Chapter 1


The aim of this study is to examine the meanings of three swearwords (shit, fuck and cunt) as they are used in contemporary Australian English. While these words form a significant part of the vocabulary of many Australians, little attention has been paid to them, and virtually no attempt has been made to categorise the different meanings attached to each of these lexemes. The semantics of swearing in general remains a neglected area of linguistic research.

This is not surprising in itself, however: nearly all aspects of swearing have been ignored by contemporary linguistics. As Davis (1989:2) aptly comments, "Very little research has been done on 'bad language' as a linguistic phenomenon, and what has has been largely unsatisfactory" (see also Taylor 1976:43 for similar comments). My aim here is to state precisely the meaning of swearwords, using a rigorous and well-tested semantic methodology.

To introduce this topic, I examine below the literature on swearing in English, with a particular emphasis on discussion of the semantics of swearwords. Given the widespread occurrence of swearing, and its relevance to a number of disciplines, it is perhaps surprising that more information on the topic is not available, but a number of recurring and salient issues can still be discerned which are relevant to semantic concerns.


Many works on swearing note that it appears to be an extremely widespread linguistic phenomenon. The comment that only Japanese, Amerindian and Polynesian languages lack swearwords is a commonplace of literature in this area (Montagu 1967:55; Bryson 1990:214). Implicit in this assumption is the somewhat problematic idea that swearing is a readily definable and universal category of linguistic behaviour.

1.1.1 Commonsense definitions

Of course, the average speaker of English does not worry about what swearing is; to them it is self-evident, as are most of the other cultural categories embodied in the English language. If pressed to define swearing, they might describe it in terms of "bad language" and "four letter words", and possibly present examples of what they consider swearwords and under what circumstances they could be used.

What is surprising is not the existence of such assumptions about swearing among speakers, but the fact that such attitudes are frequently carried over into academic studies of swearing. Numerous studies of swearing assume that what is and is not swearing is intuitively obvious and requires no examination (e.g. Fischer & Nehs 1978:81; Montagu 1967:3). Many other researchers (e.g. Bailey & Timm 1976) categorise swearing solely in terms of the lexical items used.

It is comparatively easy to demonstrate that swearing cannot be simply accepted as a category of linguistic description without qualification. Numerous commentators in the field of ethnography and sociolinguistics have demonstrated that the categories represented in English are not, as has often been claimed, exemplars of universal trends. English emotion terms, for instance, claimed by some to represent universal emotions, do not even begin to approximate the emotion terms found in other cultures (cf. e.g. Goddard 1991; Lutz 1988; Wierzbicka 1986, 1990 (ed.), 1992a: Chs. 3-4, 1992b). Similarly, speech act verbs are not the same across all cultures but can differ markedly (Verschuren 1985; Wierzbicka 1988, 1991: Ch. 5). Given these examples - and the fact that swearing is a type of speech act, and often a reflection of an emotional attitude - the assumption that swearing is a basic human instinct appears to amount to little more than Western cultural imperialism.

A concrete example may make this clearer. Jay (1992) titles his monograph on obscene language Cursing In America because, he claims, cursing is the more basic notion in the relevant area and one that is more commonly used (Jay 1992:1). In Australia, this would not apply: the most commonly used term for this type of behaviour is swearing. A cultural difference is thus indicated even between speakers of the "same language". The existence of such a difference indicates the inherent risks of assuming swearing to be an invariant category.

1.1.2 Lexicographic definitions

The Oxford English Dictionary defines swearing as follows:

To utter a form of oath lightly or irreverently, as a mere intensive, or an expression of anger, vexation or other strong feeling . . . to utter a profane oath, or use profane language habitually; more widely, to use bad language. (OED XVII:367)

It is not difficult to spot flaws in this definition. Replacing the concept of swearing with profanity essentially explains nothing. Even more damningly, the most relevant part of the definition for the ordinary use of the term ("to use bad language") is deferred until its end. A more precise definition than this is obviously needed.

1.1.3 Scholarly definitions

Those scholars who do attempt to define swearing rarely do so in a rigorous way. For example, Crystal (1987:10) defines the function of swear words as "emotional expression", making no attempt to delineate them from other emotive forms in language other than to say that "Swear words and obscenities are probably the commonest signals to be used in this way". This hardly helps to narrow the concept down.

