Impoliteness revisited: with special reference to dynamic and prosodic aspects

Journal of Pragmatics 35 (2003) 1545–1579 www.elsevier.com/locate/pragma

Impoliteness revisited: with special reference to dynamic and prosodic aspects Jonathan Culpepera,*, Derek Bousfielda, Anne Wichmannb a

Department of Linguistics and Modern English Language, Bowland College, Lancaster University, Bailrigg, Lancaster, Lancs LA1 4YT, UK b Department of Cultural Studies, University of Central Lancashire, Preston, Lancs PR1 2HE, UK Received 11 November 2001; accepted 5 July 2002

Abstract This paper focuses on impoliteness, the use of communicative strategies designed to attack face, and thereby cause social conflict and disharmony. Using television documentary recordings of disputes between traffic wardens and car owners as our data, we revisit the impoliteness framework mapped out in Culpeper [J. Prag. 25 (1996) 349]. Having justified why an impoliteness framework is needed, we explore the notion of impoliteness and consider whether the impoliteness strategies identified in Culpeper can be found in another discourse type. We argue that for impoliteness to be fully appreciated we need to move beyond the single strategy (lexically and grammatically defined) and examine both how impoliteness pans out in extended discourse and the role of prosody in conveying impoliteness. Our paper has important implications for politeness theory and discourse studies in general, and the role of prosody in discourse in particular. # 2002 Elsevier B.V. All rights reserved.

1. Introduction Although, as Leech points out, ‘conflictive illocutions tend, thankfully, to be rather marginal to human linguistic behaviour in normal circumstances’ (Leech, 1983: 105), there are contexts in which conflictive illocutions are rather more central. Conflictive talk has been found to play a role—and often a central one—in, for

* Corresponding author. E-mail address: [email protected] (J. Culpeper). 0378-2166/03/$ - see front matter # 2002 Elsevier B.V. All rights reserved. doi:10.1016/S0378-2166(02)00118-2

1546

J. Culpeper et al. / Journal of Pragmatics 35 (2003) 1545–1579

example, army training discourse (Culpeper, 1996), courtroom discourse (Lakoff, 1989; Penman, 1990), family discourse (Vuchinich, 1990), adolescent discourse (Labov, 1972; Goodwin and Goodwin, 1990), doctor-patient discourse (Mehan, 1990), therapeutic discourse (Labov and Fanshel, 1977; Lakoff, 1989), ‘workplace’ discourse (Andersson and Pearson, 1999), parliamentary discourse (Harris, 2001), ‘everyday conversation’ (Beebe, 1995), radio talk shows (Hutchby, 1996), and fictional texts (Culpeper, 1998; Tannen, 1990) (see also Kienpointner, 1997, for a good survey of types of ‘rude’ talk). In all these cases and in others, we would argue that politeness theories are unable to account fully for the interaction. This paper focuses on the notion of impoliteness, communicative strategies designed to attack face, and thereby cause social conflict and disharmony (see, also, Kienpointner’s 1997: 259–260 definition of ‘rudeness’, and that of Beebe, 1995: 159). Note that we are primarily concerned with what other researchers have referred to as ‘strategic’ (Lakoff, 1989) or ‘instrumental’ (Beebe, 1995) impoliteness, that is to say, it fulfills ‘a function that the speaker intended, and was not [simply] failed politeness’ (Beebe, 1995: 166). Also, we focus mainly on ‘genuine’ impoliteness, as opposed to ‘mock’ or ‘ritual’ impoliteness (cf. Labov, 1972; Leech, 1983: 144). A particular purpose of this paper is to build on the impoliteness framework mapped out in Culpeper (1996). One limitation of Culpeper (1996) is that it is focussed rather too narrowly on single impoliteness strategies, usually made up of particular grammatical or lexical items. In this paper, we will: (1) discuss the relationship between politeness and impoliteness, (2) investigate how impoliteness pans out in discourse, and (3) probe the role of prosody in communicating impoliteness. Although we focus on impoliteness, it will be clear that much of what we say has important implications for politeness theory, as well as for studies of discourse in general and the role of prosody in discourse in particular. This is not to say that we are arguing for a replacement of politeness theory. In fact, all the leading politeness theories at least mention the notion of impoliteness. The problem is that in practice they all focus solidly on politeness, with the result that their comments on impoliteness are descriptively inadequate and often conceptually biased (i.e. it is assumed that the concepts used to explain politeness can straightforwardly be applied to impoliteness) (see Eelen, 1999: chapter 1 and Eelen, 2001: chapter 3, for an elaboration of this point). We view our work here as partly a complement to politeness theory (specifically to the classical theory of Brown and Levinson, 1987), and partly a revision of parts of it. We begin with a brief description of our data and a justification as to why we chose it for our study.

2. The data The data for this study was taken from the BBC’s documentary television series The Clampers (1998). This series filmed the day to day activities of the London Councils’ privatised traffic wardens, including the ‘clampers’, those traffic wardens whose task it is to apply clamps to illegally parked cars and to tow them away. It also followed the activities of legal officials who adjudicate in disputes over the

J. Culpeper et al. / Journal of Pragmatics 35 (2003) 1545–1579

1547

penalties which have been applied by the traffic wardens.1 These activities often resulted in confrontation. In most cases, these confrontational episodes took the form of disputes, that is to say, ‘general disagreements in interaction which are displayed by the occurrence of some sort of opposition to an antecedent event’ (Corsaro and Rizzo 1990: 26). Two caveats regarding the data need to be borne in mind. Firstly, we have to remember that one of the likely goals of the television series is to entertain the viewing public.2 Thus, it is likely that the BBC have biased their selection of scenes for inclusion in the documentary series towards the more confrontational scenes. As a consequence, one cannot draw inferences about the norms of interaction between parking officials and car owners. Secondly, the existence of a camera crew in these scenes obviously leads to the problem of the observer’s paradox. The presence of the camera is likely, at least in some cases, to have affected the conversational behaviour of the participants. Culpeper (1996) and Bousfield (forthcoming) based much of their impoliteness work on army training discourse. Such discourse is particularly rich in impoliteness, because of the extreme inequality of power and the particular training philosophy, which aims to depersonalise the recruits. The Clampers is less rich in impoliteness, and, as a consequence, it represents a good data set for testing the margins of impoliteness.

3. The relationship with politeness theory 3.1. Why do we need an impoliteness framework? We’ve already got bald on record One particular criticism of Culpeper (1996), personally communicated by a number of academics, is that an impoliteness framework is unnecessary, since Brown and Levinson’s (1987) politeness framework already postulates a category, bald on record, which accommodates ‘impolite’ phenomena. Evidence of this argument is also in print. For instance, Thomas (1995: 171) places examples where the speaker has ‘deliberately chosen to be maximally offensive’ in bald on record. However, when we consider how the bald on record superstrategy is conceptualised, we find that such a view is problematic. Simply stating what bald on record is in a particular context is a very complex issue, not least of all because Brown and Levinson’s definition is not watertight. Their definition of bald on record utterances is those which are issued ‘in the most direct, clear, unambiguous and concise way as possible [...] following the specifications of Grice’s Maxims of Co-operation’ (Brown and Levinson, 1987: 69). Moreover, it is clear from the position of the bald on record superstrategy in their scheme that it is associated with a specific context, namely, one where the face threat is very small and so minimal politeness work is required. Specifically, 1

The full list of ‘officials’ includes traffic wardens acting as ticketers, clampers and those who tow away illegally parked vehicles; direct employees of the council, namely, public servants who work at the council’s vehicle storage pound (where towed vehicles are stored); court appointed bailiffs who operate to collect unpaid parking fines or offenders’ vehicles; and adjudicators, who hear from drivers appealing against fines levied on parking tickets, clamps, or towed vehicles. 2 For the link between impoliteness and entertainment, see Culpeper (1998).

1548

J. Culpeper et al. / Journal of Pragmatics 35 (2003) 1545–1579

they suggest that bald on record is used when: (1) the speaker and hearer recognise that face wants are suspended in the interests of some emergency, (2) the face threat is very small, and (3) the speaker is much more powerful than the hearer (1987: 69). For the purpose of discussion, let us review an example of the bald on record strategy used in an emergency situation, given in Thomas (1995: 170): [The speaker knows that a bomb has been planted in the stand at his racecourse. He thinks his young nephew is hiding in the stands] ... Toby, get off the stands. The stands are not safe. Toby, for Christ’s sake do what I say. This is not a game. Come on you little bugger ... for once in your life be told. Our first observation would be that the example does not appear to adhere to Grice’s (1975) maxims, and thus a key part of Brown and Levinson’s (1987) definition of bald on record. A maximally efficient bald on record version might be simply: ‘Toby, get off the stands’. This version would fit Thomas’s suggestion that in emergencies ‘the speaker is likely to focus on the propositional content of the message, and pay little attention to the interpersonal aspect of what is said’ (Thomas, 1995: 170). Thomas’s original version is inefficient through the use of linguistic devices that appear to attack face. In the final sentence, ‘you little bugger’ is an abusive term of address, and ‘for once in your life’ implicates, via the maxim of quality (Grice, 1975), the impolite belief that Toby is generally disobedient. Clearly, the focus is not purely on the propositional content, but also has an important interpersonal aspect. What seems to be happening here is that the speaker employs a short-term goal of causing face damage, in order to achieve the long-term benefit of saving his nephew’s life. Moreover, there are ‘maximally offensive’ examples that fall outside the specific contexts for bald on record mentioned by Brown and Levinson (1987). Thomas (1995: 171) supplies as one such example Tam Dalyell’s reference to Margaret Thatcher (then Prime Minister) in the British House of Commons: ‘I say that she is a bounder, a liar, a deceiver, a crook’. This does not fit Brown and Levinson’s (1987) contexts. There is also the issue of whether one can comfortably label such offensive phenomena as this bald on record politeness, when politeness is defined thus: [...] politeness, like formal diplomatic protocol (for which it must surely be the model), presupposes that potential for aggression as it seeks to disarm it, and makes possible communication between potentially aggressive parties. (Brown and Levinson 1987: 1, our emphasis) It is clear that bald on record does not adequately describe the variety of phenomena, including impoliteness phenomena, that can be—and have been by many researchers—placed within it.3 3 See also Watts’ (1992: 47–48) argument that the term ‘politeness’ is inadequate, given the variety of phenomena, including conflictual phenomena, which it attempts to cover.

