Corpus-Based Study of Swearing in Spoken British English

ZHANG Jidong()， LI Shuang( )

School of English Studies， Shanghai International Studies University， Shanghai 201620， China

Abstract： Swearing constitutes an important part of daily language use. However， despite several previous studies with foci on the social and cultural characteristics of swearwords， a large part of it remains to be investigated. With the abundant spoken language resources from British National Corpus(BNC)， this article observes the pragmatic and collocational features of swearwords. Three social variables， namely， social class， speech domain， and gender were put under investigation to see how the use of swearwords varies with different speech social contexts. Meanwhile， the noun and adjective collocates of several frequent swearwords were separately sorted out and discussed. The data observed show that the use of swearwords is highly related to different social contexts， and there exits certain level of differentiation between different swearwords used as intensifiers which can be generalized lexicographically.

Key words： swearing； corpus linguistics； British National Corpus(BNC)； pragmatics； collocation

Introduction

Swearing is the use of utterances containing taboowords[1]. As one of the most important linguistic and sociocultural phenomena， it remained， for a long time， to be a field that is least studied， as suggested by many linguists and sociologists. During recent decades， however， attempts have been made to study swearing from different angles and with different research methods. In swearing， Ljung[1]studied swearing from the sociolinguistic perspective while Mohr[2]conducted an in-depth research of swearing diachronically. One of the problems of the earlier studies of swearing is that most of the observations made by researchers are intuition-based. From the latest researches， we can easily find that there is a trend toward researches based on qualitative data. In Ref.[3], McEnery observed the features of bad language use with the support of large-size corpus. The study conducted by Lutzky and Kehoe[4]illustrated how corpus linguistic approach could be applied effectively to identifying swearing in online communication with reference to both contexts and collocations.

From the perspective of corpus linguistics， this study conducts a comprehensive research on two aspects of swearwords. Firstly， we focus on the pragmatic aspect of swearword， that is， how the use of swearwords is influenced by social variables， namely， the social status and gender of people involved， the domain of the speech. Secondly， the collocations of swearwords will be studied in detail. Just as we shall know a word by the company it keeps， a better understanding of swearword can be gained through the observation of its collocates. By looking at metadata in British National Corpus(BNC)， the study aims at uncovering several characteristics of swearwords and tries， as it can， to inform corpus linguistics into English language and gender research and teaching.

1 Theories Revisited

1.1 Swearing

There is neither a standard definition nor categorization of swearing. Different criteria have been adopted by different linguists based on the focus and purpose of their specific studies， which make it a challenge to straight out the study of swearing in a systematic way. According to Ljung[1]， however， “despite their different views on what swearing is and how it is best described， the studies above all set up certain basic criteria that in their opinion have to be met in order for an utterance to count as swearing”. Therefore， he listed those criteria as follows.

(1) Swearing is the use of utterances containing taboo words.

(2) The taboo words are used with non-literal meaning.

(3) Many utterances that constitute swearing are subject to severe lexical， phrasal and syntactic constraints which suggest that most swearing qualifies as formulaic language.

(4) Swearing is emotive language， its main function is to reflect， or seem to reflect， the speaker’s feelings and attitudes.

The fourth criterion， as a case in point， prescribes the function of swearing， that is， to express feeling and emotion， including positive and negative one. As Mohr[2]suggests， “Swearwords are offensive， they are vulgar， and they can certainly be overused. But they also do what no other English words can. They are the most powerful words we have with which to express extreme emotion， either negative or positive. They insult and offend others(which， like it or not， is a function of language)， they offer catharsis as a response to pain or to powerful feelings； they cement ties among members of groups in ways that other words cannot.”

1.2 Swearing and pragmatics

As one of the most common language behaviors， the study of swearing is closely related to pragmatics. “Interpersonal swearing is a complex communicative act that is influenced by contextual variables such as speaker-listener relationship， social and physical setting， and the topic discussion.”[5]

1.3 Function-based classification of swearwords

Based on different functions of swearwords， Ljung[1]divided swearwords mainly into two categories： stand-alone function and slot fillers. Stand-alone swearword， as its name suggests， functions as an utterance themselves， mainly expressing negative feelings such as anger and pain. In contrast， slot fillers constitute longer strings， in which they serve as adverbial and adjectival intensifiers expressing a high degree of a following gradable noun， adjective and adverb：

(1) You bloody fool！

(2) It sounds fucking awful.

(3) I know damn well that he will lose it.

2 Methodology

This study is based on the spoken texts of BNC， which contains a 100-million-word collection of samples of both written and spoken language from a wide range of resources. In the first half of the research， we mainly focus on the pragmatic features of five swearwords(fuck，shit，damn，bastardandfucking) on the basis of three social variables， namely， the social class of speaker and respondent， the domain of speech and the gender of speaker are studied separately. In the second half of the research， the noun and adjective collocations of three swearwords(fucking，damn， andbloody) are investigated respectively.

3 Results and Analysis

3.1 Swearing and social class

It is a widely-held view that the higher one’s social class is， the less likely he is to use highly offensive swearwords. Previous study conducted by McEnery and Xiao[6]confirmed this proposition in the case study offuckin the BNC. In our study， besides this most frequently-used swearword， another four highly offensive words， namelyshit，damn，bastardandfucking， have also been studied as shown in Table 1.

