Slurs

Adam Croom

Slurs possess interesting linguistic properties and so have recently attracted the attention of linguists and philosophers of language. For instance the racial slur "nigger" is explosively derogatory, enough so that just hearing it mentioned can leave one feeling as if they have been made complicit in a morally atrocious act. (Jennifer Hornsby has suggested that slurs might count as “hate speech” and so raise questions “about the compatibility of the regulation of [hate] speech with principles of free speech” (2001, p. 129). Indeed, the very taboo nature of these words makes discussion of them typically prohibited or frowned upon. Although it is true that the utterance of slurs is illegitimate and derogatory in most contexts, sufficient evidence suggests that slurs are not always or exclusively used to derogate. In fact, slurs are frequently picked up and appropriated by the very in-group members that the slur was originally intended to target. This might be done, for instance, as a means for like speakers to strengthen in-group solidarity. So an investigation into the meaning and use of slurs can give us crucial insight into how words can be used with such derogatory impact, and how they can be turned around and appropriated as vehicles of rapport in certain contexts among in-group speakers. In this essay I will argue that slurs are best characterized as being of a mixed descriptive/expressive type. Next, I will review the most influential accounts of slurs offered thus far, explain their shortcomings, then provide a new analysis of slurs and explain in what ways it is superior to others. Finally, I suggest that a family-resemblance conception of category membership can help us achieve a clearer understanding of the various ways in which slurs, for better or worse, are actually put to use in natural language discourse. https://doi.org/10.1016/j.langsci.2010.11.005 https://doctorcroom.com/croom-2011

Language Sciences xxx (2010) xxx–xxx Contents lists available at ScienceDirect Language Sciences journal homepage: www.elsevier.com/locate/langsci Slurs Adam M. Croom ⇑ Department of Linguistics, University of Pennsylvania, 619 Williams Hall, Philadelphia, PA 19104, USA Department of Philosophy, University of Pennsylvania, 433 Cohen Hall, Philadelphia, PA 19104, USA Institute for Research in Cognitive Science, University of Pennsylvania, 3401 Walnut Street Suite 400A, Philadelphia, PA 19104, USA a r t i c l e i n f o Article history: Received 13 June 2010 Received in revised form 10 November 2010 Accepted 11 November 2010 Available online xxxx Keywords: Euphemisms Philosophy of language Politeness Pragmatics Semantics a b s t r a c t Slurs possess interesting linguistic properties and so have recentlyattracted the attention of linguists and philosophers of language. For instance the racial slur nigger is explosively derogatory, enough so that just hearing it mentioned can leave one feeling as if they have been made complicit in a morally atrocious act. (Jennifer Hornsby has suggested that slurs might count as ‘‘hate speech’’ and so raise questions ‘‘about the compatibility of the regulation of [hate] speech with principles of free speech’’ (2001, p. 129). Chris Hom further suggests that, ‘‘the use of an epithet may count as a literal threat, and hence no longer merit freedom of speech protection under the First Amendment’’ (2008, p. 440). A close analysis of slurs is clearly required before we can make informed decisions about this serious issue.) Indeed, the very taboo nature of these words makes discussion of them typically prohibited or frowned upon. Although it is true that the utterance of slurs is illegitimate and derogatory in most contexts, sufficient evidence suggests that slurs are not always or exclusively used to derogate. In fact, slurs are frequently picked up and appropriated by the very ingroup members that the slur was originally intended to target. This might be done, for instance, as a means for like speakers to strengthen in-group solidarity. So an investigation into the meaning and use of slurs can give us crucial insight into how words can be used with such derogatory impact, and how they can be turned around and appropriated as vehicles of rapport in certain contexts among in-group speakers. In this essay I will argue that slurs are best characterized as being of a mixed descriptive/expressive type. Next, I will review the most influential accounts of slurs offered thus far, explain their shortcomings, then provide a new analysis of slurs and explain in what ways it is superior to others. Finally, I suggest that a family-resemblance conception of category membership can help us achieve a clearer understanding of the various ways in which slurs, for better or worse, are actually put to use in natural language discourse (note that in this article slurs will be mentioned but not used. Although I have considered not even mentioning such a derogatory term as nigger in the first place, I chose it because on the one hand there is a substantive literature on the term upon which to draw to aid in the analysis of slurs in general, and on the other hand, this term highlights the fact that slurs possess a forcefully potent affective component that is clearly a key aspect to their employment). Ó 2010 Elsevier Ltd. All rights reserved. 1. What are slurs? A slur is ‘‘a disparaging remark or a slight’’ that is usually used to ‘‘deprecate’’ certain targeted members. Utterances of slurs are usually explosively derogatory acts, and different slurs derogate members of different classes. For instance, racial ⇑ Address: 3600 Chestnut Street MB 1209, Philadelphia, PA 19104-6106, USA. Tel.: +1 215 898 6046; fax: +1 215 573 2091. E-mail address: adam.m.croom@gmail.com 0388-0001/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.langsci.2010.11.005 Please cite this article in press as: Croom, A.M. Slurs. Lang. Sci. (2010), doi:10.1016/j.langsci.2010.11.005 2 A.M. Croom / Language Sciences xxx (2010) xxx–xxx slurs are ‘‘derogatory or disrespectful nickname[s] for a racial group’’ and sexist slurs are ‘‘term[s] of disparagement used to refer to members of a given sexual minority, gender, sex, or sexual orientation in a derogatory or pejorative manner.1’’ Although different slurs target members of different groups, slurs are in general derogatory terms that target members of a certain class or group. This will at least for the moment serve as a working definition, for I will provide a more nuanced characterization of what slurs are in later sections of this essay. 2. Slurs and semantic types Some scholars offering analyses of slurs do so through a discussion of semantic types, attempting to characterize slurs as members belonging to one semantic type or another. Others offer analyses of slurs without postulating a theory of semantic types (recently, most linguists have taken the former approach whereas most philosophers of language have taken the later). In this section, the discussion will first be situated within the context of linguists attempting to analyze slurs in terms of semantic types. In Section 5 and later, I will re-focus the discussion within the context of philosophers of language attempting to analyze slurs otherwise. In Section 6 I’ll provide my own analysis, and finally in Section 7, I will suggest that my analysis can be combined with a family-resemblance conception of category membership that is explanatorily powerful and true to the facts of the various empirical uses of slurs (although my account does not require that all concepts be restricted to a family-resemblance account). Christopher Potts is one notable linguist that has done a great service to linguists and philosophers of language. For Potts has brought non-descriptive meaning, which had often been neglected, under the purview of rigorous systematic analysis, motivating much fruitful discussion on non-descriptive terms (such as expressive terms). Potts explains the difference between descriptive and non-descriptive terms via an analysis of semantic types, and he has accordingly attempted to analyze slurs in terms of semantic types. As Potts suggests, we can partition linguistic phenomena by ‘‘dividing the class of semantic types into descriptive types and expressive types’’ (2007, p. 168). Examples of descriptive types are dog, woman, and African American. Examples of expressive types are damn, bastard, and fuck. Speakers generally