J. Culpeper et al. / Journal of Pragmatics 35 (2003) 1545–1579

1549

3.2. Politeness, impoliteness and directness In fact, some impolite utterances are far from the directness associated with bald on record. Leech’s (1983: 108) comments on the relationship between politeness and indirectness are frequently cited. Indirect utterances, such as ‘Could you possibly answer the phone’, tend to be more polite because they increase optionality for the hearer, whilst decreasing illocutionary force (cf. Leech, 1983: 108). Less well known is Leech’s idea that indirectness can increase impoliteness: [. . .] in this case obliquity works in the opposite direction: because ‘You have something to declare’ is an impolite belief, the more indirect kinds of question [e.g. ‘Haven’t you something to declare’] are progressively more impolite, more threatening, than the ordinary yes–no question. (1983: 171) To take another example, You have shit for brains is clearly more indirect than You fool, and, context permitting, could be interpreted as very impolite for several reasons: (1) the criticism is personalized through the use of you (this is, of course, also true of You fool), (2) shit is a taboo word, and (3) the speaker flouts Grice’s maxim of quality, in order to implicate the impolite belief that the target has absolutely no intelligence. Of course, Leech would be the first to admit that the bi-directional correlation between indirectness and politeness/impoliteness does not apply in all contexts, and that other factors, aside from indirectness, affect politeness/impoliteness. For example, note that the quotation from Leech above refers to an ‘impolite belief’; in the case of polite beliefs, increased directness correlates with increased politeness (cf. Leech 1983: 109–110). Note that in the analysis of You have shit for brains we observed that shit as a taboo word partly accounts for its impoliteness. Taboo words are not an issue for directness–indirectness: ‘tabooness’ is conventionally associated with a word; it does not require the inferencing associated with indirectness. Furthermore, Blum-Kulka (1987), for example, reports that conventionally indirect strategies (e.g. Can you pass the salt?) are perceived as most polite by English and Hebrew speakers. This suggests not only that the indirectness–politeness scale is sensitive to cross-cultural variation, but also that the process of conventionalisation can ‘short-circuit’ the indirectness scale. Many impolite expressions are conventionalized to a degree (cf. Jay, 1992: 2), but whether the same would apply to the indirectness–impoliteness scale—namely, that conventionalized strategies short-circuit the indirectness scale and therefore are less impolite—is an area that could benefit from future investigation. Interestingly, it is precisely those utterances that have the directness of bald on record which are difficult to interpret as either polite, impolite, or something in between. As Brown and Levinson (1987) realized, the context is crucial. For example, if it were not part of someone’s role (as a tutor, say) to make a criticism, and if it were known that the addressee was particularly sensitive to criticism, then It was bad, as a description of the addressee’s work, would seem to be impolite. It should be noted that a key difference between politeness and impoliteness is intention:

1550

J. Culpeper et al. / Journal of Pragmatics 35 (2003) 1545–1579

whether it is the speaker’s intention to support face (politeness) or to attack it (impoliteness). Of course, identifying speaker intention is a problematic business, and this is one of the issues we will attend to in the following section, where we will also consider types of (im)politeness. 3.3. Notions of politeness and notions of impoliteness In order to help understand the notion of impoliteness, we will first briefly consider the types of action that may lead to face damage. Goffman (1967: 14) suggests to us that there are three types of action which constitute a threat to face. Firstly, ‘the offending person may appear to have acted maliciously and spitefully, with the intention of causing open insult’ (Goffman, 1967: 14). For example, in the clampers data the following interchange between S1 (a traffic warden) and S2 (who owns a car which has received a parking penalty) contains an offence which aims at aggravating the face of the intended recipient (throughout this paper italics indicate the particular part of the discourse we wish to focus on):4 [1] S1: I’m afraid I have sir= S2: you have no authority to put a ticket on a disabled err car =why didn’t you S1: I’d . what do you mean I did do it before S2: do it before then I’ve parked here every day because I have a S1: yes sir I never really S2: prescription from the chemist you don’t put any tickets on my car and then and S1: I’m not I’m not always down here sir S2: then you just come in and out of the blue and put one on ohh fuck off

Our understanding of impoliteness falls within this category. Secondly, ‘[t]here are incidental offences; these arise as an unplanned but sometimes anticipated by-product of action—action the offender performs in spite of its offensive consequences, though not out of spite’ (Goffman, 1967: 14). In the following example, S1 is an adjudicator for London’s Southwark council. He hears and decides upon appeals concerning all types of parking offences. S2 and S3 are a husband and wife (respectively) who are appealing against a ticket which they have received. This extract is taken from the final stages of their appeal hearing: [2] S1: well look I must draw this to a close . I’ve listened to you very carefully and I’m . do S2: S3: 4 Throughout our study we have used the stave transcription method for representing interaction. Within this approach, (S1) or (S2), for example, represent a particular speaker; a full-stop represents an audible pause of more than 1 second; an equals sign represents latching (i.e. one interactant speaking immediately another finishes); non-verbal actions, contextual information, and transcription difficulties (e.g. ‘indistinct’) are placed in angle brackets.

J. Culpeper et al. / Journal of Pragmatics 35 (2003) 1545–1579

1551

S1: understand entirely what your point is but I can’t allow your appeal I’m afraid= S2: =Oh why S3: S1: erm erm I can’t say that the ticket has been incorrectly issued because it’s been S2: not I I S3: S1: correctly issued not incorrectly issued S2: we’re not we’re not arguing that I S3: but it hasn’t been correctly issued S1: well in my opinion it has S2: what happens if a sign changes overnight S3: but

The adjudicator’s act of denying an appeal and ‘disagreeing’ with the interlocutors does not appear to be impoliteness, as we understand it, since the intention is not likely to be primarily to cause face damage per se to the vehicle’s owners, but to perform actions consistent with the duties and constraints his job imposes (e.g. he could not allow the appeal in the context of insufficient evidence). However, face damage is likely to be an anticipated by-product of the action of denying an appeal (indeed of ‘disagreeing’ generally). The action is done in spite of the offensive consequences, but not out of spite for the vehicle’s owners. This kind of by-product offence is captured within politeness frameworks. Note that the adjudicator takes into account the potential offensive consequences of denying an appeal by showing empathy with the vehicle’s owners (I’m . do understand entirely what your point is), expressing regret (I’m afraid), and couching his refusal as a matter of his inability to do otherwise in this context rather than as his own desire (I can’t allow your appeal and I can’t say that the ticket has been incorrectly issued because it’s been correctly issued). In other words, he predominantly uses positive politeness (cf. Brown and Levinson, 1987). Thirdly, the offending person ‘may appear to have acted innocently; his offence seems to be unintended and unwitting [. . .]. In our society one calls such threats to face faux pas, gaffes, boners or bricks’ (Goffman, 1967: 14). No clear examples of this type of offence occurred within our data. An ‘unwitting’ offence might consist of something like the following example, reconstructed from an actual conversation which took place between a high ranking member of Lancaster University’s senate, S1, and a member of the academic staff, S2: [3] S1: < sincerely >when is it due? S2: .. I’m not pregnant

Here S1 has unwittingly drawn attention to the fact that S2 looks overweight. Such an action could be construed as face threatening within a Western cultural setting.

1552

J. Culpeper et al. / Journal of Pragmatics 35 (2003) 1545–1579

Whilst Goffman’s categorisation is helpful, it does not capture the indeterminate instances one finds in real data. For example, the following exchange occurs at the very end of a dispute which has taken place between S1 (a parking official) and S2 (the owner of a clamped van). S2 has been disputing S1’s right to clamp and the ethics of clamping. Neither participant has been willing to concede to the other, and eventually S2 ‘closes’ the exchange, thus: [4] S1: that’s right I clamped your car sir and I won’t dispute that fact that I clamped your car S2: well S1: well that’s fine by me if you don’t want to talk we don’t have to talk S2: end of conversation

The problem which concerns us here is that we cannot be sure whether S2’s utterance was performed maliciously, with ‘the intention of causing open insult’ (Goffman 1967: 14) to S1 (by snubbing him), or whether S1 simply grew tired of the discussion and its apparent lack of progress, and thus decided to end the conversation in spite of the offensive consequences of doing so, though not out of spite for S1. Interlocutors do not wear their intentions on their sleeves. Speakers’ intentions, of course, are not just the basis of Goffman’s categorisation above, they are also fundamental to speech act theory. And, as a consequence of the fact that Brown and Levinson’s (1987) politeness theory is built on speech act theory, it too has speakers’ intentions at its heart. However, even within conversation analysis and other more sociologically oriented approaches to interaction, there is an admission that participants in conversation do attribute intentions and purposes of some kind to talk. Grimshaw (1990: 281), reviewing a collection of sociolinguistic investigations into ‘conflict talk’, makes the following points: [The] attribution of purposiveness to participant behaviours, will be sharply criticized by those students of talk (e.g. conversation analysts) who argue that this implies the ability to ‘‘get into people’s heads’’ and requires unwarranted inferences and claims. The researchers whose work is reported here do not contest the position that what is in people’s heads is accessible neither to analysts nor to interlocutors (nor even, ultimately, fully accessible to those whose behaviour is under investigation). I believe most of them will also argue, however, that the availability of ethnographic context and of an optimally complete behaviour record permits analysts to make such inferences and attributions which are ‘‘for-the-most-practical-purposes’’ (paraphrasing Garfinkel) no less plausible than those of actual participants. This claim is subject to qualification but the disambiguation process is that which we ourselves employ in interaction—where, it must be conceded, we sometimes err. There is no claim, then, that one can reconstruct the actual intentions of speakers, but rather that ‘plausible’ intentions can be reconstructed, given adequate evidence. One of the main reasons why we consider impoliteness in its discourse context in

J. Culpeper et al. / Journal of Pragmatics 35 (2003) 1545–1579

1553

Section 4 and prosody in Section 5 is precisely to provide that adequate evidence, something that politeness theory analyses tend not to do.5 3.4. Impoliteness work inspired by Brown and Levinson (1987) Three papers, Lachenicht (1980), Austin (1990) and Culpeper (1996), have postulated impoliteness frameworks, each using Brown and Levinson (1987) as their point of departure. Lachenicht (1980), in what has been described as a ‘rarely cited but nevertheless meritorious paper’ (Turner, 1996: 7), focuses on ‘aggravating language’, a rational and intentional attempt to ‘hurt’ or damage the addressee (1980: 607). Lachenicht (1980) views ‘aggravating language’ as an extension to Brown and Levinson’s (1987) politeness framework. He considers four ‘aggravation’ superstrategies, presented below in order of face threat : (i) Off record: ambiguous insults, insinuations, hints, and irony. This strategy is of much the same kind as the politeness strategy, and is designed to enable the insulter to meet an aggrieved challenge from the injured person with an assertion of innocence. (ii) Bald on Record: directly produced FTAs and impositions (‘Shut that door’, ‘Do your work’, ‘Don’t talk’, etc.) of the same kind as in the politeness strategy. (iii) Positive aggravation: an aggravation strategy that is designed to show the addressee that he is not approved of, is not esteemed, does not belong, and will not receive cooperation. (iv) Negative aggravation: An aggravation strategy that is designed to impose on the addressee, to interfere with his freedom of action, and to attack his social position and the basis of his social action. (Lachenicht, 1980: 619). Note that the first two strategies are not in fact part of a new impoliteness framework, but are taken from Brown and Levinson’s (1987) politeness framework. Positive aggravation and negative aggravation appear to be distinguished in terms of their orientation to positive and negative face wants (cf. Brown and Levinson, 1987). Undoubtedly, the chief merit of Lachenicht’s (1980) work lies in the fact that it provides an extensive review of linguistic strategies that may be used to aggravate face. There are, however, a number of issues concerning the consistency, speculative nature, and validity of Lachenicht’s (1980) model. For example, Lachenicht claims (1980: 631) that ‘[p]ositive aggravation informs the hearer that he is not liked, will not be cooperated with, and does not belong. Essentially it attacks his need for freedom of action, for status, and for power’. But, as indicated by other references and claims in other parts of Lachenicht (1980), attacks on ‘freedom of action’ 5 Of course, not all notions of politeness in the literature have intention at their heart. For example, Watts’ (1992) notion of ‘politic behaviour’ emphasises doing what is socially and culturally appropriate.