Table 1 Five swearwords’ distribution among speakers’ social classes in BNC

Note： RF refers to raw frequency； NF means normalized frequency per one million words; AB refers to upper class; C1 refers to upper middle class; C2 refers to lower middle class; DE refers to lower class

From Table 1， we can see that the data observed are in accordance with the generally held view that people of higher social class tend to refrain from using highly offensive swearwords while people from lower class use swearword more frequently.

Interestingly， although it is tempting to believe that people tend to behave more politely and use less highly offensive words when they are talking to someone who enjoys higher social status， this belief contradicts with the data we have observed.

The data in Table 2 shows that respondents from AB are more likely to hear offensive swearwords than people from C1. In the case offuck，shitandbastard， respondents from AB are more likely to hear swearwords than people from C2， ranking the second only next to people from DE.

Table 2 Five swearwords’ distribution among respondents’ social classes in BNC

(Table 2 continued)

Note： RF refers to raw frequency； NF means normalized frequency per one million words

3.2 Swearing and domain

Beebe[7]argued that besides venting feelings， swearwords， or rude words were intentionally used to gain power. This can also be evidenced by the following data which suggest that in occasions where power is needed(especially in the case of doing business)， highly offensive words are more preferably used than they are in other less tense occasions.

The data collected in Table 3 show that in the cases of doing business or educating and informing others， offensive words are chosen at the speakers’ disposal to gain power. In occasions where there is no need to gain power(as in leisure) or where swearwords are forbidden(as in public)， people are unlikely to use those offensive words.

Table 3 Five swearwords’ distribution in BNC’s speech domain

(Table 3 continued)

Note： RF refers to raw frequency； NF means normalized frequency per one million words

3.3 Swearing and gender

The theory of gender difference in language use have been confirmed and enriched by many qualitative data[8-9]). Comparatively speaking， women tend to use more standard language than men. Linguists offer some explanations for it and one of them is status-consciousness. “Women use more standard speech forms than men because they are more status-conscious than men. The claim is that women are more aware of the fact that the way they speak signals their social class background or social status in the community. Standard speech forms are generally associated with high social status， and so， according to this explanation， women use more standard speech forms as a way of claiming such status.”[10]Based on this explanation and the fact that swearing is a typical non-standard language， we may hypothesize that women in all social classes swear less than men.

From Table 4 we can justify our hypothesis that in all social classes， men tend to use more swearwords than women do. Besides the widely accepted explanations that women are more status-conscious than men and are equipped with more psychological taboos than men， Zhao[11]added that women spoke less swearwords in order to protect their own image.

3.4 Swearwords and collocation

Table 4 Five swearwords’ cross tabulation of speakers’ genders and social classes

Note： RF refers to raw frequency； NF means normalized frequency per one million words； LL score(log likelihood value) refers to the probability of the difference between the two corpora happening by chance. If the LL score is greater than 6.630， the probability of the result happening by chance is less than 1%； Sig. level also indicates whether the difference between the two corpora is happening due to chance. If the value is lower than 0.001， we are reasonably certain that the result is not due to chance and thus， meaningful

Table 5 Noun collocations of fucking(span： Right1-Right2)

Note: Freq. means frequency

Table 6 Concordance lines of fucking hell

Table 7 Noun collocations of bloody(span： Right1-Right2)

(Table 7 continued)

Table 8 Concordance lines of bloody hell

Table 9 Noun collocations of damn(span： Right1-Right2)

Table 10 Concordance lines of damn fool

By comparing the above data， we may draw the conclusion that the wordfuckingis very likely to be used to modify and intensify other swearwords such asbitch，wanker，cunt，shit，bastardand so on. Also， while the wordhellcan co-occur withfuckingandbloody， it can not collocate withdamn.

Table 11 Adjective collocations of fucking

Table 12 Adjective collocations of bloody(span： Right1-Right2)

Table 13 Adjective collocations of damn(span： Right1-Right2)

From the data collected above， we find thatfuckingandbloodycan be used as intensifiers for the adjectives with both connotational positive and negative meanings， and the same is not true with the case of the adjective collocations ofdamn.

4 Conclusions

This article is a specific study of swearing from the perspective of corpus linguistics. Basing on the spoken texts of BNC， we have observed the pragmatic use of swearwords by looking at several social variables， namely， social class of speakers and respondents， domain of speech and gender of speakers. In this research， we have at least three findings. (1) While people from higher social classes are less likely to use highly-offensive swearwords， they tend to hear those words more frequently than the native speakers would expect. (2) In such specific occasions as doing business， for example， where power is more preferable with the businessmen， swearwords are used more frequently to demonstrate speakers’ attitude or stance. (3) This research reconfirmed the proposition that in all social classes， gender differences in swearing behavior exists， and male tend to swear more than female. Moreover， basing on Ljung’s categorization of swearwords， we have studied the collocations of three swearwords， namely，fucking，bloodyanddamn， which are often used as intensifiers. By seeing the noun and adjective collocates of those swearwords， we have found that every swearword has its own collocational features， and when using them， we need to be more discernful not to choose the anomalous associates as their collocates. All in all， this study will help uncover some characteristics of swearwords and can shed new light on the value of corpus linguistic research in this field.