produce sentences involving descriptive types in order to pick out descriptive items. In contrast, it is thought that speakers generally produce sentences involving expressive types, not to pick out or describe items, but rather in order to express their heightened emotional state (Jay and Janschewitz, 2008; Potts, 2007; Jay, 1992, 2000; Jay and Danks, 1977; Haverkate, 1990; Ayer, 1952). These two points are suggested in the examples below: (1) (2) Rover is a skinny dog. Rover is a skinny fucker. When it comes to informing a hearer about a state of affairs or describing how the world is (1) succeeds in communicating descriptive or at-issue content in a way that (2) does not. In (1) the speaker uses the descriptive dog in order to pick out a descriptive item in the world, a dog. But since expressive types such as fucker do not seem to communicate content in the way that descriptive types such as dog do, Potts (2007) has suggested that ‘‘expressive content is not propositional, that it is distinct from the meanings we typically assign to sentences’’ (p. 177) and that their contents are not usually available for articulation in descriptive terms. That is, ‘‘expressives in general manifest this descriptive ineffability’’ (p. 176). Since such content is not descriptive and is thought to be expressive instead, the idea of Potts (2008, 2007) and others is that in cases like (2) the speaker is presumably using the expressive fucker to express their heightened emotional state.2 As Potts writes about expressives: As speakers we have strong expectations that uses of [expressives such as] damn will correlate with the speaker’s being in a heightened emotional state (or wishing to create that impression). In turn, we use it only when we are in such a state (or wish to create that impression). The total effect of these assumptions is that [an expressive such as] damn is a reliable signal of emotionality. Knowing its use conditions [. . .] largely involves being attuned to this information.3 (2008, p. 13) Because of the difference in linguistic behavior between expressives and descriptives, some scholars such as Potts (2008, 2007) have suggested that expressive types are fundamentally distinct from descriptive types such that for every semantic type it is either descriptive or expressive but not both (Potts, 2007; Kratzer, 1999).4 More recently, Timothy Williamson 1 The definition of ‘‘slur’’ comes from Random House Dictionary (2010), ‘‘racial slur’’ comes from Dictionary.com’s 21st Century Lexicon, and ‘‘sexual slur’’ comes from Wikipedia.com (however, these definitions are common knowledge among competent speakers). 2 It may be that expressives serve as verbal ‘‘gestures’’ in that ‘‘the commitments incurred by someone who makes the gesture are commitments to targeted emotional attitudes, and not necessarily to thoughts’’ (Hornsby, 2001, p. 140–141). 3 As Austin explains, ‘‘Where, as often, the procedure is designed for use by persons having certain thoughts or feelings [. . .] then a person participating in and so invoking the procedure must in fact have those thoughts and feelings, and the participants must intend so to conduct themselves’’ (1962, pp. 15, 39). 4 For instance, although Potts allows that ‘‘the expressive imbues the at-issue content with new meaning and importance, and the at-issue content clarifies the meaning of the expressive,’’ that the ‘‘two dimensions shape each other,’’ and that ‘‘expressives color [content] with subjectivity and emotionality’’ (2008, p. 13), Potts also unambiguously asserts that ‘‘descriptives and expressives are fundamentally different’’ (2007, p. 165) and that the ‘‘descriptive and expressive realms are [. . .] distinguished not only syntactically (in the semantic types), but also model-theoretically’’ (2007, p. 167). In fact, Potts claims that ‘‘expressive denotations [. . .] are quite fundamentally different from the usual denotations in semantics, in that they [. . .] are metalogical’’ (2007, p. 188) and that therefore ‘‘a demand for a paraphrase of [expressives such as], say, damn is nonsensical – conceptually equivalent to asking for a paraphrase of Chris Potts’’ (2007, p. 178). Please cite this article in press as: Croom, A.M. Slurs. Lang. Sci. (2010), doi:10.1016/j.langsci.2010.11.005 A.M. Croom / Language Sciences xxx (2010) xxx–xxx 3 (2009) and Eric McCready (2010) have criticized Potts for not having a more flexible view of semantic possibility, and have argued for the addition of semantic types of mixed (descriptive and expressive) content. I will now review the linguistic behavior of slurs and compare them with the behavior of descriptives and expressives, to discern whether slurs are most plausibly of the descriptive, expressive, or mixed type. 3. Are slurs descriptives? Let us first compare slurs with descriptive types to see if it is plausible to analyze slurs as pure descriptives. Compare the descriptive case (3) with the slurring case (4): (3) (4) S is an African American. S is a nigger. Note that slurring terms such as nigger and slut differ from descriptive terms such as African American and woman because slurs carry derogatory force whereas descriptives usually do not (Hom, 2008; Himma, 2002; Hornsby, 2001).5 To call someone African American is not usually understood as an act of derogation, but rather one of straightforward description. However, to call some a nigger is usually understood as an act of derogation, regardless of whether or not it is also understood to describe the target agent or their features. So the difference between (3) and (4) suggests that it is inapt to treat the slur as a pure descriptive. Another reason to reject slurs as descriptives is that slurs pass two paradigmatic tests that are used to identify linguistic items that are non-descriptive,6 such as expressives. McCready (2010) proposes two tests that identify non-descriptive items. The first test identifies the scope or ‘‘scopelessness’’ of a linguistic item, and the second test identifies a linguistic item’s behavior under negation or denial (McCready, 2010, p. 6). An example of the first test is given below, which shows that descriptive terms (in 5a) differ in scope from slurring terms (in 6a). (5a) (6a) If I didn’t like African Americans, then I’d probably be a racist. If I didn’t like niggers, then I’d probably be a racist. Since in (5a) the descriptive term African American is embedded within the antecedent of the conditional, and since descriptive terms are not typically a reliable signal that the speakers using them possess derogatory attitudes in the same way that derogatory or slurring terms are reliable signals of this, it is clear that a speaker uttering (5a) does not generate the inference that they are racist simply in virtue of uttering (5a). The speaker of (5a) is only committed to saying that they probably would be racist if in fact they did not like African Americans, but that is an if they can plausibly deny. So the scope of the descriptive term African American is restricted by the conditional and does not project out to generate inferences about the speaker’s communicative intent. Yet although in (6a) the slurring term niggers is embedded within the antecedent of the conditional, the derogatory force of niggers scopes out of its embedded position to generate the inference that a speaker uttering (6a) is racist or at least holds racist views. For if the speaker were not currently in possession of derogatory attitudes, there are many other non-derogatory neutral descriptive terms that the speaker could have used, for instance, by saying (5a) instead. The second test that is used to identify items that are not purely descriptive focuses on an item’s behavior under negation or denial. An example of the second test used by McCready (2010, p. 7–8) is given below, which highlights how the linguistic behavior of descriptive terms (in 7) differs under denial from terms that are not purely descriptive (in 8). In these examples McCready points out that ‘‘In ordinary denial, the truth of any at-issue part of a sentence can be called into question’’ (2010, p. 7). This is evidenced below, where the standard interpretation of B’s denial of A is given in (7b): (7a) (7b) A: John came to the party last night. B: That’s not true/That’s false. = John didn’t come to the party. Slurring terms differ from descriptive terms in that their content ‘‘does not participate in denials’’ (McCready 2010, p. 7). This is evidenced from another of McCready’s examples (2010, p. 10) shown below, where the standard interpretation of B’s denial of A is given in (8b): (8a) (8b) A: Juan is a kraut. B: That’s not true/That’s false. – German people are not bad. McCready concludes that ‘‘The result of this test also supports the conclusion that the negative part of the meaning of Kraut, and, by extension, pejoratives in general [. . .] are not part of the at-issue [or descriptive] meaning’’ (2010, p. 10). Rather, ‘‘the negative part of the meaning’’ of slurs is non-descriptive. In effect, McCready’s tests are nice elucidations of 5 Hornsby, for instance, suggests that ‘‘derogatory words [. . .] apply to people and are commonly understood to convey hatred or contempt’’ and that ‘‘for each such word, there is [. . .] another [word] that applies to the same people but whose use does not convey these things – there is, that is, a neutral counterpart’’ (2001, p. 128–129). 