviour in rendering various categories of swear words?

  3. (3)

    What roles do textual, sociocultural, and ideological contexts play in the translator’s decision-making process when rendering swear words?

Literature review

The translation of swear words presents greater challenges for translators than does the translation of ordinary language because it requires the conversion of source texts containing vulgar, colloquial, or offensive expressions along with the preservation of their original stylistic features, emotional impact, and sociocultural connotations. Translators must consider not only the literal meaning of lexical items but also their contextual implications, social effects, and underlying ideological dimensions within specific cultural contexts.

Research on the translation of swear words appears to have received insufficient attention in translation studies, largely because of its status as a “low-register” linguistic phenomenon (Gomez, 2016); however, research on the translation of swear words has garnered scholarly attention from both Eastern and Western academics. Feng (2002) reported that swear words are frequently associated with sex or sexual organs, and that their literal meaning often diverges from their actual communicative function. They function primarily as a vehicle for expressing the speaker’s psychological states—such as anger, defiance, contempt, or insult—and therefore require particular attention during the rendering process. Zhou (2022), leveraging a self-compiled Parallel Corpus of the Chinese Folk Language translated by Howard Goldblatt, applied the translator behaviour evaluation model to analyse translational practices in rendering Chinese rural dialects. This study revealed Goldblatt’s distinct domestication-oriented tendencies in the translation of swear words. Wang et al. (2024) investigated swear words in Julia Lovell’s translation of Xi You Ji (Journey to the West) through a corpus-based approach and constructed a critical framework for evaluating translator behaviour in terms of literary swear words. While this study is highly insightful, its analysis is constrained by the examination of relatively mild swear words, which limits the in-depth exploration of the translator’s behaviour; furthermore, its oversimplified definition of swear words and classification criteria exhibiting significant subjectivity represent the methodological limitations that our research must deliberately and consciously seek to avoid.

The study of swear words and their translation has also attracted considerable attention in English-language academia. The Journal of Pragmatics has long dedicated a special issue (Volume 224) to research on the translation of European swear words. These articles systematically reviewed two decades of scholarship in the field and examined such language pairs as English–Danish, English–Greek, English–Italian, and English–Spanish translations. The focus was on subtitle translation for audiovisual media, with only two articles addressing literary fiction and theatrical texts. It proved that “translation of swear words, a subarea of research within translation studies that has gradually attracted the interest of researchers and practitioners” (Valdeόn, 2024, p. 1).

Contemporary research on swear word translation predominantly evolves within two disciplinary domains: (1) the translation of swear words in literary texts and (2) the translation of swear words in audiovisual media and social media platforms. Notably, the latter category incorporates substantial corpus-based studies on swear word translation.

In research on literary swear word translation, many scholarly articles have systematically examined the rendering of profanities in novelistic texts. Greenall’s (2024) analysis of Norwegian crime fiction translations demonstrates that most swear words of the source language undergo either recontextualization or direct transfer in English translations. These strategic choices significantly alter readers’ experiential engagement with literary swear words, consequently impacting character portrayal. The same is true with the strategic application of recasting and borrowing techniques, consequently reshaping character ethos through modified face-threatening act perceptions. Fägersten’s (2017) sociolinguistic examination of English swear words in Swedish comics demonstrates how code switching to L2 profanity generates humour capital through deliberate nonalignment with native English discourse norms. This study shows that speakers consciously adopt Swedish–English interlanguage patterns; thus, shared cultural memory and communities of practice amplify the incongruity humour effect. Parallel investigations (Flannery, 2016; Judiah et al., 2024; Karavelos and Maria, 2024) further validate these observations across diverse linguistic and cultural contexts.

Research on the translation of swear words in film subtitles and social media has yielded a substantial body of literature, predominantly comprising studies that integrate both qualitative and quantitative methodologies. The mixed-methods approach (combining quantitative and qualitative analysis) has been employed to show a central concern with the strategic handling of culturally embedded swear words during cross-cultural mediation, demonstrating how translation choices are shaped by cultural, ideological, pragmatic, and ethical factors and critically impact textual meaning, face management, and audience reception (Constantinou, 2024; Olimat et al., 2023; Abdelaal and Sarhani, 2021). They contribute empirically to the understanding of translation strategies for sensitive linguistic material within specific media contexts. Additionally, numerous scholars have explored the translation of swear words in audiovisual and social media contexts using corpus-based approaches. Sahari (2023) adopted a corpus-driven methodology to analyse the Arabic subtitling of profane/taboo language in 75 Hollywood films and reported that only approximately 5% of swear words were retained in Arabic subtitles, whereas 95% were “toned down, euphemised, or omitted” during the translation process. This study subsequently identified seven predominant strategies for the translation of swear words (metaphorical transfer, preservation, implication, metonymy, semantic misrepresentation, and widening). On the basis of a corpus of Netflix subtitles and their Arabic translations, Abu-Rayyash et al. (2023) examined 1564 English obscenities, revealing three core translation strategies: omission (predominantly used), softening, and conversion to non-swear expressions. While their conclusions differ somewhat from those of Sahari (2023), they are broadly comparable. Through mixed-methods analysis, Pavesi and Formentelli (2023) reported that Italian translators consistently favour cross-linguistically viable models when navigating the relationship between structural–functional patterns and pragmatic functions. These models accentuate the highly emotive and abusive nature of swear words, thereby proposing a principled translation solution that amplifies the inherent conflict of films. Pamungkas et al. (2023) developed a corpus of English swear words on Twitter and constructed a predictive model for swear word detection.