1554

J. Culpeper et al. / Journal of Pragmatics 35 (2003) 1545–1579

concern negative and not positive face. Brown and Levinson (1987: 61) point out that positive face is the desire that an individual’s wants and needs are respected by others, whereas negative face is the desire for ‘freedom of action’. If positive and negative aggravation are supposed to relate to positive and negative face, as defined by Brown and Levinson (1987), they fundamentally fail to do so. Another significant issue is that Lachenicht’s (1980) model is based upon and describes anecdotal, constructed examples, and written material from a number of dictionaries of insults. No ‘real life’ conversational data, either written or verbal, are utilised. Turner (1996: 7) writes, ‘it might be stressed that the paper is an essay not in analysis but in constructivism and so the specific details are subject to trial and revision by the data that are collected.’ Indeed, lack of any such trials or revisions of Lachenicht (1980) relegates numerous claims made therein to purely hypothetical ones. One brief example will suffice: ‘Probably swearing is past its prime today, for the decline of religious belief has made it less useful. Today, it is manly ‘God-damns’ and ‘bloodys’ that are popular’ (1980: 641). Superficially, Austin (1990) and Culpeper (1996) appear to have a lot in common. Both talk about ‘face attack’ and both derive a framework from Brown and Levinson (1987). However, there is a fundamental difference. Austin (1990) is a hearer-based account of how utterances can be interpreted as offensive. She shows (1990: 285) how apparent compliments like You have been a capable and decorative chairman could, even if it may have been intended as a straightforward compliment, have offensive implications for the hearer in a particular context. Whilst her paper is useful reminder that Brown and Levinson (1987) underestimate the role of the hearer and of the context, Austin steadfastly overlooks the role of the speaker. Her paper is not about the communication of impoliteness, but the interpretation and perception of it. Thus, her examples include cases which may simply have involved the miscommunication of politeness (e.g. too little or too much politeness work in a particular context), in other words, what we referred to as ‘unintended’ impoliteness in Section 3.3. A particular weakness of Austin’s (1990) paper is that her interpretations of offence are untested. We view it as a priority and a challenge for future research to test the offensive effects of linguistic impoliteness. Jay’s (e.g. 1992, 2000) psycholinguistic work on taboo words, their use, and their perceived offensiveness provides a valuable starting point. Culpeper (1996), unlike Lachenicht, considers not just an extension to Brown and Levinson (1987), but explores the possibility of a parallel structure. Impoliteness superstrategies for Culpeper are ‘opposite’ in terms of orientation to face (i.e. instead of maintaining or enhancing face, they are designed to attack face), but not necessarily opposite in other pragmatic ways (e.g. from a Gricean point of view, the opposite of bald on record is off-record). The superstrategies are summarised here. 1. Bald on record impoliteness. This is distinct from Brown and Levinson’s bald on record strategy which is deployed for polite purposes in fairly specific circumstances, namely, where there is little face at stake, an emergency situation, or no intention of damaging the face of the hearer. In contrast, bald on record impoliteness is typically deployed where there is much face at stake, and where there is an intention on the part of the speaker to attack the face of the hearer.

J. Culpeper et al. / Journal of Pragmatics 35 (2003) 1545–1579

1555

2. Positive impoliteness. The use of strategies designed to damage the addressee’s positive face wants (‘ignore, snub the other’, ‘exclude the other from the activity’, ‘disassociate from the other’, ‘be disinterested, unconcerned, unsympathetic’, ‘use inappropriate identity markers’, ‘use obscure or secretive language’, ‘seek disagreement’, ‘make the other feel uncomfortable (e.g. do not avoid silence, joke, or use small talk)’, ‘use taboo words’, ‘call the other names’, etc. ). 3. Negative impoliteness. The use of strategies designed to damage the addressee’s negative face wants (‘frighten’, ‘condescend, scorn, or ridicule’, ‘invade the other’s space’, ‘explicitly associate the other with a negative aspect’, ‘put the other’s indebtedness on record’, ‘hinder or block the other—physically or linguistically’6, etc.). 4. Sarcasm or mock politeness. The use of politeness strategies that are obviously insincere, and thus remain surface realizations. Sarcasm (mock politeness for social disharmony) is clearly the opposite of banter (mock impoliteness for social harmony).7 5. Withhold politeness. Keep silent or fail to act where politeness work is expected. It needs to be stressed that, as with all politeness phenomena, impoliteness does not simply arise from any one particular strategy, but is highly dependent on context. Moreover, there is no claim that the impoliteness superstrategies belong to one dimension: precisely what those dimensions are is an area that needs further research. Also, we make no claims about how the impoliteness superstrategies should be ordered for degree of offence, or about how hard-edged the distinctions are between them. Whilst these are also areas that need further consideration, we will adduce evidence in Section 4 that some strategies can be mixed. Note that Brown and Levinson (1987: 17–20) resist the idea that politeness strategies can be mixed (for example, positive politeness markers occurring in negative politeness strategies such as indirect requests), despite claims by other researchers (e.g. Craig et al., 1986: 452–453) to the contrary. The fact these frameworks are inspired by Brown and Levinson’s (1987) model means that weaknesses associated with their model tend to be carried over into impoliteness frameworks. Specifically, this means that little is said about matters to do with sequencing in discourse or prosodic aspects. It is these very areas that this paper will address. Furthermore, as Culpeper (1996: 358) points out, Brown and Levinson (1987) are primarily geared to handling matters relating to linguistic form, and thus impolite implicatures can slip through their framework, as they themselves admit (1987: 11). Leech’s (1983) politeness model, primarily concerned with linguistic content, may be used to complement Brown and Levinson (1987). 3.5. Further evidence of the impoliteness strategies in Culpeper (1996) All of the impoliteness superstrategies outlined in Culpeper (1996), which were largely derived from a investigation of army recruit training data, were also represented in our data, as were most of the specific linguistic output strategies. First, we 6 Whilst relevant phenomena were discussed, the label for this particular strategy was not in fact given in Culpeper (1996), but in Bousfield (1999: 25). 7 Our understanding of sarcasm here is close to Leech’s conception of irony (1983: 82, 142).

1556

J. Culpeper et al. / Journal of Pragmatics 35 (2003) 1545–1579

will cite examples of these strategies, starting with bald on record impoliteness, and then we will note any other strategies which occurred. The following interaction, which we cite to illustrate ‘bald on record impoliteness’, involves S1, a clamper who is preparing to ticket a car in a road next to a primary school, and parents of children who often park in order to drop their children into the school. Prior to this interaction, a number of parents have been ticketed and one has had her vehicle removed. The conversation between S1 and S2, S3 (Parents) and S4 (the school’s headmistress) has been confrontational up to this point: [5] S1: we’ll start with you madam < to S4> I work for T F M parking okay S2: has made no attempt to respond S3: excuse me excuse me you are S4: S1: I did the first time I met you okay where’s your car S2: S3: a parking attendant alright act like one okay shut up and act like a parking attendant S4:

Bald on record impoliteness occurs when S3 says shut up and act like a parking attendant. Here we see two imperative commands that are deployed baldly with the purpose of aggravating the face of the parking attendant. It might also be noted that S3 is in a position of relative powerlessness, and so this does not match one of the contexts which Brown and Levinson (1987: 69) attribute to bald on record politeness. Frequently occurring output strategies for ‘positive impoliteness’ include: ‘ignore, snub, fail to attend to H’s needs’, ‘avoid agreement’, and ‘use taboo language, swear, be abusive’. In the following example, S1 (a clamper) returns to remove the clamp from S2’s van, following S2’s payment of the fine. S2 challenges S1 as to why he clamped the van. S1 is keen to explain that there is nothing ‘personal’ in his action. [6] S1: I can take your notes on board but there’s nothing I personally can do I simply work do my job for S2: just do your job S1: the council I do my job for the coun if you want me to explain then if you S2: I don’t care what you do S1: want be like that then I can walk away I don’t have to talk to you if I don’t want to if S2: S1: you’re going to be rude to me yeah I that’s fine then sir I S2: I don’t really want to talk to you you’re not going to do anything S1: S2: about it are you

Clearly, S2 fails to attend to S1’s need to exonerate himself. In our next example, S1 (a council official) is urging S2 (an angry car owner) to reconsider his decision that,

J. Culpeper et al. / Journal of Pragmatics 35 (2003) 1545–1579

1557

rather than pay the towing and storage costs, he will leave the council to scrap his impounded car. [7] S1: oooh come on S2: I don’t want the car fuck it you lot can keep the car no there’s no ‘come on’ because S1: S2: you’ve towed the car off my own . blasted road where I’m staying right . twice . not once but S1: S2: twice that’s two nearly three hundred pounds

S2 explicitly block’s S1’s attempt to bring about agreement. In our final example, S1 (the court bailiff) arrives at the home of S3 and S2 (who are husband and wife), in order to collect court issued penalties and other legal costs for non-payment of parking fines incurred against S2’s car. S1’s assistant is loading S3 and S2’s car onto a truck.

[8] S1: the car is going he has a S2: what the fuck you doing . excuse me what are you fucking doing S3: < S2 hits S1 in mouth- S1 starts speed dialling on the phone> S1: court order police please yeah S2: really . you want some fucking money right S3: all you have S1: you can’t S2: all you have to do is ask for the money you don’t have to fucking take the car S3: to do is ask for the fucking money right S1: get in the car madam S2: piss off S3: jackie come here

Clearly, taboo language is used repeatedly here. Of course, such language can simply be an expression of a speaker’s emotional state. That may be part of the explanation here, but we would argue that it is predominantly an impoliteness strategy to make S1 feel uncomfortable: the taboo language expresses anger targeted at the hearer (see Section 5.2.1, for further discussion of targeted anger). Frequently occurring output strategies for ‘negative impoliteness’ include: ‘condescend, scorn, ridicule’, and ‘hinder linguistically (e.g. interrupt, deny turn) or physically (e.g. block other’s passage)’. In the following example, at an appeal hearing, S2 and S3 (husband and wife) have being arguing with S1 (the adjudicator) over a parking ticket. They have refused to listen to his ‘polite’ explanations as to

1558

J. Culpeper et al. / Journal of Pragmatics 35 (2003) 1545–1579

why he will not allow their appeal (see the discussion of example [2]). S1 attempts to close the conversation and make S2 and S3 leave his office. [9]