6 Or as it is also sometimes put, they do not have at-issue content (Potts, 2007, 2008). Please cite this article in press as: Croom, A.M. Slurs. Lang. Sci. (2010), doi:10.1016/j.langsci.2010.11.005 4 A.M. Croom / Language Sciences xxx (2010) xxx–xxx two properties that Potts and others have previously identified in non-descriptive terms such as expressives. These two properties are that of nondisplaceability (Anderson and Lepore, forthcoming; Potts, 2007; Cruse, 1986) and immediacy (Potts, 2007). For instance, Alan Cruse (1986) has argued that one ‘‘characteristic distinguishing expressive meaning from propositional [descriptive] meaning is that it is valid only for the utterer, at the time and place of utterance. This limitation it shares with, for instance, a smile, a frown, a gesture of impatience’’ (p. 272). The idea is the now common one, that descriptives can be embedded (for instance, in conditional and modal expressions) in ways that slurs and expressives cannot. The second distinguishing property is immediacy: the characteristic whereby ‘‘the act of uttering an expressive morpheme is sufficient for conveying its content [. . .] expressive content is performative in [. . . that] the act of uttering an expressive is the emotive performance. Epithets’’ or slurs, Potts says, ‘‘provide an especially clear example of this’’ (2007, p. 180). The point is that to speak and think in the same terms as the racist is inappropriate. I suggest that this is because the choice of which words one accepts as legitimate for use comes prior to the choice of what one will in turn use those words to say, and so the speaker’s choice of acceptable terms itself out-scopes their choice of the particular positions in which they choose to place those terms within their utterance. So the speaker, by virtue of choosing to use nigger as part of their utterance at all rather than opting to use a non-derogatory neutral descriptive such as African American instead, is thereby typically taken to be indicating to hearers that they approve of their linguistic choice. In other words, in this case even if a speaker needed to refer to someone by their racial features, speakers have a strategic choice to refer to that person with a neutrally descriptive term instead of an alternative racial slur (Hornsby, 2001; Haverkate, 1990; Hare, 1963). Given the fact that a speaker can choose to refer to an African American as an African American instead of referring to them as a nigger, by choosing to use the slur over the neutral description the speaker usually communicates to hearers that they approve of the legitimacy of uttering the slurring term and what that typically conveys. Likewise, a speaker can also indicate to hearers that they reject a slurring term as illegitimate by opting to use, for instance, the neutral descriptive African American rather than the slur nigger (Brandom, 2000).7 Haverkate has suggested that, ‘‘speakers concentrate on making optimal choices from those sets of lexical items which share the same denotative content but differ so far as connotative meaning is concerned. It is at this level, for instance, that speakers decide to make use of a euphemism instead of a vulgarism or a neutral expression’’ and that ‘‘the strategical impact of these choices has been often overlooked’’ (1990, p. 78). Hornsby also points out that, ‘‘Since there are other [non-derogatory] words that suit us better, we lose nothing by imposing for ourselves a blanket selection restriction on them’’ (2001, p. 129). These points are made especially clear in an example offered by Mark Richard in When Truth Gives Out: Imagine standing next to someone who uses S as a slur [. . .] the racist mutters that building is full of Ss. Many of us are going to resist allowing that what the racist said was true. After all, if we admit its truth, we must believe that it is true that the building is full of Ss. And if we think that, we think that the building is full of Ss. We think, that is, what and as the racist thinks. (2008, p. 3–4) Let us reinforce these points with a few more examples. (9a) (10a) (11a) I went out with an African American. I went out with a nigger. I went out with a fucker. (9b) (10b) (11b) I went out with an African American, but I deny expressing any feelings about him. I went out with a nigger, but I deny expressing any feelings about him.⁄ I went out with a fucker, but I deny expressing any feelings about him.⁄ In the examples above the italicized items are (9) descriptive, (10) slurring, and (11) expressive. We see that by adding the continuation clause but I deny expressing any feelings about him to these examples, only (9b) remains clearly felicitous.8 This is because (as McCready (2010) and others have similarly argued) by producing the expressive fucker in utterances like (11b) above, which as an expressive possesses the immediacy property, the speaker is typically understood to be communicating the expressive content that they are in a heightened emotional state (Potts, 2007; Haverkate, 1990).9 In other words, in (11b) the further content specified by the continuation clause is rendered impotent by the fact that the speaker, by virtue of their chosen utterance, has already shown or demonstrated that they approve of the use of the expressive term and what that usually conveys. Resultantly, the negation carried by the continuation clause seems like mere lip service in comparison to the speaker’s choice to approve of the very use of the expressive term in the first place. 7 In Making It Explicit Robert Brandom (2000) argues that one could ‘‘refuse to employ the concept, on the grounds that it embodies an inference one does not endorse’’ (p. 126). 8 Williamson (2009) claims that ‘‘for someone who says ‘Lessing was Boche, although I do not mean to imply that Germans are cruel’ merely adds hypocrisy to xenophobia’’ (p. 150). As Potts rightly points out, ‘‘the immediacy property ensures that the damage is done as soon as nigger escapes his [the speaker’s] lips. The post-hoc attempt to clarify his intended meaning is thus futile’’ (2007, p. 181). 9 Although Peter Lasersohn (2005) points out that there are cases where predicates of personal taste encode a perspective other than the speaker’s – i.e. that the contents of predicates of personal taste are capable of felicitous embedding – scholars for the most part deny that expressive content is likewise embeddable (Quang, 1971; Cruse, 1986; Kaplan, 1989; Zimmermann, 1991; Kaplan, 1999; Soames, 2002; Potts, 2003, 2005). So Potts claims that, although ‘‘we need to allow for the possibility [. . .] It appears to be a marked option to evaluate expressives with a judge who is not also the speaker,’’ (2007, p. 176) and that ‘‘As a pragmatic default, the judge is the speaker’’ (2007, p. 175). Please cite this article in press as: Croom, A.M. Slurs. Lang. Sci. (2010), doi:10.1016/j.langsci.2010.11.005 A.M. Croom / Language Sciences xxx (2010) xxx–xxx 5 This point about expressives is traditionally thought to apply mutatis mutandis to slurs (Potts, 2007). Like the expressive case (11b) and unlike the descriptive case (9b), the slurring case (10b) is also not typically displaceable (Anderson and Lepore, forthcoming; Hornsby, 2001; Cruse, 1986). This is because by uttering the slur nigger, which as a slur possesses the immediacy property, the speaker is typically understood to be communicating that they are in a heightened (derogatory) state (Williamson, 2009, p. 148).10 In other