English academics favour empirical approaches that utilise corpus-based techniques to conduct quantitative analyses of swear word translation. With a predominant focus on audiovisual subtitling and social media discourse, research on the translation of swear words in literary texts remains limited. Research on swear words in Chinese academia is relatively rare, and some corpus-based studies are coupled with a notable deficiency in the examination of translator behaviour. Both traditions exhibit a marked insufficiency in the study of translator behaviour as manifested in swear word translation—a gap that offers methodological and perspectival reference points for the present study.

Research on translator behaviour

Emerging in the 1950s, research on translator behaviour initially focused on issues of fidelity and equivalence in translation and has progressively evolved from examining translation norms and processes towards sociocultural dimensions. Translator behaviour analysis grounded in empirical translation process research and cognitive processing (Wilss, 1996), and is inherently framed by the dual attributes of translation: the intralinguistic dimension governing the translation process itself and the extralinguistic dimension shaping its sociocultural context (Hermans, 1991). Baker (2006) argued that translator behaviour influences narrative construction in target cultures, and selective translation may reshape or challenge existing narrative frameworks. Munday (2012) contends that translators’ behaviour is dually constrained by the dominant ideology of the target culture and the translators’ personal ideologies, and it extends to complex forms, norms, and belief systems governing translational conduct. Although ancient Chinese scholars never developed systematic theories of translator behaviour, translation practices and historical documents richly document translators’ roles and impacts—from technical descriptors such as “tongue people” (舌人 in Chinese) and “interpreters” (通事 in Chinese) to the paradoxical status of the Buddhist translators who had to obey the royal requirement (Li, 2012). Recent years have witnessed significant theoretical and practical advancements in Chinese scholars’ exploration of translator behaviour.

Translator behaviour research constitutes a sociocultural examination of both linguistic translation acts and extralinguistic social behaviours, encompassing translator criticism, textual (intra- and extra-) analysis, and contextual factors (Zhou, 2014a). Researchers can identify and analyse patterns in translator behaviour by examining the “overt or covert intentional traces” present within texts (Zhou, 2023a, p. 17). Zhou (2014a) proposed a complementary model based on “truth-seeking” and “utility-attaining” principles. The former focuses on achieving communicative equivalence through the comprehensive or partial reconstruction of source text meaning for target readers, whereas the latter involves strategic adaptations applied to fulfil contextual requirements (e.g., socially conditioned modifications). A utility-attaining illustration is exemplified in Gu’s annotation of the historical figure “管仲” (Guan Zhong) from The Analects, rendered as “the Bismarck of ancient China” (Gu, 2012, p. 227). This representational strategy deviates from literal accuracy (truth seeking) but exemplifies the pragmatic adaptation (utility-attaining) that is essential for cross-cultural comprehension. Such contexts may prioritise target-reader accessibility over etymological precision. Building on this observed strategy, researchers can anchor their study to “translator behaviour theory” and adopt “truth-seeking” and “utility-attaining” evaluative models as an analytical framework. This necessitates a dual analytical approach: examining translators’ decision-making processes at the linguistic level while simultaneously investigating the social dimensions of their actions and thereby achieving a comprehensive characterisation of translator behaviour.

Definition

Prior to empirical examination, a three-dimensional operationalisation of key terminology is needed, encompassing the functional orientation of swear words, their taxonomic classification, and the conceptual framework of translator behaviour.

Functional orientation of swear words

Drawing upon Jakobson’s (1987) seminal framework of verbal communication, which posits that any speech act comprises six constitutive factors correlating with six distinct functions, the “referential” function is identified as the orientational dimension of language. Based on the referential characteristics of swear words (RSW), we divide swear words into two primary types: non-addressee (e.g., self-directed expletives) and with-addressee swear words. The latter type is further classified according to the degree of interpersonal conflict between interlocutors, namely high-conflict and low-conflict, while non-addressee instances are categorised as zero-conflict expressions.

Classification of swear words

With respect to the classification of swear words, Leach (1989) divided the obscenity of language into three categories: (1) dirty words, referring to sex and excretion; (2) blasphemy and profanity, including words such as “damn”; and (3) animal abuse, equating humans with other species of animals. Wen (1998) classifies swear words into sexual terms, pejoratives, direct accusations, and curses, whereas Sun (2005) systematically analyses swear words in three dimensions—the speaker’s profile, target characteristics, and rhetorical techniques. Allan and Burridge (2009, pp. 365–373) categorise swear words into four types: (1) “the expletive function”, which refers to the use of a swear word to let off emotive reaction to frustration, something unexpected or in anger (such as “ouch!”, “shit!”); (2) “abuse and insult”, which includes swearing, name-calling, any derogatory comments directed at another person to insult or hurt him or her (such as “stupid pig”); (3) “expression of social solidarity”, which is a sign of in-group solidarity, especially when directed at outgroups (such as flyting); and (4) “stylistic choice”, which makes use of bad language to spice up what is said or to make it more vivid and compelling (such as “arsehole”), and these functions always overlap. Pedro (2018) proposes a binary taxonomy distinguishing lexicalised swear words (taboo words developed by tacit consensus within speech communities) from contextual swear words (words that acquire offensive connotations only in specific situations while remaining neutral in others). By analysing the novel Aib dan Nasib, Judiah et al. (2024) identify nine thematic categories (excretion, death, religion, bodily functions, sexuality, animal references, etc.), arguing that these serve pragmatic functions, including attention-getting, denigration, provocation, social bonding, and emotional release.