S1: I I’ve finished Mr Culp . will you will you will you please leave the room . I’m not S2: say in this situation S3: S1: answering any further questions . do you want me to press the buzzer S2: S3: well that’s being S1: S2: S3:

will you please leave the room babyish isn’t it

Clearly, the item ‘babyish’ is used to express scorn at the adjudicator’s threat (we return to this particular speech event in Section 5). In the next example, S1 (a traffic warden) is ticketing cars. S2 (a mother) is currently being ‘ticketed’; S3 (another mother) has taken umbrage at this. The discussion up to this point has been about the right of the parents to park there. S4 (the school’s headmistress) has been insisting that the parents have special dispensation to park at the start and finish of the school day. In this segment, S1 has just challenged the view of special dispensation. [10]S1: you .. one at a time ladies one at a time who wants to speak to me first ladies I mean take S2: the school is trying the council makes no S3: < ————indistinct————— > S4: this is my chi S1: your pick cos you can queue up if you want I’ve only got one set of ears you know what S2: S3: S4: S1: I mean well I am trying to be but I’ve only got one S2: it’s nice to see you being so responsive the council S3: S4: S1: set of ears we’ll start with you madam < to S4> I work for T F M parking okay S2: has made no attempt to respond S3: excuse me excuse me you are S4: S1: I did the first time I met you okay where’s your car S2: S3: a parking attendant alright act like one okay shut up and act like a parking attendant S4:

J. Culpeper et al. / Journal of Pragmatics 35 (2003) 1545–1579

1559

S1 interrupts S2 by allocating a turn to S4, and thus denying it to S2. Note that S3 realises the face damaging aspect of this (to S2) and herself interrupts S1, in order to begin a criticism of his attitude. Note also that shut up, which we commented on under the heading bald on record impoliteness, is also an aggressive means of impeding speech. As an illustration of ‘sarcasm’, in the following example S4 (a van driver) has returned to his vehicle to find that S1 (a clamping official) has only just finished clamping it. After a highly confrontational conversation in which S4 demonstrates his anger, he snatches the penalty ticket and other details away from S1 and tells him to ‘go away’. S1 returns to his own vehicle to leave the scene. As he does so, S4 issues the following utterance: [11] S1: I will do S4: .. (sarcastically) have a good day

In this context, S4 is highly unlikely to be expressing a sincere polite wish; rather, he expresses via an implicature the opposite—the impolite wish that he has a bad day. Finally, we present an example of ‘withhold politeness’, Here S1 (an adjudicator) has just refused S2’s (a car owner) appeal against a parking ticket. [12] S1: well thank you very much for coming good day S2: I don’t thank you at all

S2 explicitly withholds politeness by not reciprocating S1’s thanks. This, of course, is an ‘on-record’ example of withholding of politeness. Although there were not any verifiable cases in our data, an alternative means of withholding politeness is simply to do nothing in contexts where politeness is expected. Whilst the impoliteness strategies mentioned in Culpeper (1996) accounted for much of the face attack in our data, we discovered some others. One of the most frequent strategies in our data was to ‘challenge’ (negative impoliteness) the other speaker, usually by means of a rhetorical question. The following interaction takes place between S1 (a clamper) and S2 (a car owner). [13] S1: can you just answer me one question can you see the yellow line visibly S2: S1: under your car . . . S2:

I live here why is there a yellow line anyway why do

S1: S2: I have to park my car three hundred yards up the raod it gets stolen S1: S2: broken into vandalised three times this year already why have you done S1: S2: it why do you make my life impossible how am I supposed to work doing

1560

J. Culpeper et al. / Journal of Pragmatics 35 (2003) 1545–1579

S1: can I just say you you you can clearly see the yellow line on the road S2: this S1: it’s not a new yellow line its been there for quite some time S2: so why don’t you just S1: you shouldn’t have parked on a yellow line S2: stop the ticket

S2 issues no less than five challenges to S1. Some of the challenges, why is there a yellow line anyway,why do I have to park my car three hundred yards up the road, how am I supposed to work, have some potential for just being ‘incidentally’ face-damaging utterances (i.e. the speaker is merely given vent to personal vexation). However, two of the challenges, why have you done it, andwhy do you make my life impossible, are more clearly directed at S1 (cf. ‘you’), and are more likely to be impolite utterances, deployed specifically to aggravate the face of S1. Moreover, the rhetorical nature of some of the questions, particularly the final two why do you make my life impossible and how am I supposed to work, is part of a strategy to implicate impolite beliefs (e.g. ‘you make my life impossible’). The ‘challenge’ strategy is, in fact, discussed by Lachenicht (1980: 668–671), who draws upon the discussion in Labov and Fanshel (1977).

4. Impoliteness in discourse Unlike the traditional approach of focussing on single politeness or impoliteness strategies (e.g. Brown and Levinson 1987; Culpeper 1996), in this section we investigate how impoliteness is realised in extended discourse. Of course, it is not the case that extended discourse has been totally neglected within politeness studies, although only a handful of studies have investigated this area. Aronsson and Rundstro¨m (1989), for example, examining doctor–patient discourse, show that the on-record/off-record distinction is best understood sequentially and as a matter of negotiation, whilst Hayashi (1996) shows how politeness strategies for dispreferred messages are produced as sequences of acts. More recently, Harris (2001) considers how politeness theory can be extended in order to analyse formal institutional discourse—specifically, British parliamentary discourse—including intentionally face threatening acts, and focusses on the performance and perception of (im)politeness in relatively long stretches of discourse. First, we briefly discuss how individual impoliteness strategies co-occur in and across a particular participant’s turns. Second, we look at patterns of impoliteness and reactions to it across exchanges. We believe that research into such patterns provides some of the richness of context necessary for identifying and analysing impoliteness (cf. Section 3.3). 4.1. Multiple strategies Discussing aggravating language, Lachenicht notes that ‘it is possible to combine more than one sub-strategy into an utterance’ (Lachenicht, 1980: 635). In fact,

J. Culpeper et al. / Journal of Pragmatics 35 (2003) 1545–1579

1561

combinations of impoliteness strategies turned out to be the norm in our data: in any one interaction a participant never used a single strategy just once. We identified two ways, but not mutually exclusive ways, in which impoliteness strategies pattern in a participant’s turns. A particular strategy (or combination of strategies) can be used repeatedly to form a parallelism. That is to say, words, grammatical structures, intonational contours, or indeed any feature constituting a pragmatic strategy can be used repeatedly (usually in juxtaposition) to form a parallelism (a perceptually prominent pattern where some features are held constant and others may vary). We have already seen a good example of this in [13] above, where S2 (a car owner) delivers five challenges to S1 (a clamper) as a means of attacking S1’s face. This series of challenges is issued rapidly with minimal gaps between each of the challenges. Holmes (1984: 355), discussing the modification of illocutionary force, points out that ‘repetition itself serves as a rhetorical device to increase the force of the repeated speech act’. Clearly, in this case repetition serves to boost the challenge, and thereby to exacerbate the threat to S1’s face. Repetition can also, as in [13], ‘hog’ the conversational floor, imposing on the other participant’s negative face. The key point about the repetition (the parallelism) of impoliteness strategies relates to interpersonal effect: it increases the imposition upon the target and/or emphasizes the negative attitude of the speaker towards the target. In other words, it can boost impoliteness. A particular strategy can be used in combination with other strategies. An example of this can be found in the example [8] above. S2 (a car owner) combines the strategy of ‘challenge: ask a challenging question’ (negative impoliteness) with the strategy of ‘use taboo words’ (positive impoliteness) to make what the fuck you doing. In fact, S2 also uses the above tactic of parallelism in repeating this particular combination, albeit in a slightly different form: what are you fucking doing. Whilst the taboo word fuck(ing) plays the grammatical role of intensifier (e.g. Quirk et al., 1985: 445, 589– 591), amplifying or boosting the force of the challenge, on an interpersonal level it marks the extremely negative attitude of the speaker towards the hearer. Holmes (1984: 363) again makes a pertinent remark: ‘[d]evices may reinforce one another, as when strong stress, lexical Boosters and repetition co-occur in one utterance’. The cumulative effect of using mutually reinforcing impoliteness strategies is to boost the impoliteness. It is worth noting that the ‘use taboo words’ strategy (Culpeper, 1996) seems to be the one most likely to combine with other strategies, and in future research would benefit from specific attention. As we noted in Section 3.4, Brown and Levinson (1987) resist the idea of strategy mixing.8

8 Harris (2001) observes a mixture of negative politeness and positive impoliteness, as well as rapid transitions from positive politeness to positive impoliteness, in British parliamentary discourse. Whilst this is further evidence of strategy mixing, we need to be aware that the ‘negative politeness’ to which she refers is rather different in type from the politeness or impoliteness we have been discussing here. The distancing strategies (e.g. third person address) and deference (e.g. vocatives like ‘my Right Honourable Friend’) in the House of Commons are a matter of convention, not strategic manipulation; they are ‘politic’ behaviour, not ‘politeness’ behaviour (Watts, 1992).

1562

J. Culpeper et al. / Journal of Pragmatics 35 (2003) 1545–1579

Just how frequent such multiple strategies are in other conflictive discourses; whether certain strategies are more likely to be combined or repeated (or combined and repeated); and what precisely the cumulative effects of such tactics are remain areas for future consideration. For example, we need to investigate whether we can assume that more linguistic impoliteness leads to more offence. A hint that this may not always be the case appears in Greenberg’s (1976) study of perceived verbal aggressiveness. He found that when an utterance reaches the higher levels of aggression the addition of boosters (specifically, frequency qualifiers like ‘always’) made little difference to how it was perceived. We now consider the patterning of impoliteness across exchanges. 4.2. Impoliteness and responses to it Research on both politeness and impoliteness has tended to overlook what the recipient of face threat or attack does. This is crucial, since, as every conversation analyst knows, the response to an utterance can reveal much about how that utterance is to be taken. Labov’s study (1972) assumes that one can identify a personal insult, as opposed to a ritualistic one, by the response that such insults elicit. He suggests that personal insults are followed by a denial (cf. Labov, 1972: 333–335, 347). On the other hand, Harris et al.’s (1986) study of verbal aggression notes the common assumption that the best way to save face in the light of verbal attack is to counterattack. In this section, we will consider whether these basic pairs, OFFENSIVE– DEFENSIVE and OFFENSIVE–OFFENSIVE, are reflected in our data. We will argue that there is, in fact, a more complex set of strategies for managing face attack than these pairs suggest. Also, we will note how sequences of impoliteness acts can constitute particular ‘activities’, which in turn affect the interpretation of impoliteness. Theoretically, when a recipient of an utterance perceives a strategic impoliteness act—an exacerbated face threatening act (FTA)—they have two choices open to them: they can either respond or not respond (i.e. stay silent). The option of not responding did not occur in our data, and so we will not pursue it here. But we briefly note that this option presents particular problems for both the other participants in the original speech event and the researcher, who must depend solely on contextual factors in interpreting the meaning of the silence. Participants who choose to respond to the impoliteness act have a further theoretical set of choices open to them: they can either accept the face attack or they can counter it. In accepting the face attack, the recipient may, for instance, assume responsibility for the impoliteness act being issued in the first place. Thus, repeated, strong and personalized complaints (i.e. an impoliteness act) might be met with an apology. Note that this option involves increased face damage to the responder. This option did not occur in our data, and so we will not discuss it further. The alternative option, to counter the face attack, involves a set of strategies which can be considered in terms of whether they are offensive or defensive. Offensive strategies primarily counter face attack with face attack; this is the pattern referred to by Harris et al. (1986). Such strategies are the impoliteness strategies of researchers like Culpeper (1996). Defensive strategies primarily counter face attack by defending one’s own face; this is the

