Chapter 2


In examining the meanings of commonly used swearwords, there are two major methodological issues which need to be addressed: the appropriate form of semantic representation, and what types of evidence can be utilised in justifying such representations. This chapter will seek to answer both of these questions.


The question of how meaning is to be represented is one which has largely been largely (although not entirely) neglected by modern linguistics, making the choice between semantic formalisms a relatively limited one. The approach adopted here will be the Natural Semantic Metalanguage (NSM) approach advocated by Anna Wierzbicka and her colleagues (among the numerous works of this school, Wierzbicka 1972; 1980; 1989a, b; 1991; 1992a and Goddard 1989 all give useful overviews and background).

The basic principle behind this method is the simple fact that words can only be explained by using other words, and for this explanation to avoid circularity, it must be made with words of greater simplicity than those being explained.

Attempting to explain the meanings of a large number of words in this way should ultimately result in a set of simple terms - so-called "semantic primitives" - which cannot be explained themselves, but which can be used to explicate the meanings of all other words. Ideally, such a set of primitives should be readily translatable into any human language, and should function with a minimal combinatorial syntax. The set of primitives thus becomes the foundation of all linguistic meaning, and (presumably) the basis upon which cross-cultural understanding is possible.

The reader may find some of the ideas underlying this methodology questionable. It should therefore be pointed out that the assumptions discussed above have been thoroughly tested by Anna Wierzbicka over a period of twenty years. The originally proposed set of primitives has been revised to take account of translatability and syntactic considerations and tested against a wide variety of semantic and grammatical phenomena in an ever-increasing range of languages. The number of publications in NSM theory (most by Wierzbicka, but also by scholars such as Cliff Goddard, Felix Ameka, David Wilkins, Hilary Chappell and Jean Harkins) is well over one hundred and thirty. It seems safe to say that not only is the methodology well tested and proven, but the specific set of primitives being used is also well justified and productive.

The currently proposed inventory of primitives is as follows (drawn from Goddard & Wierzbicka (to appear)):


I, you, someone, something, no, if, can, because, like, people very


this, the same, other, one, when, where, before, after, two, many all under, above


think, say, know, feel, want do, happen


have parts, kind of good, bad, big, small

While the definitions given here will occasionally make use of terms which are not in the list, such as now or sometimes, they will conform to the principle of only using simpler terms which can ultimately be defined in terms of this set of primitives in the definitions. One example of an NSM-style definition (or explication as they are often known) has already been given for the concept of swearing. Another, Wierzbicka's definition of the "great Australian adjective" bloody, is reproduced below:

(1) bloody

(a) I say: X (where X is a sentence including bloody)

(b) when I say this, I feel something

(c) because of this, I want to do something

(d) I can't say what I feel

(e) because of this, I want to say something that people say is a bad thing to say

(Wierzbicka 1992c)

It is not possible here to explain fully the justification for each of the primitives proposed, the intricacies of the methodology or its philosophical background and implications (see Wierzbicka 1989a, b, 1991, 1992a for discussion of all these aspects). It is worth noting, however, the immediate practical use to which the method can be put. It seems intuitively obvious that any account of the meaning of swearwords will crucially involve the concepts good, bad, very and feel, among others. The explication quoted above shows that this type of analysis can be plausibly applied to swearwords. The final test of the methodology in this context is, of course, the extent to which it can be used to accurately portray the differing meanings of swearwords.

While the NSM approach often gives the appearance of simplicity, this does not mean that the explications here were quickly or mechanically produced. Each has gone through numerous drafts and revisions, cross-checks by other speakers of Australian English and comparisons with the full range of available examples. Virtually every word in every explication has been carefully considered. In this regard, the reader's attention should be drawn to the potential subtlety of distinction in meaning which can be represented through the NSM approach.

This is best illustrated by an example. The following is a list of components drawn from explications to be presented subsequently. Each component is designed to convey an essential element of the meaning of all swearwords, namely the social prohibition attached to their use:

(2a) Other people would say this is a bad thing to say

(2b) People would say this is a very bad thing to say

(2c) People would think this is a bad thing to say

(2d) People could say this is a bad thing to say

Initially, it may seem that all these components (the NSM label for a single, sentence-like element forming part of an explication) convey much the same idea, and that the choice between them is relatively unimportant. In fact, however, this is not the case. In example (2a), the use of "other" which is absent from (2b) and (2c) conveys the idea that within the speaker's own social group, such language is not unacceptable, while outside that group, it is. The distinction between something being a "bad thing to say" in (2a) and a "very bad thing to say" in (2b) allows for the gradation of obscenity in swearwords.