Building upon established frameworks for swear word classification (SWC), we systematically categorise swear words in Stories of China at War into five distinct typologies derived from their semantic-pragmatic features: (1) expulsive swear words, which are words with expulsive connotations and limited in number (such as “fuck off”); (2) abusive swear words, which reflect a form of verbal behaviour that involves scolding another person with vicious words and usually contain offensive, insulting and demeaning words designed to psychologically harm or humiliate the target; (3) derogatory swear words, which refer to words used to demean the target of speech, reducing it to an animal or other object; (4) bantering swear words, which reflect a reciprocal social practice in which one party holds customary sanction—at times a requirement—to engage in jesting or ridicule, while the other party bears a corresponding obligation to interpret such speech as non-offensive (Radcliffe-Brown, 1940), such as “小鬼” (naughty kids) in Chinese, which means a naughty child in English; and (5) expletive swear words, a category of profanity primarily deployed for emotional expression rather than for targeting other individuals (Andersson and Trudgill, 1992) (such as “hell” or “shit!”).

Conceptual framework of translator behaviour

In terms of translator behaviour (TB), the “truth-seeking” and “utility-attaining” perspectives exhibit a dialectical interdependence, and “they are conceptually distinct yet operate on a continuum”, wherein the translator’s choice between truth-seeking or utility-attaining strategies ultimately reflects the degree of translational socialisation and behavioural tendencies in the translation process (Zhou 2014a, pp. 76–77). This study applies Zhou’s “truth-seeking” and “utility-attaining” evaluative paradigm to analyse translator behaviour in Stories of China at War.

We categorise the translator’s translation strategy (TSSW) into four primary types: (1) equivalence (direct transfer of source-text meaning with minimal adaptation, e.g., literal translation); (2) euphemism (such words function as lexico–pragmatic substitutes for socially dispreferred expressions and serve critical face-management functions. Their deployment mitigates potential face threats—whether to the speaker’s positive face or to the hearer’s/third party’s negative face through communicative offence avoidance.) (Allan and Burridge, 2009); (3) omission (lexical-level removal due to untranslatability or taboo constraints); and (4) substitution (a method under the target-language-oriented domestication strategy, which is widely employed in translations across genres and types, particularly in cultural translation, e.g., culture-specific epithet swaps.) (Zhang and Sun, 2018). Notably, omission and substitution can be further subdivided into contextual omission vs semantic omission, and contextual substitution vs semantic substitution, respectively. While semantic omission and semantic substitution are self-evident phenomena, we briefly explain contextual omission and contextual substitution. Cappelen and Dever (2019) introduce the concept of “context control”, suggesting that interlocutors can determine referential meaning on the basis of contextual cues. Berman (2014) proposes four analytical perspectives for examining context: sociocultural, organisational, communicational and textual. Naibaho (2022) categorised meaning in translation into five types: lexical, grammatical, contextual, discourse and situational, and noted that lexical and grammatical meanings operate within microlinguistic contexts, whereas contextual, utterance, and situational meanings function across both micro- and macrolinguistic contexts. Therefore, contextual omission in the present study encompasses two typological manifestations: (1) at the macrodiscursive level, translators deliberately eliminate redundant or verbose expressions to accelerate narrative pacing and enhance target-text readability, preferentially preserving dialogic exchanges and pivotal narrative elements; and (2) situational lexical pruning, involving the omission of speaker-specific swear words on the basis of contextual appropriateness. In contrast, contextual substitution denotes obligatory adjustments that are necessitated by cross-linguistic pragmatic or syntactic constraints. A representative case involves replacing the source-text swear word “鬼子” (devil) with neutral third-person pronouns (e.g., “they”, “he”) to maintain target-text coherence.

To clarify the research trajectory, we constructed an analytical framework for translator behaviour in swear word translation (AFTBSWT) (Fig. 1).

Fig. 1: The analytical framework for the translator's behaviour in swear word translation presents a systematic classification of translator decision-making in the context of swear word translation.
Fig. 1: The analytical framework for the translator's behaviour in swear word translation presents a systematic classification of translator decision-making in the context of swear word translation.The alternative text for this image may have been generated using AI.
Full size image

Beginning at the left, the Speaker is categorized by the presence or absence of an addressee, leading to zero-conflict (no addressee) or conflict-varying situations (high-conflict or low-conflict, with addressee). Swear word classification (SWC) encompasses specific types such as abusive, bantering, derogatory, expulsive, and expletive. Translational behaviour (TB) is divided into two overarching considerations: truth-seeking and utility-attaining, which inform the selection of specific strategies for translation of swear words (TSSW)—including equivalence, omission (contextual or semantic), euphemism, and substitution (contextual or semantic). These strategies generate the target text (TT) through the interaction between the source text (ST) and translation objectives. RSW referential characteristics of swear words, ST+TT interaction between source and target text.

Data

Regarding the research data, Chinese wartime literature of the 1940s, characterised by its pronounced national consciousness and social critique, constitutes a significant component of global anti-fascist literature. Chi-Chen Wang, recognised as one of the pioneers in American East Asia studies (Pritchard, 1963), is hailed by later scholarship as the earliest torchbearer who brought modern Chinese fiction writers to the American readership (Wang, 2007). His translated anthology Stories of China at War (Wang 1947), featuring works by Lu Xun, Mao Dun, Zhang Tianyi, and others, serves as a critical lens for Western audiences to comprehend wartime Chinese society. This translation garnered substantial acclaim in the English-speaking world, receiving a one-star commendation in the 1945 bestseller selection by the American Association for Asian Studies (Cameron, 1945).