J. Culpeper et al. / Journal of Pragmatics 35 (2003) 1545–1579

1563

Fig. 1. A summary of response options.

pattern referred to by Labov (1972). As we will illustrate below, such strategies seek to deflect, block or otherwise manage the face attack. However, note that these strategic groupings are not mutually exclusive. Offensive strategies have, to some degree, the secondary goal of defending the face of the responder; defensive strategies may have, to some degree, the secondary goal of offending the speaker of the original impoliteness act. As a consequence, the distinction is best conceived of as a scale. The response options we have described are represented in Fig. 1. Now that we have mapped out a complete theoretical set of response options, we will focus on some of the specific counter strategies used in our data, first considering more clearly offensive strategies, then more clearly defensive. There are in fact no clear examples of counter strategies that are clearly primarily offensive, which means that there are no clear examples of the OFFENSIVE–OFFENSIVE pair. This is not surprising, since, whilst a clamper has the power to ticket, clamp or even tow away an owner’s illegally parked vehicle, they do not in their particular socio-discoursal role have the legitimate power to respond to the impoliteness of car owners with clear, unambiguous impoliteness. Clearly, the social context constrains the response options that are actually available to a participant in an interaction. A striking feature of army recruit training discourse is that ‘do not respond’ is often the only option available to recruits, who are in a position of very low relative power (see Culpeper, 1996 and Bousfield, forthcoming). In order to illustrate the OFFENSIVE–OFFENSIVE pair, we will briefly present an example from Beebe (1995: 154, sourced from Wendy Gavis, field notes): A man in a compact red car was trying to maneuver into a parking spot right next to a crosswalk at a corner in New York City. A woman was crossing the street with her two children as the man attempted to park. She was very thin. He had a big ‘‘beer belly’’. The two were arguing over whether she, the pedestrian, or he, the driver, had the right of way. Finally the woman yelled: Woman: Man: Woman: Man:

Oh shut up you fat pig Go fuck yourself Go on a diet Go fuck yourself

1564

J. Culpeper et al. / Journal of Pragmatics 35 (2003) 1545–1579

This example has a relatively simple sequential organisation, thus: (Woman) OFFENSIVE (bald on record impoliteness+derogatory nomination) Oh shut up you fat pig (Man) OFFENSIVE (implicated impolite belief+use taboo word) Go fuck yourself (Woman) OFFENSIVE (bald on record impoliteness+implicated impolite belief) Go on a diet (Man) OFFENSIVE (implicated impolite belief+use taboo word) Go fuck yourself Note also that the Man repeats a combination of strategies, and that the Woman’s are parallel with respect to the fact that they both express the impolite belief that he is fat. Thinking about the patterning of strategies across exchanges, a parallelism begins to emerge in the final three turns—initial imperative Go and then the rest of the sentence. Lein and Brenneis (1978; see also Brenneis and Lein, 1977) noted that ‘repetition’ was a characteristic pattern across exchanges in children’s disputes. With OFFENSIVE–OFFENSIVE pairs, presumably the objective is to at least match the other in kind—a ‘tit for tat’ strategy. Another possibility, again observed by Lein and Brenneis (1978), is ‘escalation’, where each speaker uses a stronger strategy than the previous speaker. For example (Lein and Brenneis, 1978: 301):

James: Art: James: Art:

I sock you in your nose. I sock you in the mouth. You gonna have a black eye you keep on. You gonna have a bloody nose, and a bloody mouth and knocks one of your teeths out. I’m gonna knock ‘em down your throat.

Impoliteness escalations, such as the above, are referred to as ‘conflict spirals’ in the conflict literature. Andersson and Pearson (1999), for example, studying workplace interactions, postulate the factors that create conflict spirals (e.g. perceptions of ‘interactional injustice’, negative affect, desire for revenge) and facilitate them (e.g. ‘hot temperament’, situational informality). Turning now to defensive counter strategies and the OFFENSIVE–DEFENSIVE pair. One possibility is direct contradiction. Let’s take an example from Brenneis and Lein (1977: 56–7), who refer to this pattern as ‘inversion’:

Dave: Larry: Dave: Larry: Dave: Larry:

I am, you dumb-dumb. I’m not no dumb-dumb, dodo. Yes, you are. No, I’m not. Yes, you are. No, I’m not.

J. Culpeper et al. / Journal of Pragmatics 35 (2003) 1545–1579

1565

No clear examples of this emerged in our data, perhaps because it is a relatively unsophisticated strategy and one which is unlikely (as the above example illustrates) to bring a swift end to the altercation. One specific defence strategy we did identify in our data was ‘abrogation’: the abrogation of personal responsibility for the action(s) or event that caused the interlocutor to issue a face damaging utterance in the first place. In Labov’s (1972) terms, this is a type of denial. It works by deflecting the FTA. In our data, abrogation involves attempting to switch either social role (from being addressed as private citizen to that of a public servant) or discoursal role (where an interactant emphasises that they are merely acting in a representative role such as a ‘mouthpiece’). Abrogation by social role switching is like saying ‘I’m not to blame, I’m just following orders!’, whilst abrogation by discoursal role switching is like saying ‘Don’t shoot the messenger!’ An example of defensive social role switching by S1 (a clamper) (and of implicit discoursal role switching) can be seen in [14] below. This example also illustrates, on the part of S2 (a car owner), how role triggering or role enforcing can be a more offensive strategy.

[14] S1: =well you see I’m just doing a job but I’ve come along S2: don’t you think this is a bit stupid= yeah S1: here and yeah S2: well so was Hitler all I’m asking you as a person don’t you think this is a bit stupid S1: well <exhales loudly > yes and no S2:

The sequential organisation of this example is thus: (S2) OFFENSIVE (enforce role switch+implicate impolite belief) don’t you think this is a bit stupid (S1) DEFENSIVE (abrogation) well you see I’m just doing a job (S2) OFFENSIVE (implicate impolite belief+reiterate enforced role switch) Yeah well so was Hitler all I’m asking you as a person don’t you think this is a bit stupid (S1) DEFENSIVE (compromise) well < exhales loudly > yes and no Note that the identification of the first offensive strategy ‘enforce role switch’ is only sustainable in the light of his second turn, when he makes the metapragmatic comment I’m asking you as a person to clarify his intent. In S1’s final utterance, the hedge well and the exhalation signal that agreement (the preferred response) is not forthcoming, and the following yes and no attempts to negotiate a compromise. The interaction is interrupted at this point. The vast majority of the interactions in our data follow that basic OFFENSIVE– DEFENSIVE pattern, and so we will not illustrate it further. However, we will draw attention to some other specific defensive strategies that occurred in our data.

1566

J. Culpeper et al. / Journal of Pragmatics 35 (2003) 1545–1579

Another type is to opt out on record. This strategy attempts to ‘seal off’ the FTA. In the following example, S2 (an adjudicator) refuses the appeal of S1 (a car owner) on the basis that he ‘prefers the council’s evidence’, despite S1’s claims that the traffic warden who ticketed him (this woman in the extract) has falsified her records: [15] S1: S2:

so you’re quite happy that this woman perjured herself in writing I’m not saying

S1: so you are happy S2: anything more Mr Langarth and that is the end of the proceeding I’m not S1: you’re happy that she perjured herself S2: saying anything I I’m not making any further comment S1: S2: thankyou .. If you’ll be kind enough to wait in the foyer

A further type of defensive strategy is insincere agreement, which manages the impoliteness act by allowing its speaker to ‘let off steam’. A variant of this strategy is to express surface agreement with the face attack. In this example, S1 (a car owner) has stopped his car to heckle S2 (a traffic warden), who has just ticketed him for illegal parking. [16] S1: One day you’re going to get a lot of problems .. and you create the problems . we don’t S2: S1: create the problems . ban the bloody cars all over London and we have no worries S2: yeah S1: you give us authority to put them < indistinct > and then you come and give us a S2: all right all right I agree with you S1: ticket bar the cars take the cars off London do it S2: all right I agree with you sir yeah too many c S1: today not tomorrow don’t give us S2: yup too many cars yeah too many cars on the road S1: this bloody headache all the time S2: yeah too many cars .. anyway you gonna shift

Note the lack of synchronisation between S1’s accusations and S2’s responses, particularly towards the end where the repeated yeah too many cars does not seem to address S1’s immediately prior discourse. This betrays the fact that S2 is on ‘automatic pilot’, simply allowing S1 to let off steam. Anyway, towards the end of the extract, marks S2’s shift back to serious business.

J. Culpeper et al. / Journal of Pragmatics 35 (2003) 1545–1579

1567

Yet another defensive strategy, though one which has some similarities with the previous strategy, is to ignore the implied face attack. This is particularly clear in cases of sarcasm, where the surface meaning of the utterance is accepted, rather than the implied sarcastic barb. In example [11] above, a van owner, who has just been clamped, says have a good day, to which the clamper replies I will do. Clearly, the clamper complies only with the surface meaning of the van owner’s sarcastic utterance. This defensive counter strategy may also be offensive at the same time, given that it is such a blatant misunderstanding of the van owner’s meaning. This is not a rare occurrence: counter strategies which operate both offensively and defensively with respect to face occurred elsewhere in our data, as our final example in this section will show. In our final example in this section, we aim to show a much more complex case than those we have discussed hitherto. S1’s car has been towed to the car pound, after it was clamped for being illegally parked. Previously, S1 refused to pay the release fee, stating that he preferred to spend his money on a new vehicle. He has returned to the pound to take his belongings out of the car. As he leaves, he strongly threatens S2 (a clamper):

[17] S1: touch my fucking new car and I’ll bust your fucking head off yeah S2: jackanory yeah S1: yeah you fucking shit just cos you’ve got a fat belly anyway S2: yeah full of it aren’t you eh S1: S2: eh it’d fit in your mouth quite easily wouldn’t it

On the face of it, S2’s response, jackanory, looks like a defensive strategy, which might be formulated: ‘dismiss: make light of face damage, joke’. Jackanory is an intertextual reference to a children’s television programme in which fictious stories were read out. S2 implies that S1’s threat is the stuff of children and fiction—it carries no weight. Whilst this strategy does dismiss the impoliteness act, it also belittles the speaker’s attempt in producing it. ‘Belittle the other’ is in fact one of the negative impoliteness strategies suggested in Culpeper (1996: 358). So, simultaneously S2 appears to be offensive. Clearly then, this does not fit the pairs suggested by either Labov (1972) or Harris et al. (1986). Looking at S2’s responses full of it aren’t you eh and it’d fit in your mouth quite easily wouldn’t it, the situation becomes even more complex, since we begin to see a pattern characteristic of ‘mock’ impoliteness or ritualistic banter. Labov (e.g. 1972: 340) points out that strong ritualistic insults are outrageously bizarre and obviously untrue, and Eider (1990: 75) suggests that in ritual insulting participants ‘continually challenge each other to come up with a better, more clever response to an accusation or insult’. These aspects fit S2’s responses: he engages in mock impoliteness. On the other hand, S1’s impoliteness is not obviously untrue (the target, S2, appears to be slightly overweight) or clever: he engages in ‘genuine’ impoliteness. Contrary to Labov (1972), who suggested it was

1568

J. Culpeper et al. / Journal of Pragmatics 35 (2003) 1545–1579

important that all parties considered the insults to be untrue in ritualistic banter, Kochman (1983) (and see Eider, 1990, for further evidence) argues that the key to doing ritualistic banter lies in taking insults—even true, personal insults—in a non-serious way. In our example, we would argue that S2 practices a more global defensive strategy than the others we have discussed. His utterances constitute a particular ‘activity type’ (Levinson, 1992), that of ritualistic banter, and in so doing reduce the impact of S1’s impoliteness. At the same time, S2 shows off his considerable linguistic dexterity. Finally, we should note one contextual factor that may have facilitated the occurrence of this particular interaction, namely, that S1 is walking away from S2. Most of the talk occurs when they are several metres apart—clearly, a safer situation for more risky communicative strategies. We conclude this section with two brief points. First, it should be remembered that in other discourses other impoliteness patterns may emerge, as well as other defensive strategies. Second, one area that we have neglected, but would obviously benefit from research, is how these confrontational encounters are resolved (see Vuchinich, 1990 and Hutchby, 1996: chapter 6, for some first steps).