words, in (10b) the further content specified by the continuation clause is rendered impotent by the fact that the speaker, by virtue of their chosen utterance, has already shown or demonstrated that they approve of the use of the slurring term and what that usually conveys in general and in application to this target in particular. Resultantly, the negation carried by the continuation clause seems like mere lip service in comparison to the speaker’s choice to approve of the very use of the slurring term in the first place.11 4. Are slurs expressives? In Section 3 we saw that it is implausible to characterize slurs as pure descriptives. Indeed, for precisely the reasons pointed out in Section 3 many scholars reject the claim that slurs are pure descriptives (Anderson and Lepore, forthcoming; Potts, 2007) and some have suggested that slurs should instead be treated as pure expressives (Potts, 2007). (At this point we are discussing pure descriptives and pure expressives; a discussion of mixed content will follow later.) Now, it was argued that pure descriptives are used to pick out descriptive items whereas pure expressives are used to indicate the speaker’s heightened emotional state (Potts, 2008), and that since pure descriptives are fundamentally distinct from pure expressives (Potts, 2007; Kratzer, 1999) it was thought by some scholars that descriptives are not in the business of indicating the speaker’s heightened emotional state and that expressives are not in the business of picking out descriptive items. Further, since it was thought by these scholars that expressives are typically used to indicate the speaker’s heightened emotional state and so do not pick out descriptive items (Jay and Janschewitz, 2008; Potts, 2007; Kratzer, 1999; Haverkate, 1990; Ayer, 1952), this view suggests that speakers are capable of felicitously producing utterances involving pure expressive types regardless of there being a target with certain descriptive features at all. The idea that seems to be suggested by these scholars is that pure expressives are typically indicative of the speakers that use them, and so not particularly of the audience or their features. This idea is suggested in the examples below, with expressives and descriptives in predicate position ((12a) and (13a) respectively) and NP position ((12c) and (13c) respectively): (12a) (13a) S is a fucker, but I deny saying anything about his [x]d. S is an African American, but I deny saying anything about his [x]d. (12c) (13c) That fucker is my colleague, but I deny saying anything about his [x]d. That African American is my colleague, but I deny saying anything about his [x]d. Let [x]d represent a variable that admits only of purely descriptive types. Substituting descriptives for [x]d in the examples above shows that the substitution of certain descriptive contents blocks the felicity of descriptives (13a) and (13c), but does not block the felicity of expressives (12a) and (12c). This is suggested in the examples involving substitutions below: (12b) (13b) S is a fucker, but I deny saying anything about his [race]. S is an African American, but I deny saying anything about his [race].⁄ (12d) (13d) That fucker is my colleague, but I deny saying anything about his [race]. That African American is my colleague, but I deny saying anything about his [race].⁄ 10 In Reference, Inference and the Semantics of Pejoratives, Williamson describes Frege as holding the view that ‘‘‘This dog howled the whole night’ and ‘This cur howled the whole night’ express the same thought (the same sense). They differ in that only the latter conveys an attitude of contempt for the dog on the part of the speaker’’ (2009, p. 148, emphasis added). 11 In The Semantics of Racial Epithets, Hom (2008) discusses what he calls non-derogatory, non-appropriated (NDNA) uses of slurs. Examples of NDNA uses include ‘‘Yao Ming is Chinese, but he’s not a chink’’ and ‘‘There are lots of Chinese people at Cal, but no chinks’’ (2008, p. 429). Although Hom concedes that in such cases the slur still has derogatory content – indeed, his own semantic account suggests that slurs always have derogatory content – Hom nonetheless claims that they lack derogatory force. However, it is unclear to me that these examples completely lack derogatory force (although they surely have comparatively less derogatory force than other more straightforward racist claims). This is because what is intended to be expressed by an utterance involving an NDNA use of a slur can alternatively be expressed by means of an utterance that does not require a slur at all. For example, speakers usually have a strategic choice to say something like ‘‘Yao Ming is Chinese, but does not deserve being derogated for being Chinese’’ instead of ‘‘Yao Ming is Chinese, but he’s not a chink’’ in order to avoid having to use the slur chink at all. Similarly, speakers usually have a strategic choice to say something like ‘‘There are lots of Chinese people at Cal, but no Chinese people deserving of racial derogation’’ instead of ‘‘There are lots of Chinese people at Cal, but no chinks’’ in order to avoid having to use the slur chink at all. The derogatory force that these NDNA examples can still be understood to carry is expressed even more clearly with an alternative example involving the slur nigger. For consider that speakers usually have a strategic choice to say something like ‘‘S is African American, but does not deserve being derogated for being African American’’ instead of ‘‘S is African American, but he’s not a nigger’’ in order to avoid having to use the slur nigger at all. Similarly, speakers usually have a strategic choice to say something like ‘‘There are lots of African American people at Cal, but no African American people deserving of racial derogation’’ instead of ‘‘There are lots of African American people at Cal, but no niggers’’ in order to avoid having to use the slur nigger at all. So speakers that choose to use an utterance involving the slurring term itself – instead of opting out and using an alternative utterance untainted by slurs – may still typically be taken to indicate that they sanction the use of the slurring term itself. And clearly this could carry derogatory force to those hearing the slur being used (hearers may wonder why the speaker did not make more of an effort to avoid using the slur altogether in that conversational context). Although I lack the space here to give Hom’s discussion of NDNA uses of slurs a full treatment, Hom is right to point these cases out. In Sections 5–7 of this article I offer my own treatment of how slurs can have varying degrees of derogatory and non-derogatory force. Please cite this article in press as: Croom, A.M. Slurs. Lang. Sci. (2010), doi:10.1016/j.langsci.2010.11.005 6 A.M. Croom / Language Sciences xxx (2010) xxx–xxx The idea is that since the particular descriptive features of the target (S) are inessential to the speaker indicating their own emotional state, in expressive cases (12b) and (12d) the speaker can felicitously call a target a fucker while denying that their utterance has anything to do with certain (e.g. racial) descriptive features of that target at all. In other words, these examples are suggestive that, because pure expressives work primarily to indicate the emotional state of the speaker, they can be felicitously uttered regardless of there being a target with certain (e.g. racial) descriptive features at all. Now, even if we assume that fucker is a pure expressive, the slur nigger clearly is not. This is indicated below: (12b) (13b) (14b) S is a fucker, but I deny saying anything about his [race]. S is an African American, but I deny saying anything about his [race].⁄ S is a nigger, but I deny saying anything about his [race].