Taking the Anti-Japanese War as the backdrop, Stories of China at War adopts a populist and accessible style in both form and content, with a diverse selection of narratives. Its characters span all strata of Chinese society—intellectuals, peasants, guerrilla fighters, soldiers, and others—collectively portraying a multifaceted resistance movement. These short stories interweave the destinies of ordinary individuals with the national cause, vividly articulating “the essence of personal value and the meaning of life” (Zhu and Wang, 2009, p. 112). Owing to the generally low literacy rates in pre-1949 China and the brutalising effects of war, people often developed, consciously or unconsciously, uncivilised instincts and habits, resulting in the frequent use of swear words. Such linguistic behaviour is a common feature of war literature that serves to express emotional catharsis and the horrors of conflict. Consequently, Wang’s translation exhibits a notably high frequency of swear words, e.g., “鬼子” (devils) and “妈的” (shit), whose cultural specificity and emotionally violent nature pose dual challenges in cross-cultural translation: semantic loss and ethical controversy. This makes the text an invaluable resource for studying the translation of Chinese wartime swear words. Wang translated 9 of the 16 stories in this translation, which we use as the primary data, with the translations by other translators serving as references. This study examines Wang’s strategies for rendering swear words and the mechanisms of cross-cultural mediation for linguistic violence in wartime literature translation.

Design

While qualitative analysis undoubtedly facilitates in-depth textual examination, its inherent limitations, including tendencies toward subjective interpretation and incomplete coverage, can be effectively counterbalanced through quantitative methodologies. As a quantitative paradigm, corpus-based translation studies provide robust empirical tools for investigating swear word translation via corpus linguistics, ensuring methodological rigour and analytical comprehensiveness. Farahani and Kazemian (2021) constructed a contrastive parallel corpus to examine how metadiscourse features are used and distributed across English discourse and its Persian translations. By analysing translator–audience interactions within translated texts, they identified four types of translational shifts: (dis)information change, (in)visibility change and (de)emphasis change. Their framework offers a useful point of reference for contrastive and corpus-based translation studies. The mixed-methods approach combining quantitative and qualitative research presents numerous advantages: On the one hand, it empowers researchers to concurrently gather both quantitative and qualitative data within a single data collection phase, thereby leveraging the strengths of both data types for the study; on the other hand, by utilising these distinct methodologies, researchers are able to derive insights from diverse data types and various hierarchical levels within the research context (Creswell and Creswell, 2018). The quantitative component primarily employs corpus-based methods to establish a Chinese‒English parallel corpus of swear words. Specifically, the nine articles translated by Chi-Chen Wang in Stories of China at War were digitised into editable textual data. After rigorous proofreading, the Tmxmall online alignment editor was utilised to achieve sentence-level alignment between source texts and translations, with ParaConc serving as the alignment retrieval tool and AntConc 4.2.0 for statistical analysis (this software is freely accessible via online download and can be deployed for corpus-based analysis; when provided with annotated TXT-format documents, it can perform quantitative computations, including word frequency, keywords and keyness). Subsequent quantitative analysis was conducted on the basis of retrieval and statistical outcomes. We presented the original swear words with their frequencies alongside their translated counterparts and corresponding frequencies in Table 1.

Table 1 Classification and frequency of swear words in Stories of China at War.
Full size table

For qualitative data analysis, MAXQDA software was used. This tool facilitates text coding and categorisation, enabling the creation of new variables through variable management modes. It supports in-depth thematic analysis and quantitative thematic evaluation and provides visualisation tools (Word Clouds, MAXMaps) to display thematic progression, key terminologies, and conceptual relationships (Zhang and Hsu, 2019). We imported the sentence-aligned documents mentioned above into MAXQDA and subsequently performed systematic coding based on a four-tiered framework with 16 subcategories, shown in Fig. 1, for swear word analysis. The predefined hierarchical classification was then applied to annotate the textual data, culminating in the visualisation of the coded content (Fig. 2).

Fig. 2
Fig. 2The alternative text for this image may have been generated using AI.
Full size image

Interface of coding by MAXQDA.

A colleague specialising in translation studies was invited to perform parallel coding to ensure reliability. Upon completion, interrater reliability was calculated using Pearson’s correlation coefficient to compare the two sets of coded data regarding translation strategies and translator behaviours, thereby mitigating subjective bias and enhancing objectivity. For the 127 swear words coded, the Pearson correlation coefficient reached 0.862 (which is within the 0.70–0.99 range), indicating high intercoder agreement and statistical reliability without significant divergence (Table 2).

Table 2 Statistical correlation coefficients between researcher and participant.
Full size table

Results and discussion

Statistical analysis reveals that the number of tokens of Chi-Chen Wang’s translation in Stories of China at War is 28,674, with approximately 0.3% comprising swear words, whereas the original text contains 75,185 tokens with 0.43% swear words. These findings align closely with Jay’s (2018) statistical results, indicating that both the original text and its translation in Stories of China at War exhibit proportions of swear word usage that align with their occurrence in daily communication. This establishes the corpus as a representative subject for research.

MAXQDA enables the coding, analysis, and visualisation of collected corpus data. On the basis of the statistical results, the translator behaviour manifested in the rendering of swear words in Stories of China at War can be graphically represented (Fig. 3).

Fig. 3
Fig. 3The alternative text for this image may have been generated using AI.
Full size image

Visualisation of the translator behaviour regarding swear words in Stories of China at War.