5. Prosody and impoliteness 5.1. Prosody: its components and functions No utterance can be spoken without prosody, and it is therefore desirable at some point to include this dimension of speech in pragmatic analysis. So far our discussions of impoliteness strategies have referred solely to the (actions and) words of participants as recorded in the orthographic transcription of our data. In this section we would like to consider the contribution not of what was said but of how it was said. The importance of this aspect with respect to impoliteness had been made very clear to us at conference presentations of many of the examples cited in this paper: members of the audience who had not heard the examples usually thought them less impolite than we did. However, before exploring the ways in which prosody can convey impoliteness, it is perhaps useful to consider the resources available. The components of prosody include pitch (intonation), loudness, speed and voice quality. All are important, both separately and in their complex interaction, but we will focus here mainly on features of pitch or intonation, and briefly on loudness. Pitch operates as part of a conventionalised linguistic system (intonational phonology), and also paralinguistically to express, for example, speaker emotions. The two are not easy to disentangle (see Ladd, 1996), since the same channel is used for both. In general, however, we find that local pitch changes tend to have a grammatical or discoursal function (e.g. highlighting sentence accent or topic shift), while pitch changes that affect a whole utterance or string of utterances are more likely to convey paralinguistic information. The most elusive function of intonation, the ‘attitudinal’ function, is what we would expect to be looking for in the data described here— this is clearly data with ‘attitude’. But this aspect of intonational meaning is notoriously hard to pin down.

J. Culpeper et al. / Journal of Pragmatics 35 (2003) 1545–1579

1569

The meanings expressed through prosody are not propositional but very general. Some are language specific and relate to phonological choice: Cruttenden (1997), for example, suggests an underlying meaning of ‘open’ and ‘closed’ for rising and falling contours respectively.9 Other meanings are more iconic and physiological in origin: for example, overall high and low pitch are physiologically associated with small vs. large, or child vs. adult. This may account for some contextually determined effects of high and low pitch, such as associating high pitch with deference (behaving in a ‘small’ way), and low pitch with assertiveness (behaving in a ‘big’ way). Brown and Levinson (1987) make a similar claim, but mistakenly ascribe local meanings to high and low pitch directly, rather than indirectly via inferences derived from the context. As we will see below, low pitch does not necessarily convey ‘comfort and commiseration’ (1987: 268). What we find is that the individual meanings or ‘attitudes’ conveyed by intonation are arrived at by a complex process of pragmatic interpretation, relating the very abstract underlying meanings of intonation to both the message and the context in which it is uttered (see Wichmann, 2000b; Wichmann and Cauldwell, 2001). We will attempt to clarify these relationships for each of the examples below. The examples for this section are taken for the most part from an appeals tribunal. We chose to focus on these sections for two reasons: first, they contain a range of different ways in which prosody can contribute to interpersonal meaning, and second, the quality of the sound recording, unlike the noisy street encounters, allows corroborative instrumental analysis of the pitch data. Our post hoc impressions of the one example taken from a street scene are corroborated by participant response. The prosodic transcription of our examples was made auditorily by the third author and checked by the first author. Some disputed contours were subsequently analysed instrumentally to corroborate (or not) impressionistic judgements. The illustrations are intended to make the contours readable for non-specialists, and represent the melody as a continuous line, much as a hearer would perceive it. The physical reality is of course different. A fundamental frequency contour is not continuous but frequently interrupted by the presence of non-sonorant segments in the speech. 5.2. Prosodic strategies for impoliteness In our data we find a number of examples where negative impoliteness strategies are signalled prosodically, including: ‘hinder linguistically’, ‘threaten’, and ‘invade the other’s space’ (cf. Culpeper, 1996). We also find an example of positive impoliteness using the strategy of ‘dissociating from the other’ (cf. Culpeper, 1996). We will consider these examples in detail below, and also draw attention to the patterning of strategies. 5.2.1. Negative impoliteness 5.2.1.1. Hinder linguistically. One of the ways in which speakers can be negatively impolite is to deprive someone of their freedom to speak, for example by interrupting or denying a turn. Intonation has the resources to reinforce or mitigate such acts of 9

This analysis of rises and falls relates to Southern British English, the variety most often spoken in this data.

1570

J. Culpeper et al. / Journal of Pragmatics 35 (2003) 1545–1579

impoliteness. Take the following example which is an earlier part of the conversation presented in [9]. The tribunal officer wishes to draw the hearing to a close against the wishes of his clients, and makes this quite explicit in the following exchange: Client: I’m not finished yet as far as I’m concerned Tribunal officer: well I have I’m finished and I’m running this tribunal and that’s the end of that

The tribunal officer’s turn is divided into four tone groups, and each of these carries the ‘same’ tune, i.e. a realisation of the same underlying phonology: high onset and falling nucleus.10

The falling nucleus (or, in Autosegmental-metrical terms, H*LL% contour11) is typical of the end of a sentence and effectively divides the turn prosodically into four independent utterances or ‘spoken sentences’. The meaning of ‘closure’ associated with this contour conveys a succession of closed, or final, statements, reinforcing the propositional content and the force of each one. This alone does not necessarily constitute impoliteness beyond that implied in the message itself. However, each of the falling nuclei is highly marked. The starting point of a fall (given the same low endpoint) has been shown to convey different degrees of finality (Wichmann, 2000a); the most marked is the fall which begins low in the speaker’s range and usually lower than the preceding syllable. In Autosegmental-metrical terms this is known as a downstepped fall12 (!H*L L%) and conveys a sense of extreme finality or closure. It can be used to suppress normal turn-taking in conversation in order to make temporary ‘conversational space’ for the speaker (Wichmann, 1998), or to bring a whole discourse to an end. The clauses are therefore not just closed but very closed. The prosody thus intensifies the force of the tribunal officer’s words. An additional effect is achieved through repetition to form a parallelism—one of the multiple strategies mentioned in Section 4.1. In the successive repetition of a pitch contour, we have a case of intonational parallelism, which has just as much of an intensifying function as lexical repetition or structural parallelism. The role of 10

The terms ‘onset’ and ‘nucleus’ are used in the British system of intonation. The onset is the first accented syllable in a group and the nucleus is typically the last. In each of these tone groups the underlying phonology is the same; the difference in realisation relates to the differing numbers of syllables onto which it is mapped. 11 In the currently preferred Autosegmental-metrical framework contours are expressed as the interpolation between a series of pitch targets, e.g. high (H) followed by low(L). The * symbol indicates an accented syllable; the% symbol indicates a so-called ‘boundary tone’ - the pitch target at the end of a tone group. 12 The exclamation mark ! indicates a ‘downstepped’ , i.e. lower than predicted, pitch accent. A falling contour transcribed as !H* L thus begins lower than a normal H*L.

J. Culpeper et al. / Journal of Pragmatics 35 (2003) 1545–1579

1571

intonation here is thus twofold—it signals unmitigated finality, and it does so iteratively, thus intensifying further the verbal message. Since ending a conversation is potentially an attack on negative face, we would always expect some minimal expression of politeness in this situation, either in the text or in the prosody. The strategy used here, however, is not simply an absence of mitigation but an emphatic denial of the client’s wishes, and cannot therefore be accounted for by politeness theory. 5.2.1.2. Threaten. The tribunal officer’s attempts to close the hearing meet only with the client’s stubborn refusal to comply. Although the officer had begun the closure sequence in a much more indirect and conciliatory fashion (see the discussion of [2]), he now shifts to a more direct approach: [18] S1: I’ve finished Mr Culp .will you will you will you please leave the room I’m S2: in this situation S3: S1: not answering any further questions . do you want me to press the buzzer S2: S3: well S1: will you please leave the room S2: for what reason S3: that’s being babyish isn’t it S1: I’ve finished I’ve finished .. the hearing S2: are you turning us out for what reason I’m furious S3: give us

Despite the interrogative form, will you please leave the room can only be interpreted as a command, given the use of please and the context (e.g. the relative power of the official). Moreover, the first time it is said the prosody reinforces its commanding force. It carries the same contour as described earlier: high onset and a markedly low (‘downstepped’) final fall which intensifies the sense of finality. The intonation signals that it is not the speaker’s intention to offer an ‘out’ to the hearer, but that the matter is closed.

The client still refuses to leave and the command is repeated. This time the intonation is different: it ends in a very slight rise (confirmed by instrumental analysis).

1572

J. Culpeper et al. / Journal of Pragmatics 35 (2003) 1545–1579

According to Aijmer (1996), a final rise on a request can operate as a mitigating device for more direct (or conventionally indirect) requests, as in:

while if the request is very indirectly expressed, i.e. already mitigated, a falling nucleus appears to be acceptable, as in:

This effect of contour is, of course, again related to the notion of openness and closure. A request which is prosodically ‘open’ can offer the hearer a chance to respond, perhaps to negotiate, but if it is prosodically ‘closed’ it signals that no further discussion is expected, i.e. the ‘matter’ is closed. On the face of it, the tribunal officer’s second will you please leave the room, is mitigated by the final rise in intonation. However, given the fact that the officer has made repeated attempts to close the proceedings, and even threatened to use his ‘buzzer’, presumably to have his clients forcibly removed, it would be perverse to interpret this second request as less forceful, or more polite, than the first. The apparent ‘politeness’ can therefore not be genuine. Knowles (1987) (later echoed by Aijmer 1996) suggests that a final rise may have the effect of downtoning a command, but ‘if there is no possible interpretation other than as a command, the use of the rise can be seen to be not genuine and therefore possibly patronising’ (1987: 192). There may be an element of being patronising here—suggesting that the clients are not able to behave in an ‘adult’ way—and this in itself constitutes a specific impoliteness strategy. However, the use of non-genuine politeness, Culpeper’s (1996) ‘mock politeness’, in this context lends an additional meaning which can be explained by returning again to the abstract meaning of falls and rises. The ‘open’ meaning ascribed to a rise (and also fall-rise) suggests that there is something unsaid, usually interpreted (in the case of requests) as offering the addressee an opportunity to respond, and hence constituting a mitigation of the FTA. It can also mean, however, that the speaker has more to say, and the ‘more to say’ could in this context very well be ‘or else. . .’. The only possible interpretation of the rise in this context is not as politeness but as an (insincerely veiled) threat. 5.2.1.3. Invade auditory space. In the exchanges on the street between traffic wardens and car owners, emotions often run high, and voices are raised. The ‘raising’ of a voice normally involves both an increase in pitch and loudness and can have both internal and external motivation. It may be motivated externally by the need to communicate across a distance, or internally by the presence of extreme emotions, both positive and negative. We are very sensitive to ‘appropriate’ levels of loudness in everyday interaction: speech that is louder than physical distance warrants is