Allan & Burridge (1991), operating within a framework of comparing euphemism and dysphemism, provide a definition for the latter concept which might be seen as relevant to swearing:

A dysphemism is an expression with connotations that are offensive either about the denotatum or to the audience, or both, and it is substituted for a neutral or euphemistic expression for just that reason. (Allan & Burridge 1991:26)

While this definition does attempt to break down the elements of dysphemistic language, the proliferation of "ors" hardly makes it very specific.

Andersson & Trudgill (1992:53) define swearing as (a) refer to something taboo in a given culture, (b) to be interpreted non-literally and (c) used to express strong emotions and attitudes. The main fault with this definition is that it fails to account for the intuition of many speakers that a word such as cunt is a swearword whether it is being used literally or not (cf. Taylor 1976:43).

By far the most useful definition of swearing that I am aware of is that proposed by Anna Wierzbicka in her English Speech Act Verbs: A Semantic Dictionary, utilising the NSM methodology which will also be used in the present study:

(1) swearing

(a) I'm thinking about X

(b) I feel something bad because of that

(c) I want to say something bad because of that

(d) I want to say some words because of that that people think one should not say

(e) I say: [#@!?*!]

(f) I imagine that by saying these words I could cause something bad to happen to something

(g) I imagine that by saying these words I could cause myself to feel better

(h) I say this because I want to show what I feel

(Wierzbicka 1987:252-253)

This definition has several advantages over the others proposed here. It recognises that swearing is a deliberately "bad" act, designed for emotional expression. It explicitly acknowledges the importance of the speaker's attitude and the fact that particular words will be used in such a context. Its inclusion in a dictionary of English speech act verbs also implicitly makes it clear that there is nothing universal about such a definition.

For practical purposes, it is this definition of swearing which will be accepted for the remainder of this study. Nonetheless, some improvements could possibly be made to this definition (particularly component (f)). These will be taken up in Chapter 6. As an added safeguard, therefore, the focus will be on items which can unquestionably be classified as swearwords (cf. Sections 1.2.3 and 2.2.1 below).

1.1.4 Swearing and slang

A brief note needs to be made on the relationship between swearing and slang. The extent to which these categories overlap and differ has been of some concern to scholars of swearing (e.g. Allan 1986:17; DeKlerk 1992:287). Dumas & Lighter (1978), in a useful article, examine difficulties with the term "slang" as used by previous writers and propose that slang must meet four criteria: (1) it must be undignifying, (2) it must demonstrate a familiar or dismissive attitude towards the referent, (3) it must be taboo in higher status conversation and (4) it must be used rather than a conventional synonym. If we accept these criteria, swearwords may or may not qualify as slang. While swearwords meet the first three criteria, whether they are used in place of a conventional synonym is questionable. For instance, although the sentences He fucked her and He made love to her could feasibly describe similar situations, the background assumptions behind each are so different that to describe them as synonyms seems inaccurate in the extreme (this issue is taken up in Chapter 4).

For the purposes of this study, it will be assumed that swearing by and large does not fall into the category of slang. However, certain highly idiomatic uses of swearwords may function as slang for certain speakers. In the end, insofar as meaning is concerned, the distinction is unimportant because the meaning of a term will need to be explained regardless of what general descriptive label is attached to it.


There is a frequent assumption in scholarly works that the meaning of swearwords is not a matter for semantics at all, but falls into the pragmatic or social arena. Jay (1992), for instance, restricts his semantic discussion to differences between male and female usages. Jay is also more concerned with swearing as an act than with the function or meaning of individual swearwords.