Among the 127 swear words identified in the source text, 106 instances reflect utility-attaining translator behaviour, whereas 21 exhibit a truth-seeking tendency (Fig. 3). Wang’s translation of swear words thus predominantly adopts a utility-attaining approach supplemented by a truth-seeking trend. This statistical finding notably deviates from traditional translation theories emphasising “fidelity” or “equivalence”, suggesting that swear word translation diverges from the rendering of other linguistic and cultural elements. Translators evidently employ distinct considerations when handling such marked language. The corpus of this study comprises Chinese literary works of wartime and their translations. The source text contains numerous derogatory swear words, such as “鬼子” (devils), “东洋乌龟” (Eastern sea turtles), “乌龟养的” (raised by a turtle), and “强盗” (bandits), predominantly uttered by farmers, soldiers, and guerrilla fighters. The translator demonstrably acknowledged speaker identity considerations and endeavoured to speak out for the just war while maintaining formal equivalence at the lexical level. All 21 statistically documented swear words aligning with source-text referentiality in this study exemplify this category. Crucially, such semantic correspondence to the source text intrinsically constitutes “truth seeking” (Zhou, 2023b)—a paradigm epitomising Wang’s truth-seeking consciousness during translation. This approach strategically enables target readers to apprehend the Chinese nation’s collective indignation against Japanese aggression and the resolute determination to enact collective resistance.

Additionally, based on the coded corpus data, MAXQDA can present the overall characteristics of the translator’s behaviour with respect to swear words in Stories of China at War through visualised analytical diagrams (Fig. 4).

Fig. 4
Fig. 4The alternative text for this image may have been generated using AI.
Full size image

Overall characteristics of translator behaviour in Stories of China at War.

The data in Fig. 4 reveal that in Stories of China at War, swear words with addressees constitute the largest category (112), accompanied by a high frequency of low-conflict (95) and derogatory (94) swear words. The number of expletive swear words is 17, while swear words with no addressees are the least frequent (15), with corresponding translation strategies predominantly involving semantic omission (16). This finding demonstrates Wang’s systematic omission of expletive swear words, prioritising utility-attaining effects over truth-seeking at the semantic level. However, this does not indicate a complete separation between truth-seeking and utility-attaining, as “utility-attaining emerges when truth-seeking proves impossible or unnecessary”, and these two principles exist on a continuum, allowing translations to gravitate towards either end (Zhou, 2014b, p. 29).

With respect to substitution and omission strategies, semantic substitution dominates Wang’s approach (26), followed by contextual substitution (17). Omissions primarily occur at the contextual level, reflecting Wang’s distinctive translation style. The original author primarily adopted a vernacular and popularised writing style to accommodate the reading habits and preferences of general Chinese readers. In contrast, Wang’s translation specifically targeted a university-educated readership; therefore, he strategically omitted numerous lexical items and phrases from the source text while preserving the author’s essential communicative intent (Huang, 2022). As for euphemistic translation strategies, three instances of swear words underwent linguistic purification. Therefore, these visualised analytical diagrams of swear word translation patterns in Stories of China at War demonstrate the complexity of translation.

Equivalence

Literary texts generally exhibit a predominant tendency towards “truth seeking” (Zhou, 2014a), which is primarily manifested through “natural equivalence between source and target texts in both linguistic form and semantic meaning” (ibid., p. 225). In this study, equivalence in swear word translation refers to the fundamental correspondence at the lexical level, where translators employ target-language terms that directly mirror the original expressions. Wartime fiction employs grassroots narratives to expose social injustices, with swear words serving as direct linguistic instruments for articulating (1) class antagonisms (e.g., peasants versus landlords) and (2) national conflicts (e.g., civilians versus Japanese aggressors).

Example 1

ST: “再纠合五个人来, 那么再下一回起码可以杀……那些乌龟养的。”

TT: Hsiao Hei-tzu said, “We’ll get five men to join us. We shall then kill…..the sons of turtles!”

In wartime fiction, swear words are not only tools for emotional venting but also symbols of an ideological struggle. Translator behaviour theory emphasises that translation is a dynamic process in which the translator navigates between linguistic “truth seeking” and social “utility-attaining”. Japanese invaders’ atrocities and aggression provoked intense anger among the Chinese, which makes it unrealistic to expect polite forms of address towards them. In the original text, the speaker, a member of China’s lower class, naturally employs the swear word “乌龟养的” (raised by turtles) to refer to Japanese soldiers. There is no overt conflict between the speaker and the listener. Rather, they share a common target of condemnation, placing this utterance in a low-conflict context. Wang’s adoption of an equivalent strategy, translating the phrase “sons of turtles”, degrades the subject to an animalistic level, reflecting his truth-seeking approach in translation. This choice preserves the cultural connotations and emotional intensity of the original, simultaneously reinforcing the boundary between “us” and “the enemy” while reproducing the raw, coarse quality of vernacular language among the oppressed.

Example 2

ST: “滚蛋, 八嘎牙鹿!”

TT: “Get out! Baka desu!” the blackcap cursed, affecting the idiom of his masters.

The phrase “滚蛋” (get out) functions as an expulsive swear word, whereas “八嘎牙鹿” (Baka desu) belongs to the category of abusive swear words. Within the original context, both are high-conflict swear words with addressees. Wang did not adapt these instances of linguistic violence to align with the target readers’ ethical norms; instead, he employed a literal translation strategy to achieve equivalence. The original term “八嘎牙鹿” is a transliteration of the Japanese “ばかやろう” (which means idiot in English). In Wang’s translation, “Baka” derives from the Japanese “ばか” (meaning “fool” in English), whereas “desu” is an English rendering of the Japanese copula “やろう” (meaning “crude person” in English). This approach reflects a truth-seeking tendency in translator behaviour.