1573

J. Culpeper et al. / Journal of Pragmatics 35 (2003) 1545–1579

perceived negatively and as an invasion of our auditory space. This lies behind the common response to overheard conversations on mobile phones in public places, where speakers tend to be much less sensitive to encroaching on the auditory space of others than when talking to people who are physically present. These transgressions are, however, for the most part unintentional. Where it is clearly intentional it is likely to be interpreted as a deliberate attack on negative face. We choose the following example to illustrate this, because of the independent participant evidence (there’s no point in shouting), since the background street noise rendered instrumental analysis of the relative pitch and amplitude impossible (the capitals indicate loud speech):

[19]

S1: S4:

I’m afraid you have to have a voucher to park in this area HOW ARE YOU

S1: S4:

SUPPOSED TO EARN A BLEEDING LIVING IN THIS FUCKING

S1: S4:

COUNTRY EH

well there’s no point in shouting at me yeah I never asked

S1: S4:

you to park here yeah I’m simply trying to help you if you don’t want me

S1: S4:

to help you then I don’t need to help you

There is ample evidence here that the invasion of space is motivated by anger and not distance. The prosody (raised pitch and loudness) combines with the expressions of impoliteness already present in the utterance (the challenge, how are you supposed to [. . .], and the use of taboo words, bleeding, fucking) to intensify the impoliteness. Jay (e.g. 1992: 97, 108) claims, rightly in our view, that a person who shouts in anger is not only invading the space of the hearer, but making sure that the hearer is aware of his or her anger. It is not just self-expression but a strategy to make someone else feel responsible for the speaker’s extreme state. The invasion of auditory space alone—if it is intentional—constitutes an example of negative impoliteness. If it occurs in combination with insults and other strategies, the effect is compounded. 5.2.2. Positive impoliteness 5.2.2.1. Deny common ground or disassociate from the other. One of the impoliteness strategies we have posited is to deny common ground or to dissociate from the other speaker, and in so doing attack positive face (cf. Culpeper, 1996). This can be expressed prosodically by modifying overall pitch range (and covarying loudness). Accommodation of pitch range between participants has already been shown to be a feature of ‘compliant’ interaction. Brazil (1985), for example, claims that in compliant behaviour speakers use relative pitch concord across turns. This means that

1574

J. Culpeper et al. / Journal of Pragmatics 35 (2003) 1545–1579

the current speaker adopts the same ‘key’ (high, mid or low relative to his or her own voice) as that used at the end of the last speaker’s turn. This is illustrated in the (invented) examples (a) and (b) below:

A similar claim is made by Couper-Kuhlen (1996) who shows, in a study of repetitions, that when speakers repeat another speaker’s words at the same relative pitch it is perceived as supportive. Absolute pitch matching, on the other hand, i.e. imitating the pitch of the other exactly and not just relative to one’s own voice, is perceived as mimicry and an attempt to distance oneself from the other. We would like to suggest that not only mimicry (hyper-accommodation), but also a simple failure to accommodate can be a feature of non-compliant behaviour. Pitch concord is a signal of prosodic ‘common ground’, and by denying that concord a speaker is denying common ground or disassociating from the interlocutor. Thus pitch range can be exploited strategically by speakers to signal both closeness and distance. We find several examples of prosodic disassociation in the data. Here is an example from the appeals tribunal where the client does not accept the officer’s ruling; the official begins his response at the same (relative) pitch as the client, but he checks himself after only one syllable—the word I—and immediately shifts back to the low register he has used throughout the hearing (the arrow indicates a sudden shift in pitch level):

In these confrontational interactions it is presumably important that the arbitrator does not enter into an argument with the clients, but merely hears their account and then passes judgement. We see this ‘refusal to argue’ reflected in the refusal to accommodate prosodically to the client. If clients are annoyed or angry, as happens in these clips when their appeal is turned down, they raise their voices. It is striking that even when the interaction becomes affectively charged, the official responds almost consistently with a quiet, low pitched voice. One might say that the gauntlet of battle has been thrown down by the client, but the tribunal officer refuses to pick it up. The brief raising of the voice in the example above is a momentary lapse. In the next example, the raised voice is the result of having to speak over an interruption (which co-occurs with the text in square brackets) and the voice shifts down again at the end of the clause, when he has regained the floor.

J. Culpeper et al. / Journal of Pragmatics 35 (2003) 1545–1579

1575

Since the last clause (do you want me to press the buzzer) constitutes a threat, this shift down in pitch is highly marked. One normally associates a threat with increased emotional involvement, and that in turn usually involves a widening of pitch range. A narrowing of range in this situation may suggest to the listener a degree of restraint, which is often more threatening than uncontrolled anger. In addition to the positive impoliteness of prosodically denying common ground, we also have a verbal threat intensified by low pitch, constituting negative impoliteness. Brown and Levinson’s claim, therefore, referred to earlier, that low pitch conveys ‘comfort and commiseration’ is clearly not valid here. 5.3. The contribution of prosody to the expression of impoliteness There appear to be three main ways in which speakers exploit the ‘attitudinal’ role of intonation in this interaction to express impoliteness. First, the force of a speech act is related to the choice of pitch contour. We have given as an example two ways of uttering a command, one reinforcing its commanding force, the other turning it into a (insincerely veiled) threat. Second, there are related discoursal issues; for example, signalling whether something is ‘open’ (non final), or ‘closed’ (final)—exploited here to bring a conversation to an end and thereby block the hearers’ wishes. We can account for these meanings largely in phonological terms—in terms of choice of intonational category, such as rising vs. falling tone. Third, there are global prosodic parameters—matters of phonetic realisation rather than phonological choice—such as high pitch and extreme loudness over an utterance or series of utterances, which can be seen as an invasion of auditory space, or the strategic denial of pitch concord, which can be seen as a prosodic means of increasing the distance between interlocutors.

6. Conclusion and further directions Thomas (1995: 179) concludes that the claim for politeness theories is that ‘people employ certain strategies [. . .] for reasons of expediency—experience has taught us that particular strategies are likely to succeed in given circumstances, so we use them’. This, in our view, is also true of impoliteness. In some circumstances, such as the car parking disputes discussed here, people choose to use impoliteness to expedite their goals. In so doing, they orientate to the hearer’s face in a fundamentally different way from politeness: rather than supporting it, they attack it. Our starting point for this paper has been that politeness theory (even bald on record politeness) does not adequately describe communicative strategies designed to attack face. We

1576

J. Culpeper et al. / Journal of Pragmatics 35 (2003) 1545–1579

have argued that most of the impoliteness strategies identified in Culpeper (1996) also occur in car parking disputes: there are regular ways in which people attack face. Of course, this is not to say that even a frequent impoliteness strategy will be impolite in all contexts. What the underlying dimensions of these impoliteness strategies are is little understood. What is clear is that an impoliteness framework is not simply a mirrorimage of a politeness framework, such as Brown and Levinson’s (1987). How one orders strategies for degree of impoliteness is not known. As with politeness, there appears to be no simple correlation with directness. Tabooness, for example, relies on conventionalisation, which may short-circuit indirectness. A particular problem we inherited from Brown and Levinson (1987), and one that is becoming increasingly well-known, is the distinction between positive and negative face. For example, it is clear that a negative impoliteness strategy (e.g. blocking their conversational path) might work primarily by impeding the hearer’s freedom (an issue of negative face), but also has secondary implications for positive face (e.g. the speaker is not interested in the hearer’s views). And it is possible that those ‘secondary’ positive face implications may, in some contexts, cause more face damage than that done to negative face. Clearly, much research remains to be done. As for developing a theory which can encompass both politeness and impoliteness, we would suggest that the recent work by Spencer-Oatey (2000) on ‘rapport management’ offers the most promising way forward, since it has sufficient sophistication to accommodate both, yet is also supported by solid empirical work (e.g. Spencer-Oatey, 1993, 2002). A major objective of this paper has been to broaden the scope of previous impoliteness studies, which, like most politeness studies, have tended to focus narrowly on single lexically or grammatically-based strategies. If intention is a factor that distinguishes impoliteness from politeness, as we argue, then we need a richer understanding of an interactant’s behaviour, including their prosody, and of the discoursal context, in order to infer intentions. Hence, we examined patterns of impoliteness in and across exchanges. Interactants regularly employ the global strategies of repeating and/or combining impoliteness strategies within a participant’s turn or turns. We considered response options to impoliteness, and, in particular, the enactment of two pairs: OFFENSIVE–DEFENSIVE and OFFENSIVE– OFFENSIVE. It is the response to an utterance, and indeed the construction of the whole speech activity, that may determine how that utterance is to be taken, including whether it be polite, impolite or something in between. Note here that our approach, by analysing the reactions of interactants to impoliteness, begins to shift away from the sole emphasis on speaker production that characterizes so much of politeness theory and move towards incorporating the hearer’s perception. Grimshaw (1990) points out that ‘participants in conflict talk have the same resources available for that interaction as do all conversationalists’ (1990: 10). We have shown here that this extends to prosodic resources, and we have been able to show some of the ways these resources are used to express impoliteness. It is sometimes the prosody that makes an utterance impolite—giving truth to the common view that the offence lay in how something was said rather than what was said. Specifically, we have discussed the role of prosody in relation to illocutionary force and to dis-

J. Culpeper et al. / Journal of Pragmatics 35 (2003) 1545–1579

1577

coursal issues, and we have commented on the (lack of) prosodic concord between participants. Of course, both the prosodic resources and the abstract meanings they convey are relatively limited. Ladd (1996) observes that intonational meanings are ‘very general, but they are part of a system with a rich interpretative pragmatics, which gives rise to very specific and often quite vivid nuances in specific contexts’ (1996: 39– 40). If we are to find a way of identifying the contextual factors which give rise to such specific meanings, there is still much to done in the study of intonation and discourse. This is true whether we are concerned with politeness or impoliteness. Both are deliberate strategies on the part of the speaker to exploit the shared conventions of talk. In the course of our work, we have noted numerous areas in need of further research. We conclude by briefly listing some of these: (1) The perceived offensiveness of impoliteness strategies needs to be investigated. (2) Clearly, our findings in this paper relate to car parking disputes. Different phenomena and patterns remain to be found in other discourse types. (3) We have largely ignored what occurs prior to an impoliteness act, and how confrontational episodes are resolved. (4) There is much more to be said about turn-taking and about non-verbal aspects.