I am not aware of any formal definitions proposed for individual swearwords outside of lexicographical works, other than those of the NSM school of semantics (discussed in Chapter 2), and some of the work of Allan & Burridge (1990) (discussed in Chapter 4). Taylor (1975), discussing these issues informally, concludes that the referential meanings of many swearwords can be described by componential analysis, a technique which has frequently been shown to be questionable (Kempson 1977; Wierzbicka 1980: Ch. 1). Taylor (1976) discusses the meanings of swearwords in a similarly general fashion in terms of concepts such as 'vehemence' and 'company'. For connotative meanings, Taylor proposes, among others, the following definitions:

(2) Fuck! - exclamation of (a) surprise, (b) disgust, (c) disappointment (Taylor 1975:43)

(3) Shit! - exclamation of (a) surprise, (b) disgust, (c) disappointment (Taylor 1975:43)

(4) cunt - person whom ego disapprove of at x level of vehemence (Taylor 1975:36)

The inadequacies of these definitions are all too obvious. The definitions for Fuck! and Shit! are identical, giving no indication of how the meanings of these words differ. The psychological validity of quantifiable "levels of vehemence" is obviously open to question. And why ego rather than I? I will therefore move on to examine dictionary definitions before turning to examine trends of meaning patterns which have been observed by previous researchers.

1.2.1 Dictionary definitions

The following is an example of a dictionary definition of fuck, taken from that most respected of all dictionaries, the Oxford English Dictionary:

(5) fuck - to copulate, trans.; . . . used profanely in imprecations and exclamations as the coarsest equivalent of damn; . . . Various other casual, intensive etc. uses (OED VI:237)

While the basic meaning of fuck is indeed something like "to copulate", the extended meanings of this word are however barely touched upon. The language learner would discover little from learning that fuck is supposedly a coarse equivalent of damn, and even less from the comment "various other casual, intensive etc. uses".

One might suspect that specialist dictionaries of slang and swearing might be better in this regard. In fact they are as bad or worse. Consider the following definition from Partridge's Dictionary Of Slang And Unconventional English:

(6) fuck my luck! - Oh what a pity! (Partridge 1984:432)

The use of the genteel, even priggish "Oh what a pity!" to define this particular term goes beyond inadequacy to inaccuracy.

It is also worth pointing out that etymological information regarding the origin of swearwords is not used here as a source of data for their meanings. This difficulty partly stems from the fact that swearwords, being primarily a feature of spoken language, are not often documented in a way which allows their development to be traced reliably (see Hughes 1991 and Montagu 1967 for discussions). There seems no good reason to assume that historical background would be of relevance to the multiplicity of contemporary meanings of these words. This becomes particularly apparent if the increasing use of swearwords in the past few decades is taken into account. Contemporary usage must be a more reliable guide than outdated and sketchy historical information.

1.2.2 Referential and emotive meanings

An important distinction in meaning which has been noted by many writers concerns the difference between referential and emotive meanings for most swearwords. Many words in this area have a specific referent but have also been extended to be used solely to express the attitude of the speaker. Dong (1971a) illustrates this point well with the ambiguous sentence Fuck Lyndon Johnson, which can be very roughly glossed as "Have sex with Lyndon Johnson" or "I don't care about Lyndon Johnson" (Dong 1971a:4).

Dong proposes that the referential (sexual) version be called fuck1 and the other fuck2. This distinction can be referred to as denotative vs. connotative or referential vs. emotive. Because the latter pair of terms makes clearer the potential for separate meanings, it will be used throughout this work.

It is also worth commenting briefly on the relationship between swearing and interjections. The classificatory scheme for interjections proposed by Ameka (1992:111) would view swearwords such as Shit! or Fuck! as only secondary interjections, since they also have independent non-interjectory meanings. This appears to largely correspond to the distinction between referential and emotive meanings.

Depending on one's view of meaning as a whole, the status of referential and emotive uses can vary. Jay (1981:30) sees the emotive use as having primacy over referential use, in contradistinction to many others. Staley (1978) goes to the opposite extreme and claims that emotive uses have no "lexical meaning", which only makes sense if it is interpreted as the trivially obvious claim that the meanings in question are not referential. Jakobson (1980:22-23) also claims that swearwords are "zero parts of speech" because the use of them is frequently lost in people whose brain's right hemisphere is inactivated. The fact that this part of the brain also controls emotional reactions (Jakobson 1980:24), which can also be expressed through swearwords, might, however, be an alternative explanation of this fact.

Allan & Burridge (1991) see referential meanings, their primary concern, as constant with regard to any referent. They claim that shit and feces, for instance, mean the same thing (Allan & Burridge 1991:30) but can be used in different contexts. Yet the fact that the following two sentences can be used in an identical context appears to contradict this:

(7a) John analyses feces at work.

(7b) John analyses shit at work.