In addition to literal translation, Wang also utilised amplification by adding explanatory notes about the speaker’s identity (“汉奸”, literally “traitor”). This strategy serves a dual purpose: it compensates for semantic gaps during language transfer, ensuring contextual coherence, and vividly exposes the speaker’s moral depravity in worshipping aggressors as masters. Thus, Wang’s method strikes a balance between “truth-seeking” (preserving the original’s cultural and emotional force) and “utility-attaining” (enhancing readability and ideological clarity for the target audience).

Substitution

When employing swear words to attack their targets, speakers often linguistically degrade the subjects, reducing them from humans to animals or supernatural beings, and common examples include “王八蛋” (turtle’s egg), “畜生” (beast), and “鬼子” (devils) (Wen, 1998). In Stories of China at War, the most frequent swear word is “鬼子” (devils, 60 times). Other derogatory terms for the Japanese include “乌龟养的” (raised by turtles, 13 times), “东洋乌龟” (Eastern ocean turtles, 9 times), and “乌龟强盗” (turtle bandits, 4 times).

A textual analysis reveals that Wang rarely translated “鬼子” (devils) literally as “devils”, instead opting for substitutes such as “Japs”. The term “鬼子” (devils) originally referred to devil-like creatures, and given the Japanese army’s atrocities during WWII, labelling them “devils” was not unfounded. However, the brutality of Japanese troops solidified a collective Chinese stereotype, stripping “鬼子” (devils) of its literal meaning and transforming it into a metonym for WWII-era Japanese soldiers. However, rendering it “devils” risked oversimplifying national trauma into religiously charged vocabulary, reducing the collective memory of colonial suffering to individualised anger and thereby undermining the construction of power discourse in cross-cultural communication. Consequently, Wang chose “Japs”—an American slur combining “Japanese” and “Apes” (Hornby, 2012)—to amplify the original text’s anti-Japanese sentiment and sociopolitical critique. This strategy enhanced the translation’s acceptability while preserving its scornful and satirical force as wartime literature. In Stories of China at War, Wang also included seven articles translated by other translators. These texts contain relatively few swear words, with the most frequently used being “鬼子” (devils) and “汉奸” (traitors), whose translations largely align with Wang’s approach.

Example 3

ST: “好端端的去杀什么鬼子, 害得别人受苦, 那种人真是……”小黑子故意埋怨着, 声音发抖, 眼睛很不安的望着醉八仙。

TT: “It’s so stupid, whoever did it!” Hsiao Hei-tzu pretended, watching closely Pa-hsien’s reaction. “What’s the good of killing one or two of them? It only brings trouble to everyone. Such people …”

The ST term “鬼子” (devils) is replaced with “one or two of them”—a shift attributable to contextual coherence and sentence-level fluency, and this constitutes a case of contextual substitution. As linguistic fine-tuning constitutes an alignment of expressive modes between languages (Zhou, 2014a), replacing the source text with target-language expressions reduces reading inconvenience for target readers while enabling them to acquire the essential information of the original text with minimal cognitive effort.

Example 4

ST: 是不是城里那帮土鳌?”

TT: Those pigs in the city?”

The Chinese term “土鳖” (hick or country bumpkin) of ST, etymologically referring to a species of insect, has been predominantly adopted in the Northern Chinese dialect as a swear word to mock individuals perceived as rustic or unsophisticated. This word is a derogatory swear word because it degrades humans to inferior life forms. Faced with the absence of a direct lexical equivalent in English, Wang employed strategic substitution by using “pig” to ensure intercultural comprehensibility as well as textual illocutionary force (the effect that speakers intend to achieve through their words).

Omission

Omission is a deliberate strategy whereby translators selectively exclude certain elements from the source text to achieve more natural, fluent, or culturally appropriate expression in the target language. The original purpose of swear word creation is for verbal abuse, yet as linguistic forms detached from the actual act of cursing, they develop their own unique trajectories of evolution and ultimately become meaningless verbal habits (Wen, 1998). For instance, Chinese swear words such as “妈的” (shit) or “他妈的” (fuck) have been transformed into common colloquial expressions or expletive swear words among Chinese speakers. In the source text, variants of “他妈的” (fuck) and “妈的” (shit) appear 17 times collectively, yet Wang systematically employed omission strategies in rendering these swear words, deliberately excluding them from the target text.

Example 5

ST: 石头想这次为了响应发展枪械突击队的号召, 我石头可得粗粉细粉漏两水, 据说挖来枪枝最多的, 还要奖给×旗奖章……他是多么渴慕那个刻着和旗上一样的标记的奖章呵……他妈的, 英勇的记号呢!……妈的, 一股子臭劲……

TT: Shih Tou was preoccupied with his mission. He had volunteered for the firearms extension corps, and he was determined to do his best to win one of those coveted ribbons awarded to those who returned with the most guns.

The use of swear words has a number of motivations, and “one of them is the auto-cathartic ‘letting off steam’” (Allan and Burridge, 2006, p. 2). In the original text, the speaker is a guerrilla fighter. The phrases “妈的” (shit) and “他妈的” (fuck) serve as the protagonist’s habitual or subconscious verbal expressions without specific referents and function as non-addressee swear words. Wang employed semantic omission as a translation strategy when rendering these terms. From the perspective of translation ethics, Wang’s approach violates translational fidelity, as his rendition fails to remain faithful to the source text. However, swear words serve social functions (such as emotional catharsis and identity construction) and contribute to character portrayal. Intellectuals rarely use expletive swear words such as “妈的” (shit) and “他妈的” (fuck), whereas such language frequently appears among the lower-class populace and anti-Japanese guerrilla fighters. Driven by his subjective ideological considerations, Wang omitted these swear words in his translation, aiming to construct a positive and righteous image of the anti-Japanese fighters. In contrast, other translators demonstrate such considerations far less in their practice. Howard Goldblatt, for instance, adopts a markedly different approach when rendering the rustic and swear words in Mo Yan’s works. His translations liberally employ expletives such as “shit” and “fuck” to convey the original text’s vulgarity (Zhou, 2022). Similarly, in Tracks in the Snowy Forest, a Chinese war novel translated by Western scholars, the original text contains numerous swear words such as “他妈的” (fuck), and the translator rarely omits these expressions in the English version, instead rendering them “son of bitches” (Shapiro, 1965).