References Aijmer, Karin, 1996. Conversational Routines in English. Longman, London. Andersson, Lynne, Pearson, Christine, 1999. Tit for tat? The spiraling effect of incivility in the workplace. Academy of Management Review 24 (3), 452–471. Aronson, Karin, Rundstro¨m, Bengt, 1989. Cats, dogs, and sweets in the clinical negotiation of reality: on politeness and coherence in pedriatric discourse. Language in Society 18, 483–504. Austin, Paddy, 1990. Politeness revisited—the dark side. In: Bell, Allen, Holmes, Janet (Eds.), New Zealand Ways of Speaking English (Multilingual Matters 65). Multilingual Matters, Philadelphia, pp. 277–293. Beebe, Leslie M., 1995. Polite fictions: instrumental rudeness as pragmatic competence. In: Linguistics and the Education of Language Teachers: Ethnolinguistic, Psycholinguistics and Sociolinguistic Aspects. Georgetown University Round Table on Languages and Linguistics. Georgetown University Press, Georgetown, pp. 154–168. Blum-Kulka, Shoshana, 1987. Indirectness and politeness in requests: same or different? Journal of Pragmatics 11, 131–146. Bousfield, Derek, 1999. ‘‘They Need Strangling!’’ Impoliteness in the BBC Television Series ‘The Clampers’. Unpublished MA dissertation, Lancaster University, UK. Bousfield, Derek. Impoliteness in interaction. Unpublished doctoral thesis, Lancaster University, UK (forthcoming). Brazil, David, 1985. Phonology: intonation in discourse. In: Teun van, Dijk (Ed.), Handbook of Discourse Analysis, Vol. 2. Dimensions of Discourse. Academic Press, London, pp. 57–75. Brenneis, Donald, Lein, Laura, 1977. ‘You fruithead’: a sociolinguistic approach to children’s dispute settlement. In: Ervin-Tripp, Susan, Mitchell-Kernan, Claudia (Eds.), Child Discourse. Academic Press, New York, pp. 49–65. Brown, Penelope, Levinson, Stephen C., 1987. Politeness: Some Universals in Language Usage. Cambridge University Press, Cambridge. Corsaro, William A., Rizzo, Thomas A., 1990. Disputes in the peer culture of American and Italian nursery-school children. In: Grimshaw, Allen D. (Ed.), Conflict Talk: Sociolinguistic Investigations of Arguments and Conversations. Cambridge University Press, Cambridge, pp. 21–66. Couper-Kuhlen, Elizabeth, 1996. The prosody of repetition: on quoting and mimicry. In: Couper-Kuhlen, Elizabeth, Selting, Margaret (Eds.), Prosody in Conversation. Cambridge University Press, Cambridge, pp. 366–405.

1578

J. Culpeper et al. / Journal of Pragmatics 35 (2003) 1545–1579

Craig, Robert, Tracy, Karen, Spisak, Frances, 1986. The discourse of requests: assessment of a politeness approach. Human Communication Research 12, 437–468. Cruttenden, Alan, 1997. Intonation, second ed. Cambridge University Press, Cambridge. Culpeper, Jonathan, 1996. Towards an anatomy of impoliteness. Journal of Pragmatics 25, 349–367. Culpeper, Jonathan, 1998. (Im)politeness in drama. In: Culpeper, Jonathan, Short, Mick, Verdonk, Peter (Eds.), Studying Drama: From Text to Context. Routledge, London, pp. 83–95. Eder, Donna, 1990. Serious and playful disputes: variation in conflict talk among female adolescents. In: Grimshaw, Allen D. (Ed.), Conflict Talk: Sociolinguistic Investigations of Arguments and Conversations. Cambridge University Press, Cambridge, pp. 67–84. Eelen, Gino, 1999. Ideology in Politeness: A Critical Analysis. Unpublished doctoral thesis, Antwerp. Eelen, Gino, 2001. A critique of politeness theories. St. Jerome Publishing, Manchester. Fraser, Bruce, 1990. Perspectives on politeness. Journal of Pragmatics 14, 219–236. Goffman, Erving, 1967. Interaction Ritual. Aldine Publishing, Chicago. Goodwin, Charles, Goodwin, Marjorie Harness, 1990. Interstitial argument. In: Grimshaw, Allen D. (Ed.), Conflict Talk: Sociolinguistic Investigations of Arguments and Conversations. Cambridge University Press, Cambridge, pp. 85–117. Greenberg, Bradley S., 1976. The effects of language intensity modification on perceived verbal aggressiveness. Communication Monographs 43, 130–139. Grice, H.Paul, 1975. Logic and conversation. In: Cole, Peter, Morgan, Jerry (Eds.), Syntax and Semantics 3: Speech Acts. Academic Press, New York, pp. 41–58. Grimshaw, Allen D. (Ed.), 1990. Conflict Talk: Sociolinguistic Investigations of Arguments and Conversations. Cambridge University Press, Cambridge. Harris, Linda, Gergen, Kenneth, Lannaman, John, 1986. Aggression rituals. Communication Monographs 53, 252–265. Harris, Sandra, 2001. Being politically impolite: extending politeness theory to adversarial political discourse. Discourse and Society 12 (4), 451–472. Hayashi, Takuo, 1996. Politeness in conflict management: a conversation analysis of dispreferred message from a cognitive perspective. Journal of Pragmatics 25, 227–255. Holmes, Janet, 1984. Modifying Illocutionary Force. Journal of Pragmatics 8, 345–365. Hutchby, Ian, 1996. Confrontation Talk: Arguments, Asymmetries, and Power on Talk Radio. Lawrence Erlbaum Associates, Mahwah, NJ. Jay, William, 1992. Cursing in America: A Psycholinguistic Study of Dirty Language in the Courts, in the Movies, in the Schoolyards and on the Streets. John Benjamins, Philadelphia. Jay, William, 2000. Why we Curse: A Neuro-psycho-social Theory of Speech. John Benjamins, Philadelphia and Amsterdam. Kienpointner, Manfred, 1997. Varieties of rudeness: types and functions of impolite utterances. Functions of Language 4 (2), 251–287. Knowles, Gerald, 1987. Patterns of spoken English. Longman, London. Kochman, Thomas, 1983. The boundary between play and nonplay in black verbal duelling. Language in Society 12, 329–337. Labov, William, 1972. Language in the Inner City: Studies in the Black English Vernacular. Blackwell, Oxford. Labov, William, Fanshel, David, 1977. Therapeutic Discourse: Psychotherapy as Conversation. Academic Press, New York. Lachenicht, L.G., 1980. Aggravating language: a study of abusive and insulting language. International Journal of Human Communication 13 (4), 607–688. Ladd, D. Robin, 1996. Intonational Phonology. Cambridge University Press, Cambridge. Lakoff, Robin, 1989. The limits of politeness. Multilingua 8, 101–129. Leech, Geoffrey N., 1983. Principles of Pragmatics. Longman, London. Lein, Laura, Brenneis, Donald, 1978. Children’s disputes in three speech communities. Language in Society 7, 299–323. Levinson, Stephen C., [1979] 1992. Activity types and language. In: Drew, Paul, Heritage, John (Eds.), Talk at Work. Cambridge University Press, Cambridge, pp. 66–100. Mehan, Hugh, 1990. Rules versus relationships in small claims disputes. In: Grimshaw, Allen D. (Ed.),

J. Culpeper et al. / Journal of Pragmatics 35 (2003) 1545–1579

1579

Conflict Talk: Sociolinguistic Investigations of Arguments and Conversations. Cambridge University Press, Cambridge, pp. 160–177. Penman, Robyn, 1990. Facework and politeness: multiple goals in courtroom discourse. Journal of Language and Social Psychology 9, 15–38. Quirk, Randolph, Greenbaum, Sidney, Leech, Geoffrey, Svartvik, Jan, 1985. A Comprehensive Grammar of the English Language. Longman, London and New York. Spencer-Oatey, Helen, 1993. Conceptions of social relations and pragmatics research. Journal of Pragmatics 20, 27–47. Spencer-Oatey, Helen, 2000. Rapport management: a framework for analysis. In: Spencer-Oatey, H. (Ed.), Culturally Speaking: Managing Rapport Through Talk Across Cultures. Continuum, London and New York, pp. 11–46. Spencer-Oatey, Helen, 2002. Managing rapport in talk: using rapport sensitive incidents to explore the motivational concerns underlying the management of relations. Journal of Pragmatics 34, 529– 545. Tannen, Deborah, 1990. Silence as conflict management in fiction and drama: Pinter’s Betrayal and a short story ‘‘Great Wits’’. In: Grimshaw, Allen D. (Ed.), Conflict Talk: Sociolinguistic Investigations of Arguments and Conversations. Cambridge University Press, Cambridge, pp. 260–279. Thomas, Jenny, 1995. Meaning in Interaction: An introduction to Pragmatics. Longman, London and New York. Tracy, Karen, 1990. The many faces of facework. In: Giles, Howard, Robinson, William P. (Eds.), Handbook of Language and Social Psychology. Wiley, Chichester, pp. 209–226. Turner, Ken, 1996. The principal principles of pragmatic inference: politeness. Language Teaching 29 (1), 1–13. Vuchinich, Samuel, 1990. The sequential organization of closing in verbal family conflict. In: Grimshaw, Allen D. (Ed.), Conflict Talk: Sociolinguistic Investigations of Arguments and Conversations. Cambridge University Press, Cambridge, pp. 118–138. Watts, Richard J., 1992. Linguistic politeness and politic verbal behaviour. In: Watts, Richard J., Ide, Sachiko, Ehlich, Konrad (Eds.), Politeness in Language: Studies in its History, Theory and Practice. Mouton de Gruyter, Berlin and New York, pp. 43–69. Wichmann, Anne, 1998. Using intonation to create conversational space: projecting topics and turns. In: Renouf, Annette (Ed.), Explorations in Corpus Linguistics. Rodopi, Amsterdam, pp. 217–232. Wichmann, Anne, 2000a. Intonation in Text and Discourse. Pearson Education, London. Wichmann, Anne, 2000b. ‘The attitudinal effects of prosody, and how they relate to emotion.’ In: Proceedings, ISCA Workshop on Speech and Emotion. Newcastle, N. Ireland. Wichmann, Anne, Cauldwell, Richard, 2001. Wh-questions and Attitude: The Effect of Context. Presented at CL2001 International Corpus Linguistics Conference, Lancaster. Derek Bousfield is a PhD student and part-time teaching assistant in the Department of Linguistics and Modern English Language at Lancaster University, UK. His research topic is centrally concerned with politeness theory, the title of which is Impoliteness in Interaction. His other interests include the pragmaticstylistic interface, with special focus on how this operates in dramatic texts. He is also co-convenor of the Pragmatics and Stylistics Research Group at Lancaster University. Jonathan Culpeper is a senior lecturer in the Department of Linguistics and Modern English Language at Lancaster University, UK. Much of his research focuses on ‘spoken’ interaction as it appears in writing. His particular interests are within politeness theory, historical pragmatics, and stylistics. His most recent major publication is Language and Characterisation: People in Plays and other Texts (2001). Anne Wichmann is a Reader in Speech and Language at the University of Central Lancashire, Preston, UK. Her research is concerned with the prosody of discourse and conversation. This is reflected in her major publication Intonation in Text and Discourse (2000). She works chiefly on intonation, and currently focusses on the analysis of ‘attitudinal’ intonation within a pragmatic framework.