The most obvious explanation for this is that the two terms have identical referential meanings, but the incorporation of the speaker's assumptions brings about a differentiation in meaning. (See Wierzbicka 1991 for examples of how frequently such distinctions are built into the languages of the world.) This will be the initial assumption regarding referential meanings in this work.

Finally, while this distinction exists in many cases, it is not always clear cut. Consider the following common Australian usage:

(8) Clark beat the shit out of Kent.

While the primary usage of shit here appears to be emotive, expressing the speaker's attitude towards the event, the expression appears to have its origin in the referential meaning of shit (i.e. it is possible that Clark quite literally caused the shit to come out of Kent's bowels through the force of his kicking). While separate meanings will need to be established in such cases, the possibility that (to borrow Dong's terminology once more) fuck1 and fuck2, or any other similar pair, will share elements of meaning as well as form will need to be kept in mind.

1.2.3 Degrees of obscenity

It seems intuitively obvious that not all swearwords can simply be classified, dictionary style, as "vulgar". Some are decidedly more vulgar than others. Writers on swearing have recognised this. Crystal (1987:61) speaks of a continuum extending from mild expletives through to the two maximally taboo words, fuck and cunt". Taylor (1975) has ranked a wide variety of swearwords according to their degree of obscenity on a six-point scale. Empirical testing has demonstrated the salience of such notions for ordinary speakers (Jay 1992: Ch. 5; Baudhuin 1973). In general speakers agree that cunt is a stronger swearword than fuck, fuck than shit, shit than damn and so forth.

While the general principle of a "hierarchy of obscenity" seems clear, it is not so obvious that words can be arranged simply on a mathematical scale. The differences in degree of obscenity would appear to be qualitative as well as quantitative. This is reflected, for example, in the above discussed distinction between fuck1 and fuck2; a recent court case in Australia deemed that the use of fuck1 was obscene but of fuck2 was not, suggesting a difference of degree between the two (Allan & Burridge 1991:132).

Such differences are not reflected in a simple hierarchical model. Taylor's ranking, for example, classifies both fuck and cunt as having a maximal taboo loading (Taylor 1975:43), ignoring the fact that for most Australian speakers cunt is decidedly worse than fuck. The same problem exists for Sewell (1984), who classifies both shit and fuck as "strong profanity" but ignores the possibility of differences between them. (The general failure of semi-mathematical models such as componential analysis in explaining meaning also points to the flaws in such assumptions.) Any adequate explanation of the meaning of swearwords will need to be able to capture such differences in a more precise manner.

Having considered specifically semantic issues, I will now turn to considering other aspects of swearing which have been noted in the literature. This will not be a comprehensive survey; rather, it will seek to highlight findings which may be of relevance to my ultimate aim of defining swearwords.


Relatively little serious study of the interaction of swearing with morphology and syntax have been undertaken (Andersson & Trudgill 1992:61-63). Perhaps the most famous (or infamous) examples are those collected in the James McCawley festschrift (Zwicky et al 1971). A number of articles therein discuss such issues as acceptable syntactic structures for the verb fuck. Entertaining though they sometimes are, the contributions to this work are of limited use to the scholar interested in the semantics of swearing. Firstly the generative syntax analyses they propose have long since been shown to be untenable. Secondly, the examples they present are often so patently unlikely as to have little relevance (cf. Newmeyer 1986:136-137):

(9) Fuck seven old ladies by midnight or I'll take away your teddy-bear. (Dong 1971a:5)

A more serious fault shows up in the following set of examples:

(10) What do you do with your pencil?

(i) I fuck children with it.

(ii) *I screw runaway teenie-boppers with it.

(iii) *I bugger kiddies with it.

(iv) *I ball nuns with it.

(v) *I lay monks with it. (Wang 1971:40)

Quite aside from their unlikelihood, Wang's semantic intuitions fail to match Australian expectations. Examples (ii) and (iii) are both quite acceptable to the author and all the (Australian) informants he asked. It seems obvious once again that American and Australian linguistic intuitions on this subject differ.