Notably divergent from his common practice of omitting expletive swear words, Wang consistently rendered—often with literal fidelity and even intentional intensification—abusive swear words uttered by ordinary civilians targeting the Japanese aggressors as the addressee. This strategic choice unequivocally served to construct the populace’s profound abhorrence of the invaders within the target text. Such renditions functioned to viscerally convey the overwhelming voices and outrage of the Chinese people to the target readership, thereby fostering explicit resonance with the perceived justness of the anti-aggression war.

Furthermore, quantitative analysis reveals four instances of the expletive swear word “混蛋” (bastard or scoundrel) within the source text, all uttered by resistance fighters. Wang consistently omitted these specific expletive swear words in his translation. This deliberate omission pattern serves to further corroborate Wang’s distinct utility-attaining translational inclination.

Euphemism

The euphemistic translation behaviour examined in this study constitutes a purificatory translation strategy adopted by translators, which is motivated by subjective ideological inclinations and considerations of target-reader reception. This approach effectively attenuates or even eradicates the swear words present in the source text.

Example 6

ST: 醉八仙…说, “…你瞧, 这班乌龟养的有好日子过! 还不老早给我滚到他东洋妈的屁股眼里去, 如今二十几个乌龟养的住在镇上, 周围十里的人就得把他们当作佛爷似的供奉着, 真像坐地天皇一样, 你说………”

TT: “… those sons of turtles would not have such an easy time of it. They’d long ago have been kicked out of here and back to where they belong. Now with only about twenty of them in the town, they have the surrounding countryside completely cowed. They are treated as if they were gods, heavenly deities on earth. Now, tell me…”

Swear words contain a significant number of terms related to genitalia, sexual acts, or sexual relationships, as well as physiological excretions (Wen, 1998). In terms of utility-attaining considerations for translator behaviour, Wang adopted a de-violencing strategy towards swear words involving sexual violence and bodily humiliation in the source text. This approach systematically neutralised the vulgarity of the source text through metaphorical or abstract expressions and represented a kind of euphemistic translation behaviour. Wang did not entirely omit certain swear words uttered by resistance fighters of the source text. Instead, he substituted them with terms that carry only faint traces of abusive connotations. This approach yields threefold efficacy: (1) It preserves the ethos of the addressor, a resistance hero embodying grassroots patriotism in the source context, through euphemistic reformulation that reinforces positive characterisation. (2) It mitigates intercultural friction by enhancing reader reception and textual accessibility. (3) It maintains linguistic propriety while cultivating a constructive cultural imaginary of the Chinese language among the target readership.

Our examination of Wang’s other translated works reveals that this translational approach persists throughout his career. He consistently employed omission or euphemistic translation for highly offensive swear words involving genitalia, sexual acts, or physiological excretions. For instance, in Contemporary Chinese Stories translated by Wang during the same period, the swear word term “屁眼” (asshole) was rendered “stinking holes” (Wang, 1968, p. 124), significantly mitigating its offensive force while introducing unintended humour.

A more striking example appears in Wang’s Dream of the Red Chamber (Wang 1958, p. 64); herein, the extremely graphic vulgarism “我们肏屁股不肏屁股,管你什么相干” (whether we fuck asses or not is none of your damn business) was neutralised as “what business is it of yours whatever we choose to do?” This contrasts markedly with Hawkes’ (2014, p. 223) rendition, “whether we fuck arseholes or not… what fucking business is it of yours”.

With respect to the frequently used expletive “妈的” (shit), Wang’s translation strategies varied across works. “The genital organs of humans are always subject to some sort of taboo; those of women are usually more strongly tabooed than those of men… because they are source of new human life” (Allan and Burridge, 2006, p. 7). A parallel sociolinguistic phenomenon exists in China wherein a substantial proportion of invectives demonstrate marked gender associations, particularly those linked to female anatomy or kinship roles. The epithet “妈的” (shit, or literally “mother’s”) exemplifies this pattern; when combined with personal pronouns, e.g., “他妈的” (fuck, or literally “his mother’s”), it undergoes semantic intensification to become a highly provocative verbal assault. Regarding the expletive “妈的” (shit) discussed previously, Wang employed the omission strategy exclusively in his translation of Stories of China at War. Across his other translations, Wang predominantly rendered the term through literal translation as “mother’s”, supplementing it with explanatory notes characterising the term as a Chinese colloquialism that is analogous and “is unprintable in Chinese as its translation would be in English” (Wang, 1968, p. 231). These divergent approaches become particularly evident when Wang’s and Edgar Snow’s translations of “离婚” (Divorce) are compared. Wang (ibid., p. 33) rendered “你妈的” (fuck) as “your mother’s”, whereas Snow (1936, p. 87) chose the explicit “You—rape your mother”, clearly demonstrating their contrasting translational orientations.