⁄ Notice from the examples above that whereas the expressive case (12b) is felicitous on the grounds that pure expressives do not pick out certain specific descriptive features (such as race) and therefore can be felicitously uttered while denying some particular set of descriptive features (such as race) to its target, the slurring case (14b) is not likewise felicitous. This shows that the slurring case (14b) is not like the expressive case (12b), but is instead rather like the descriptive case (13b). The intuition behind this result is straightforward enough, for slurs are usually understood to target those descriptive features of members belonging to a certain class: nigger typically slurs African Americans, chink typically slurs Asian Americans, and so on. In contrast with slurs, however, it has been argued that pure expressives do not target descriptive classes (Jay and Janschewitz, 2008; Potts, 2007; Kratzer, 1999; Haverkate, 1990; Ayer, 1952). This point is not specific to racial slurs but is a general point applying to others such as sexist slurs. Consider the following examples: (15a) (16a) (17a) S is a fucker, but I deny saying anything about her [x]d. S is a woman, but I deny saying anything about her [x]d. S is a slut, but I deny saying anything about her [x]d. (15b) (16b) (17b) S is a fucker, but I deny saying anything about her [sexual behavior/physiology]. S is a woman, but I deny saying anything about her [sexual behavior/physiology].⁄ S is a slut, but I deny saying anything about her [sexual behavior/physiology].⁄ It is suggested from these examples that whereas the expressive case (15b) is felicitous on the grounds that pure expressives do not pick out certain specific descriptive features (such as sexual behavior/physiology) and therefore can be felicitously uttered while denying some particular set of descriptive features (such as sexual behavior/physiology) to its target, the slurring case (17b) is not likewise felicitous. The slurring case (17b) is not like the expressive case (15b), but is instead rather like the descriptive case (16b). The point can also be put as follows: whereas slurring terms such as nigger and descriptive terms such as African American are sensitive to certain descriptive contents – for instance, those regarding racial features – this is not the case for pure expressive terms as they have usually been conceived by scholars such as Haverkate, Janschewitz, Jay, Kratzer, and Potts. Whereas the felicity of utterances involving slurs and descriptives can be blocked by continuation clauses containing the negation of the relevant descriptive content, the felicity of utterances involving pure expressives such as fuck cannot typically be blocked in this way. This difference between slurs and pure expressives suggests that slurs are not aptly characterized as of the pure expressive type. Indeed, this point is supported by the fact that, whereas some scholars hold that expressive terms such as fuck and damn do not differ in their contents (Potts, 2007), slurs seem to be distinguished from one another by virtue of (differences among) their descriptive contents. What makes a racial slur r a racial slur is determined by the content of r, just as what makes a sexist slur s a sexist slur is determined by the content of s (Hom, 2008; Himma, 2002). This point can be clarified with the examples below: (18a) (19a) S is a slut, but I deny saying anything about her [x]d. S is a nigger, but I deny saying anything about her [x]d. (18c) (19c) S is a slut, but I deny saying anything about her [race]. S is a nigger, but I deny saying anything about her [race].⁄ (18d) (19d) S is a slut, but I deny saying anything about her [sexual behavior/physiology].⁄ S is a nigger, but I deny saying anything about her [sexual behavior/physiology]. Although the sexist case (18a) and the racist case (19a) both involve utterances the felicity of which are blockable by some descriptive content, it is evident from these examples that they are not both blockable by the same descriptive content. That is to say, that content which blocks the felicity of the sexist slur in (18d) does not block the felicity of the racial slur in (19d), and that content which blocks the felicity of the racial slur in (19c) does not block the felicity of the sexist slur in (18c). Resultantly, since slurs are distinguishable from one another by virtue of their descriptive conditions, it follows that slurs must have descriptive contents. Further, these contents must possess sufficient differences among their varieties such that they can be aptly distinguished from one another by competent speakers. Resultantly, slurs are not best characterized as of the pure expressive type. Please cite this article in press as: Croom, A.M. Slurs. Lang. Sci. (2010), doi:10.1016/j.langsci.2010.11.005 A.M. Croom / Language Sciences xxx (2010) xxx–xxx 7 Early in the linguistics literature, Potts and other scholars assumed that we could neatly partition all of linguistic phenomena into either pure descriptive or pure expressive types and that these two types were fundamentally distinct (Potts, 2007; Kratzer, 1999). However, in Section 3 we saw that it was implausible to characterize slurs as pure descriptives, and in Section 4 we saw that it was implausible to characterize slurs as pure expressives. That is, in agreement with McCready (2010) our analysis of slurs leads us to suspect that they do not fit nicely into either one of these two mutually exclusive semantic categories.12 This is precisely because slurs possess an interesting combination of properties borrowed from both descriptive types and expressive types. Thus, our analysis suggests that slurs are of a mixed type, containing both descriptive and expressive elements. 5. Semantics, pragmatics, and philosophical accounts of slurs In this section we will now consider the predominant philosophical positions that analyze slurs outside the framework of semantic types. Generally, such accounts of slurs come in two flavors: semantic and pragmatic. Semantic theorists claim that the derogatory content of a slur is part of its literal meaning and that ‘‘their derogatory content [. . .] gets expressed in every context of utterance’’ (Hom, 2008, p. 416). This view suggests that the derogatory content of a slur can be explained independently of context (Hom, 2008; Hornsby, 2001).13 In contrast, pragmatic theorists deny that the derogatory content of a slur is part of its literal meaning. Pragmatic theorists typically argue that derogatory content is not part of the literal meaning of the slur, but is rather that which gets implicated or pragmatically communicated by way of the slur. Because conversations are cooperative activities operating within a social/linguistic normative framework, speakers can usually deviate from their conversational norms (e.g. violate a Gricean Maxim) to communicate – without literally saying – something else to a hearer. We as hearers are all very good at rationally working out what our fellow interlocutors are intending to communicate by way of what they are saying to us, and we as hearers do this by exploiting the extra-semantic features of the discourse. For instance, in figuring out what a speaker could have possibly meant to implicate by what they literally said, a hearer might take into consideration common knowledge assumed to hold between the interlocutors (at least for the purposes of the current conversation), tone of voice and bodily comportment, salient features of the environment, and so on. Accordingly, since the pragmatic theorist conceives the derogatory content of a slur to be a pragmatic phenomenon, she denies the semantic theorist’s claim that the derogatory content of a slur can be explained independently of context. Semantic theories of slurs are appealing because they can explain why it is that slurs carry derogatory content and force across such various conversational contexts. It is because according to this account slurs ‘‘literally say bad things, regardless of how they are used’’ (Hom, 2008, p. 416) and because they literally ‘‘prescribe harmful practices’’ to their targets (p. 440). According to the semantic view of Christopher Hom (2008), slurs: both insult and threaten their intended targets in deep and specific ways by both predicating negative properties to them and invoking the threat of discriminatory practice towards them. [. . .] For example, the epithet [i.e. racial slur] ‘chink’ expresses a complex, socially constructed property like: ought to be subject to higher college admissions standards, and ought to be subject to exclusion from advancement to managerial positions, and . . ., because of being slanty-eyed, and devious, and good-at-laundering, and . . ., all because of being Chinese. (p. 431) Jennifer Hornsby (2001) also advocates a semantic approach when she argues that: A unified account of a derogatory word cannot be achieved by identifying a pragmatic ingredient to be added to a semantic one given by the [derogatory] word’s neutral counterpart, because only the [derogatory] word itself provides the outlook