The use of "people would say" in (2b) and "people would think" in (2c) corresponds to the distinction between words which are actively and vocally prohibited (such as cunt) and others which are felt to be obscene but not so often publicly proscribed (such as shit). Note also from (2d) that the choice of "would" as opposed to "could" is also motivated. Presumably, "people could say" that almost any utterance is bad, in an appropriate context (consider, for example, the taboo on the discussion of death among the recently bereaved). The use of "would" captures the idea that all speakers are aware of the social prohibitions attached to swearwords, even if they use them freely and frequently. Quite subtle semantic distinctions in swearing can thus be captured using the proposed set of semantic primitives.

Whether such components can be generalised by extracting their common elements as a means of understanding the core meaning of the concept swearwords is a question which will be addressed in Chapter 6. The reader should, however, immediately be aware that distinctions of this type are of vital importance to the NSM approach, and indeed provide a powerful demonstration of how NSM semantics can convey meaning where a binary-feature decompositional approach would require numerous ad-hoc distinctions without independent linguistic justification.

Some technical conventions should also be noted. In order to distinguish the different senses of a word, a label will be attached in brackets e.g. shit (referential), cunt (insult). The use of an exclamation mark (shit!) will indicate an exclamatory use. The labels are only extremely rough glosses for convenience, and should not be taken as definitive statements of meaning. Within the explications, the use of curly braces {} indicates phonetic content i.e.

I say: {shit!}

means that the speaker actually utters those word(s). The inclusion of this component is not always straightforward; while single word utterances such as shit! can be readily incorporated using this device, swearwords with a wider range of contexts are not so easily accommodated. Dots grouped around a swearword will indicate the use of a swearword within a sentence:

I say: {... fucking ...}

This is, however, not an ideal solution. This problem will be addressed in the discussion of individual explications, and the implications of the existence of a "phonetic component" will be discussed in Chapter 6.


The major aim of this study is to specify the meaning or meanings of three swear words as currently used in Australia, using the methodology of NSM, and to examine the implications such explications have for certain assumptions about swearing. It is not claimed that every possible usage of each word has been explicated, but every attempt has been for a large number of typical Australian uses to be included within the space available. Before beginning, though, it is necessary to justify both the selection of individual words for study and to discuss the types of evidence which will recur throughout the study.

2.2.1 The selected items

The three swearwords to be examined in this study - shit, fuck, and cunt - represent only a small part, although a well-known and well-used one, of the lexicon of swearwords. I have chosen to focus in on these words for a number of reasons:

(a) Each of the items unquestionably represents a swearword in all its possible uses. It seems obvious that the words shit, fuck and cunt are potentially offensive to most speakers even when used strictly for referential purposes. The difficult question of what precisely swearing is and isn't (see Section 1.1) can thus be partially deferred until we have the empirical evidence of actual explications.

(b) The four words chosen appear to represent a hierarchy of offensiveness (see Section 1.2.3). Shit is increasingly acceptable in Australia, as witnessed by its use in various broadcast media; fuck is more restricted in its use, although ABC-TV's more liberal policy towards offensive language (as discussed in Section 1.6) has given it some recent prominence. Cunt is extremely restricted and unlikely to be used in any form of "public speech". It is probable that this hierarchy will correlate with certain components of the explications.

(c) Each of the items included has four letters. Although this may seem a trivial reason, the widespread currency of the expression "four-letter words" suggests that such words are in some respects canonical swearwords.

(d) Each of the words has both a referential and a number of emotive meanings; the extent of the relation between them is obviously an important area to study. To this end, each of the following three chapters will conclude with a discussion of whether referential and emotive meanings can be captured by a single explication, under the heading "A unitary explication?".

2.2.2 The types of evidence

Several sources of evidence are used to justify the proposed explications. An initial source is the (somewhat dispersed) comments on the meanings of the words in question found in the literature. These, however, are rarely comprehensive enough to construct an explication. For this reason, frequent use will be made of examples to indicate the semantic constraints which apply to particular words and constructions.