Translator behaviour embodies dual attributes encompassing linguistic actions and social roles (Zhou, 2014a). In translation practice, whenever a translator deviates from the source text, it invariably manifests the strong tension of purposive factors, indicating the translator’s participation in translation as a social activity. During the 1940s, when China’s war of resistance against Japan was at its peak, the English-speaking world urgently needed to understand wartime Chinese fiction (Hsia, 2011). Wang resolutely resigned from his prestigious position at the Metropolitan Museum to dedicate himself wholeheartedly to translating and promoting Chinese literature during the war. He also worked tirelessly to cultivate American talents well-versed in East Asian culture, in support of the just war (Wang, 2007). This reveals that although Wang could not return to China to join the resistance, he used his pen as a weapon, transforming literary works into forces and arms of resistance. Wang’s utility-attaining behavioural tendency as a translator is precisely manifested in his normative approach to the translation of swear words. In Stories of China at War, Wang’s translations of profanity demonstrate low truth-seeking to the source text but high utility-attaining effectiveness, with considerable rationality in his translational behaviour, and these findings are consistent with Zhou’s (2014b) theoretical framework concerning the balanced relationship among these three dimensions. Wang’s translations aimed to shatter Western classicised imaginations of Chinese literature by presenting modern China’s spirit of resistance. Advocating readability first, he frequently adapted and rewrote source texts to accommodate Western reading habits.

Conclusion

This study investigates the characteristics of translator behaviour through an analysis of swear word translations in wartime fiction. A methodological framework for investigating the translation of swear words is developed, and a combined qualitative and quantitative analysis is conducted to examine the features of the translator’s behaviour in rendering swear words in Stories of China at War. The results reveal that out of 127 instances of swear words in the source texts, 106 reflect a utility-attaining tendency in the translator’s behaviour, whereas 21 exhibit a truth-seeking inclination. Wang’s translation employs four primary strategies—equivalence, substitution, euphemism, and omission—with the strategy selection being significantly influenced by the addressor’s and addressee’s identities. For swear words directed at Japanese invaders, Wang predominantly adopts equivalence and substitution, demonstrating a dual emphasis on truth-seeking and utility-attaining. This choice not only preserves the cultural connotations and emotional intensity of the original text but also reinforces the dichotomy between enemies and allies. Moreover, it retains the raw, vernacular quality of the source language. In contrast, when the addressor is an anti-Japanese civilian and the swear words lack a specific addressee (or include expletives that might undermine character portrayal), Wang frequently opts for omission to sustain the speaker’s positive image. For highly offensive language that could compromise the construction of positive characters, Wang employs euphemisms to purify the original text to a moderate extent. Such adjustments serve multiple purposes: safeguarding the speaker’s reputation, mitigating cross-cultural conflict, enhancing reader acceptability, and nurturing a positive perception of the Chinese language and culture among target readers. Ultimately, Wang’s translation behaviour is significantly moulded by the wartime context and his Chinese identity, reflecting a utility-attaining orientation.

Beyond the specifics of Wang’s practice, this study highlights broader tensions in wartime literary translation. Wang’s strategic choices successfully balance the translatability of swear words and their cultural adaptability. His translations also lay bare a universal dilemma inherent in wartime literary translation: the cross-cultural transfer of linguistic violence risks either diluting its critical essence or reconstructing it as a new cultural symbol. This phenomenon highlights the necessity for translators to navigate a dynamic equilibrium between cultural fidelity and ethical responsibility. While Wang’s four-strategy framework represents a pragmatic compromise between cultural truth-seeking and reader-centred utility-attaining, it also reveals a deeper contradiction in wartime literature translation. Specifically, during cross-cultural transposition, linguistic violence may lose its capacity to convey historical critique if it is overadapted to target cultural norms. Therefore, translators must proactively explore more critical strategies—for instance, preserving the cultural specificity of swear words while constructing an explanatory framework through paratexts such as annotations, prefaces or afterwords. The goal of these strategies is to transform such linguistic items into a cross-cultural medium for transmitting historical memory and social critique. This approach not only honours the source text’s historical and cultural roots but also enables target readers to engage with the ethical and political dimensions of wartime discourse in a nuanced manner.

Limitations and further research

Despite the interdisciplinary approach integrating translation studies, linguistics, and cultural analysis, which offers novel perspectives on swear word translation in wartime fiction, the corpus-based techniques employed here remain limited. These techniques struggle to accurately identify swear words due to the context-dependent and culturally variable nature of such language, where meaning shifts with speaker intent and situational norms. They also fail to support nuanced sentiment analysis that captures subtle emotional valences. These limitations hinder the comprehensive capture of the inherent complexity and linguistic dynamism characteristic of swear words, potentially leading to oversimplified categorisation of translator strategies. Consequently, a systematised theoretical framework is imperative for synthesising insights from multiple disciplinary domains, including sociolinguistics (to understand swear words’ social functions) and ethical translation theory (to guide value-sensitive analysis), providing a more robust foundation for research into the English translation of swear words. Concurrently, it is essential to leverage advanced technologies, including artificial intelligence (AI) for automated context-aware swear word detection and natural language processing (NLP) for fine-grained sentiment mapping. Both of them can help increase the analytical precision and computational efficiency of corpus methodologies. Furthermore, we acknowledge inherent limitations stemming from the circumscribed scope of the corpus and the insufficient examination of translators’ historical and contextual milieu—for example, how wartime political ideologies or target-audience cultural expectations shaped strategy choices. Addressing these constraints leads to critical directions for future research, i.e., expanding the corpus to include diverse wartime texts and translators, integrating contextual factors into analysis, and refining tech-aided tools—all of which can deepen the understanding of swear word translation dynamics.