from which one can make sense of the variety of associated [derogating] speech acts. The contours of the space of possible speech acts done with sentences containing particular derogatory words are discernable only from the standpoint of someone who can know what the words [literally] mean. Any derogatory word, like any other, has its potential for making speech acts because of what it [literally] means. (p. 135) Hornsby rejects pragmatic theories of slurs on the basis that, with slurs, ‘‘it is not merely that one does not count oneself among the word’s users so that one is not in a position to make their claims,’’ but rather that ‘‘one cannot endorse anything that is done with these words’’ at all (2001, p. 130). Yet the semantic account, at least as it is formulated by these scholars, remains problematic. One crucial problem with such accounts is that they seem unable to explain how it is that slurs can mean something non-derogatory, for instance, how they can communicate non-derogatory meaning between close in-group interlocutors. If derogation is part of the literal content of slurs, how can slurs mean something non-derogatory? (One of my main purposes in the next two sections, Sections 6 and 7, is to explain how slurs can mean something non-derogatory in certain contexts.) Semantic theorists have offered little by way of explanation to this question, and the fact that slurs are in certain cases felicitously used to mean something non-derogatory (e.g. in an appropriative manner) is now a well documented linguistic phenomenon (Brontsema, 2004; Kennedy, 2003) which a theory of slurs must account for if it is to be explanatorily adequate. For instance Russell Simmons, founder of Def Jam Records, reports from the perspective of hip-hop culture that: 12 Although in agreement with McCready on this point, I will later offer an analysis of the mixed nature of slurs that is very different from his. Typically, semantic theorists claim that a slur such as nigger really means something like African American and despicable because of it (Saka, 2007, p. 121) or African American and a fit object for derision because of it (Blackburn, 1984, p. 148). 13 Please cite this article in press as: Croom, A.M. Slurs. Lang. Sci. (2010), doi:10.1016/j.langsci.2010.11.005 8 A.M. Croom / Language Sciences xxx (2010) xxx–xxx When we say ‘nigger’ now, it’s very positive. Now all white kids who buy into hip-hop culture call each other ‘nigger’ because they have no history with the word other than something positive [. . .] When black kids call each other ‘a real nigger’ or ‘my nigger,’ it means you walk a certain way [. . .] have your own culture that you invent so you don’t have to buy into the US culture that you’re not really a part of. It means we’re special. We have our own language. (quoted in Jackson, 2005) Other prominent African American entertainers, such as Dave Chappelle (Leung, 2004), Nas (McLaughlin, 2008), 50 Cent, Ice Cube, and Richard Pryor (Jackson, 2005) have also claimed at some point in their careers that their use of slurs were not intended nor typically understood to carry derogatory meaning or force.14 In line with these observations, Erving Goffman (1967) has pointed out that ‘‘individuals who are on familiar terms with one another and need stand on little ceremony’’ with one another can exchange jokes and mock insults, ‘‘apparently as a means of poking fun at social circles where the ritual [insult] is seriously employed’’ (pp. 86). For instance, imagine that speaker A and hearer B are close friends that are aware of each others general beliefs and dispositions. If B knows A wellenough to know that A is not racist and is generally a decent person, then surely it is safe for B to assume that A meant to create rapport with B rather than to derogate B by A’s particular use of the slur. That is, as close friends, it is common knowledge between A and B that A in general intends to create rapport with B and not derogate B. This is, indeed, how one in general creates and maintains friendships, and probably how A and B in particular became friends in the first place. Additionally, Geoffrey Leech (1983) has argued that interlocutors may act superficially impolite with one another in order to foster a sense of social intimacy and to reduce relative inequalities between them, and Jonathan Culpeper suggests that ‘‘the more intimate a relationship, the less necessary and important politeness is [. . .] lack of politeness is associated with intimacy, and so being superficially impolite can promote intimacy. Clearly this only works in contexts in which the impoliteness is understood to be untrue’’ (1996, p. 352) such as in communicative exchanges between close friends or in-group members. Finally, Jonas Pfister (2010) recently argues that, ‘‘what may seem to be impolite at a (superficial) level of what is said, may nevertheless be polite at a (deeper) level of what is implicated’’ (p. 1278). So, for instance, if it is understood between African American interlocutors that no offense is intended towards one another, they may strategically produce utterances involving slurs to foster or promote intimacy and in-group solidarity. Indeed, a hearer B will tend to interpret an insult from a speaker A as mere banter if B thinks that A likes them, because the more one interlocutor A likes another interlocutor B the more likely it is that A will be concerned with B’s face and so be cautious not to offend B (Culpeper, 1996). In such appropriative or in-group uses, a slur is a form of ‘‘mock impoliteness’’ since it is understood as intentionally nonoffensive (Culpeper, 1996). For instance, in-group racial members share in many of the same discriminatory problems and face the same discriminatory prejudices. So creating a sense of solidarity and togetherness via in-group uses of slurs – the use of which is often restricted to only in-group members – can help speakers signal to each other that they are not alone and that others like them share in their pains, perspectives, and history of prejudices. As Brown and Levinson suggest, this is how ‘‘we get conventionalized (ritualized) insults as a mechanism for stressing solidarity’’ (1978, pp. 229), and David Sally (2003) has further argued that speakers engage in this sort of ‘‘risky speech’’ as a way ‘‘to reinforce their solidarity and rapport [by] us[ing] common ground not so much for translation as for security and assurance’’ (p. 1237; see also Clark, forthcoming). Indeed, there is a growing literature demonstrating that narrower psychological and social distance among interlocutors is correlated with a greater use of insults (Slugoski and Turnbull, 1988; Labov, 1972), swearing (Jay and Janschewitz, 2008), sarcasm (Kreuz et al., 1999; Jorgensen, 1996), teasing (Drew, 1987) banter (Leech, 1983), and mock impoliteness (Culpeper, 1996). It is evident, then, that in close relationships or in relationships between those that are closely alike, it may be assumed among interlocutors that there will be minimal danger of face threats during conversation. And it is this that can resultantly give rise to the employment of insults, jokes, or slurs as a way of building social intimacy among interlocutors. Furthermore, if the speaker is too polite they may actually insult the hearer by implying that the social distance or relative power between them is greater than it is (Brown and Levinson, 1978). For instance, think of cases where one romantic partner or family member is upset with the other and wants to communicate this. In such cases partners usually indicate this to each other by using more formal and socially distant terms (such as their partner’s first, last, and even middle name, if one is really in trouble) as opposed to using more socially close terms (such as darling or sunshine) that are reserved as terms of endearment for those that the speaker is or wishes to be close to (Brown and Levinson, 1978). So similarly, albeit through different means, in-group communication involving slurs can be seen as operating ‘‘as a positively polite stressing of in-group knowledge and commonality of attitudes’’ (Brown and Levinson, 1978, pp. 28). I suggest that insofar as semantic accounts such as Hornsby’s and Hom’s fail to account for the potential non-derogatory content of slurs (e.g. that content which is often communicated between in-group interlocutors that like one another), they remain unsatisfactory and at best