In application, many of these are of the familiar 'unacceptability' kind often used in syntactic research, where a given structure is shown to be possible or impossible in a given variety of a language. An example is given below:

(2a) Max fucked Betty.

(2b) Max fucked for pleasure.

(2c) *Max fucked Tuesday.

(2d) ?Max fucked.

Sentences (2a) and (2b) shows that fuck (referential)can be used transitively with a human object, and also instransitively with a prepositional phrase, while sentence (2c) shows that it can never be used in reference to an abstract concept. Although this may be trivially obvious to the reader, it demonstrates the principles involved. Example (2d) is more complex; it certainly seems a possible utterance, but it would be more acceptable in some sort of contrastive context:

(2e) Mark made love, but Max fucked.

(2f) Although he claimed to be a virgin, Max fucked occasionally.

(2d) is therefore marked as a questionable sentence. In short, * indicates an unacceptable sentence, ? a dubious sentence, and ?? a very dubious sentence.

While this approach has much to recommend it in terms of delineating elements of meaning, it is open to the criticism that the subjective reactions of an individual can become reified as the linguistic competence of an entire community. Although the use of this type of contrast has long been sanctioned in linguistic research, the problem needs to be addressed when undertaking research into swearing, since the lack of standardisation or even discussion of this topic among individuals (who may have set ideas on the overall acceptability of swearwords which block any awareness of variability in their use) makes it all too likely that the instincts of the researcher may lead them astray as well as guiding them.

For this reason, as many as possible of the original examples in this thesis will not be 'constructed' but will be taken from the speech and writings of others. More specifically, the following sources will be used:

(i) Transcribed examples of swearing noted by the author during the course of research. These will appear in the text as follows, with the gender of the speaker, their approximate age (where known) and an approximate date:

(3) Female, 22: Will you close the fucking door? (3/93)

Where some explanation of context is necessary, it will be given after the example in square brackets:

(4) Male, 18: Fuck! [on ironing board collapsing on his foot] (3/93)

?? for the speaker's age indicates that it is unknown; an age such as "30s" indicates a very approximate guess. Since these examples are only intended to demonstrate the use of a particular construction or combination, they have been written in standard English orthography and no special effort has been made to transcribe prosodic features other than the normal use of punctuation. My own usage has not been recorded in these examples.

(ii) Examples taken from Australian broadcast sources e.g. video and television programs. The same information regarding gender, age and date will be supplied where available.

(iii) Examples taken from printed sources. For the most part these are comprised of magazine articles taken from five main publications, all of which are Australian: Australian Women's Forum, Australian Playgirl, Australian Forum, Hot Metal and Inside Sport. More popular (and therefore perhaps more representative) magazines undoubtedly exist, but their very popularity makes it less likely that they would include swearwords. The more specialised audience and subject matter of the selected magazines means that they make more frequent use of swearwords, but the audience for each magazine is still that of a reasonably widespread public. In the case of the first three, all of which have a heavy focus on sexual matters, referential uses are particularly predominant. Emotive uses are more common in Inside Sport and Hot Metal, a music magazine, but can be found in all these sources. Background information is again provided on the context of the example if it seems relevant.

It should be noted that no attempt has been made to comprehensively encompass differing social groups and age groups in the selection of examples (as would be necessary in a sociolinguistic study, for instance). The author's own circumstances mean that there is a particular bias towards younger speakers and, to a lesser extent, towards males, in the transcribed examples from actual conversation.

Information on gender and age has been provided simply to demonstrate that swearing transcends such social boundaries; most of the swearwords under examination are used by both men and women, old and young. It is the content of the examples that will form evidence, and only to the extent that that content is influenced by the speaker's age and gender (for instance, in determining whether cunt can be used to insult females and males) is such information relevant to semantic structure.

With our semantic principles established, the explanatory work can begin. Chapters 3 through 5 will concern themselves with the meanings associated with shit, fuck and cunt respectively. Chapter 6 will then draw a number of generalisations based on the results of this research, and suggest potential areas for future study.

NEXTClick here to go to the next chapter

MAILClick here to contact Gus

OOPS! Go back up a level

GUSWORLDReturn to Gusworld Central