The potential for swearwords to take part in "infixing" in English has been the subject of some discussion. McMillan (1980) gives a very full set of examples of this phenomenon, involving the insertion of certain English lexemes into others e.g. fan-fuckin'-tastic. The infixed items are frequently (but not necessarily) swearwords. McCarthy (1982) goes on to develop an account of this phenomenon in terms of a theory of metrical phonology. The issue of "expletive infixation", as it is often called, will be taken up again in Chapter 4.

Taylor (1975:33; cf. also Crystal 1987:61) comments that swearwords show a high degree of combinability with each other. Once one swearword has been introduced into a sentence, others can easily follow; as Allan & Burridge (1991) also note, such terms implicate each other to a high degree:

(11) He's a cunt, a real shit of a guy!

(Allan & Burridge 1991:142)

While this example shows two different swearwords being used, it is also possible for the same swearword to trigger itself repeatedly as well as triggering others, as the following example (taken from actual conversation) shows:

(12) Male, 18: She was a fucking textbook fucking boxer's fucking girlfriend - she looked so fucking dumb! (5/93)

Such sentences have been taken as evidence that the meaning of swearwords is not terribly specific or non-existent (cf. Section 1.2.2 above). The questionable validity of this viewpoint will be one of the concerns of this study.


Although its use is more sanctioned under some circumstances than others, the general attitude to swearing is that it is a socially unacceptable behaviour. Such attitudes are reflected, for instance, in the frequent findings of American psychologists that counsellors who swear are perceived as both less attractive and less effective (Heubusch & Horan 1977; Kottke & McLeod 1989; Paradise, Cohl & Zweig 1980). They are also reflected in various psycholinguistic studies which found that swear words were less easily recognised than non-obscene ones (see Paivio & Begg 1981:133). While there are certain exceptions (e.g. the fact that in America it is an offence to sack workers for swearing during union negotiations, see Graham 1986), swearing remains largely sanctioned behaviour. This is reflected in the frequent use of the term "bad language".

The extent to which this attitude is codified in the actual meanings is a matter of some dispute. In their textbook An Introduction To Language, for instance, Fromkin, Collins, Rodman & Blair (1990:269) claim that there is

no linguistic reason why the word vagina is 'clean' whereas cunt is dirty.

The point that it is only by convention that any particular item becomes classified as "dirty" is well taken. But insofar as the meanings of these words include the notion that the speaker's attitude towards the intended referent is "bad", there is a "linguistic reason why vagina is 'clean' and cunt is 'dirty'". The role of speaker attitudes becomes even more important when these terms are used in a non-referential sense; if there was no linguistic encoding of the notion of badness in the term cunt or vagina, then either could be appropriately extended as a term of personal disparagement, but it is onlycuntthat can be used in this way.

Of course, this very fact of being forbidden is essential to the function of swearing. The emotional release it provides for would be far less likely if it was accepted. Criticism of obscenity from a scholarly point of view thus seems rather churlish and narrow-minded.

Yet one disturbing trend in scholarly writings on swearing is a tendency to abandon academic objectivity and to incorporate the writer's own (inevitably negative) attitude towards swearing. Hughes (1991:187), for example, writes:

'The habit of swearing' and the use of 'foul language' now thrive with positively indecent health. (emphasis mine)

Even more extremely, Allan & Burridge (1991) characterise various social groups using (or not using) swearwords as "certain macho hooligans" (149) and "self-righteous religious bigots" respectively (37). Other scholars (e.g. Paivio & Begg 1981:133) show a remarkable reluctance to reproduce actual swearwords in print. It is hard to believe that any writer with this kind of attitude has really devoted themselves seriously to the study of swearing. The implication appears to be that while swearing is an interesting diversion, it is not of particular scholarly interest, and not worth such in-depth study as other aspects of language.

The current analysis will attempt not to be judgemental in this fashion. The offensiveness of swearwords needs to be recognised if their meanings are to be examined, but it should not be used as an excuse for failing to recognise the systematicity and subtlety of meaning distinctions in this area. In this regard, swearing is as "good" a part of language as any other.


The aspect of swearing which has most often been examined from a scholarly perspective is doubtless that of their use. The predominant area of research so far has been into gender variation in swearing behaviour, but a limited amount of work has been carried out on the contexts of swearing.

1.5.1 Gender variation in swearing

The rise of research into language and gender has meant that widespread folklinguistic beliefs concerning the language use of males and female