incomplete (I discuss such non-derogatory content in Sections 6 and 7). Furthermore, there are other reasons why we should be unsatisfied with Hom’s analysis of the content of slurs. For instance, I remain skeptical that a racist using the slur chink to derogate Chinese Americans in different contexts is literally expressing that their target, for instance, ‘‘ought to be subject to higher college admissions standards, because they are good at laundering, which is due to their being Chinese American’’ (Hom, 2008, p. 431). That is, Hom’s analysis of the content of the racial slur chink is far too specific to help us understand what the slur chink systematically means across different conversational contexts. As Herman Cappelen and Ernest Lepore (2005) have argued ‘‘The semantic content of a sentence [or expression] S is the content that all 14 And surely despite the fact that these entertainers are neither linguists nor philosophers they are still to be considered as competent language users. Please cite this article in press as: Croom, A.M. Slurs. Lang. Sci. (2010), doi:10.1016/j.langsci.2010.11.005 A.M. Croom / Language Sciences xxx (2010) xxx–xxx 9 utterances of S share [. . . or] express no matter how different their contexts of utterance’’ (pp. 143). Accordingly, since what is typically expected from an analysis of the semantic or literal content of S is an analysis general enough to capture the systematic linguistic behavior of S across different conversational contexts, and since Hom’s analysis of the content of chink does not provide us with content general enough to satisfy this criteria (due to the implausibility of the slur chink literally expressing the particular properties Hom attributes to them) it is quite clear that Hom has not yet provided us with a satisfying analysis of the literal content of slurs. Since the semantic accounts of slurs we have surveyed have not yet provided us with a satisfying analysis of the literal content of slurs, and since they were further unable to account for the fact that slurs are derogatory in some cases but not in others, this might suggest that a pragmatic approach may be more apt for slurs. Three popular pragmatic approaches have been suggested: radical contextualism, Fregean minimalism, and pragmatic minimalism (Hom, 2008). In Nigger: The Strange Career of a Troublesome Word, Randall Kennedy (2003) argues that the meaning of a slur is context sensitive, such that the content of a slur – and hence whether the slur has derogatory or non-derogatory content – varies according to the particular features of the conversational context. For instance, Kennedy claims that the slur nigger ‘‘can mean many different things, depending upon, among other variables, intonation, the location of the interaction, and the relationship between the speaker and those to whom he is speaking’’ (2003, pp. 43). Hom (2008) calls this view ‘‘radical contextualism,’’ and George Carlin (1990) seems to have suggested this kind of view roughly twenty years ago in They’re Only Words: I get tired of people talking about bad words and bad language. Bullshit! It’s the context that makes them good or bad [. . .] For instance, you take the word nigger. There is absolutely nothing wrong with the word nigger in and of itself. It’s the racist asshole who’s using it that you ought to be concerned about. We don’t care when Richard Pryor or Eddie Murphy say it. Why? Because we know they’re not racist [. . .] We don’t mind their context because we know they’re black. This contextualist approach is appealing to some because, compared to semantic approaches, it seems to more straightforwardly explain why it is that slurs do not need to be derogatory across every conversational context. Because radical contextualism denies that slurs contain derogatory content and force as part of their literal meaning and instead claim that they are determined by the contextual factors of the interlocutor’s conversation, radical contextualists can explain how appropriative (non-derogatory) uses of slurs are possible by pointing out that such utterances involving slurs occur in non-derogatory contexts. Thus, some have argued that radical contextualism is better suited for explaining non-derogatory uses of slurs than semantic theories. However, most theorists will tend to be unsatisfied with such a radically contextual pragmatic theory, because such radical contextualism appears to siphon the meaning from the very linguistic items that we are interested in analyzing. To some extent, any adequate analysis of linguistic items must be restricted from being too radically contextual on pain of dissolving a stable object of study. Where pragmatic theorists differ then is in how radical their contextualism is. Hom argues that Kennedy’s account of slurs is too radical. The fundamental problem with radical contextualism is that it ‘‘offers little in the way of predicting the content [of a slur] in a context’’ and that, resultantly, the adoption of such a view indicates ‘‘a premature surrender in the search for a principled analysis of epithets, and should be left as a last resort’’ (Hom, 2008, p. 419–420). ‘‘Fregean minimalism’’ (as Hom (2008) terms the view) is an alternative and potentially more promising pragmatic strategy. The idea here is that, for instance, the pair of terms African American and nigger share the same sense but differ from one another in their evaluative coloring or tone (Frege, 1892, 1956; see also Williamson, 2009; Hom, 2008; Hornsby, 2001; Dummett, 1973). So when a speaker says (a) ‘‘S is an African American,’’ and (b) ‘‘S is a nigger,’’ the Fregean claims that what the speaker says in both cases is literally the same, but that (a) differs from (b) in the associated feelings or mood expressed. So on this view, the description African American and the slur nigger are literally synonymous, but have different evaluative colorings and associations tagged on to each of them. Importantly, Frege held that evaluative ‘‘colouring and shading are not objective,’’ and that since the evaluative coloring associated with each term will vary from speaker to speaker, the coloring associated with slurs such as nigger are not truth-conditional – and therefore not semantic or literal – aspects of the speaker’s discourse (1892, 59–61; see also Williamson, 2009; Dummett, 1973).15 Fregean minimalism is appealing to some scholars because it offers an account of why, for instance, the terms African American and nigger are both typically ascribed to African Americans while still acknowledging that the evaluative attitudes often expressed by speakers using each of these terms will differ. However, this view is problematic because if the terms nigger and African American share the same literal meaning and differ only in the subjective feelings or evaluative associations individual speakers attach to each, then it seems implausible that this view can adequately ‘‘account for how competent speakers uniformly and objectively understand the derogatory force associated with epithets [. . . that is] The word [nigger] is derogatory regardless of one’s personal associations or feelings toward African Americans’’ (Hom, 2008, p. 421). Another problem with Fregean minimalism is that it suggests that certain highly controversial racist claims literally express analytic truths (Hom, 2008). Since the Fregean minimalist holds that the terms African American and nigger express the 15 In Reference, Inference and the Semantics of Pejoratives, Timothy Williamson nicely describes Frege’s view: ‘‘In Fregean terminology, the lexicographers’ account of ‘Boche’ is in effect that it has the same reference as ‘German’ but a different tone. Indeed, since the differences between ‘Boche’ and ‘German’ apparently play no role in determining reference, and so make no difference to the way in which the terms contribute to the truth-conditions of sentences in which they occur, a Fregean might even count ‘Boche’ and ‘German’ as having the same sense. Frege himself gives just such an account of another pejorative term (1979, p. 140): ‘cur’ has the same sense and reference as ‘dog’ but a different tone. According to Frege, ‘This dog howled the whole night’ and ‘This cur howled the whole night’ express the same thought (the same sense). They differ in that only the latter conveys an attitude of contempt for the dog on the part of the speaker’’ (2009, p. 148). Please cite this article in press as: Croom, A.M. Slurs. Lang. Sci. (2010), doi:10.1016/j.langsci.2010.11.005 10 A.M. Croom / Language Sciences xxx (2010) xxx–xxx same sense, then on this view it should follow that statements involving these terms should express the same thought. Hom (2008) points this out with the following example: (a) (b) African Americans are African Americans. African Americans are niggers. What is problematic here is that whereas rationally competent speakers will be compelled to accept (a) as trivial and knowable a priori, this is not likely to be the case with (b) (Hom, 2008; see also Williamson, 2009). Most scholars will find it unappealing to go on record endorsing that racist claims such as (b) literally express analytic truths knowable a priori. According to a third pragmatic strategy advocated by Timothy Williamson called pragmatic minimalism, the descriptive African American and the slur nigger share the same reference, so on this view there is nothing derogatory as part of the literal content of what is said in the utterance of a slur (Williamson, 2009). Rather, according to this view derogatory force is implicated in conversational contexts where slurs are uttered as a result of an operating conventional implicature (Williamson, 2009; see also Hom, 2008). This view is nicely stated by Williamson: to assert ‘Lessing was a Boche’ would be to imply that Germans are cruel [. . .] Since the false implication that Germans are cruel does not falsify ‘Lessing was a Boche,’ it is not a logical consequence of ‘Lessing was Boche.’ Rather, in Grice’s terminology, ‘Lessing was a Boche has the conventional implicature that Germans are cruel [. . . that is,] ‘Lessing was a Boche’ and ‘Lessing was a German’ differ in conventional implicatures while being truth-conditional equivalent. (2009, pp. 149) McCready (2010) also endorses this view. McCready claims, for instance, that ‘‘Kraut is a pejorative term for German people on its nominal use. By saying [. . .(3) ‘‘He is a Kraut’’] I asserted that the referent of he is German, and express that I have negative feelings about him’’ (p. 6). McCready further explains that in slurring cases such as (3), ‘‘what is expressed by the sentence [. . .] is that the speaker takes German people to be bad’’ (p. 9). Although this is a popular view, it too faces serious problems. First, since the pragmatic minimalist holds, for instance, that the terms African American and nigger have the same literal meaning, they face the same problem as the Fregean minimalist in that they seem committed to the view that racist claims such as ‘‘African Americans are niggers’’ literally express analytic truths that are knowable a priori. This, I agree with Hom (2008), is an unappealing result. Secondly, since conventional implicatures differ from conversational implicatures in that only the latter are cancelable whereas the former are not (Williamson, 2009; Grice, 1975, 1961), if the utterance of slurs conventionally implicates derogatory force then ‘‘Derogation ought to occur in every context of use for epithets without any means for cancelation’’ (Hom, 2008, p. 424). However, there are clearly cases of felicitous yet non-derogatory uses of slurs that are genuinely meaningful, as we’ve already discussed above (Hom, 2008; Brontsema, 2004; Kennedy, 2003). So given that slurs can be used to build rapport among in-group interlocutors, it should now be clear that pragmatic minimalism is unappealing to the extent that it is unable to account for the empirical fact that there are genuine and felicitous uses of slurs that are non-derogatory and nonthreatening. Furthermore, since pragmatic minimalism holds that the terms nigger and African American have the same literal meaning (Williamson, 2009), on this view it should resultantly be infelicitous to ascribe the term nigger to someone that is not African American. Similarly, in the case of sexist slurs the pragmatic minimalist holds that the terms faggot and homosexual have the same literal meaning, and accordingly on this view it should be infelicitous to ascribe the term faggot to someone that is not homosexual. I suggest that these are problematic commitments for the pragmatic minimalist (and others sharing these commitments, such as Fregean minimalists) because, for instance, it is an empirical fact that in certain cases the term nigger is felicitously ascribed to non-African American individuals. Indeed, such uses of this slur are discussed in some detail by Michael MacDonald (2000) in his national bestseller All Souls: A Family Story from Southie. For example, MacDonald describes his experiences of growing up in his linguistic community and reports the following: Danny told me that the people that ended up in D Street were ‘‘white niggers’’ [. . .] Of course, no one considered himself a nigger. It was always something you called someone who could be considered anything less than you. I soon found out that there were a few black families living in Old Colony. They’d lived there for years and everyone said that they were okay, that they weren’t niggers but just black. (2000, pp. 60–61) Likewise, it is also an empirical fact that in certain cases the sexist slur faggot is felicitously ascribed to non-homosexual individuals (Szekely, 2008, ‘‘Offensive Words’’). Indeed, such uses of this slur are discussed in some detail by Louis Szekely (a.k.a. Louis C.K.) in his standup performance Chewed Up (2008). For example, Szekely describes his experiences of growing up in his linguistic community and reports the following: faggot didn’t mean gay when I was a kid, you called someone a faggot for being a faggot, you know? [. . .] ‘‘you’re not supposed to use those for that’’ [said in an annoying mocking voice as if from another person, then Louis C.K. replies normally] ‘‘shut up faggot!’’ [. . .] I would never call a gay guy a faggot, unless he was being a faggot. But not because he’s gay, you understand. (2008, ‘‘Offensive Words’’) So pragmatic minimalism is problematic insofar as it fails to account for the empirical fact that there are meaningful and felicitous uses of racial slurs that need not target an individual belonging to the race typically associated with that racial slur (see, for instance, MacDonald, 2000), and that there are meaningful and felicitous uses of sexist slurs that need not target an Please cite this article in press as: Croom, A.M. Slurs. Lang. Sci. (2010), doi:10.1016/j.langsci.2010.11.005 A.M. Croom / Language Sciences xxx (2010) xxx–xxx 11 individual belonging to the sexual group typically associated with that sexist slur (see, for instance, Szekely, 2008). Given that it is our aim to capture the linguistic behavior of slurs and provide an accurate representation of their systematic use in actual natural language discourse, I suggest that it would be unwise to neglect the aforementioned uses of slurs simply because they do not fit well with the early models of slurs that have been offered by scholars thus far. We have now surveyed five different accounts of slurs – Hornsby’s and Hom’s semantic accounts, Kennedy’s radical contextualism, Fregean minimalism, and Williamson’s pragmatic minimalism – and have found them all to be inadequate in one way or another. However, there is a final approach that we should consider at this point, proposed by Anderson and Lepore. According to Anderson and Lepore, ‘‘slurs are prohibited words not on account of any content they communicate, but rather because of relevant edicts surrounding their prohibition’’ (forthcoming, p. 2) and that ‘‘When a word is prohibited, then whoever violates its prohibition risks offending those who respect it. Presumably, prohibitions include deeply embedded occurrences of the [derogatory] word’’ (forthcoming, p. 18). Anderson and Lepore attempt to account for slurs as prohibited words because it presumably explains, first, why we cannot ‘‘unilaterally detach the affect, hatred and negative connotations tied to most slurs and use them interchangeably with their neutral counterparts’’ (Richard, 2008, pp. 62), and secondly, ‘‘why occurrences of slurs inside semantic attributions or within quotations can still inflict an offense’’ (Anderson and Lepore, forthcoming, p. 18). According to these authors, we cannot detach the negative connotations from slurs because slurs are prohibited pure and simple. Although I agree with Anderson and Lepore’s claim that slurs are typically prohibited words, to simply point this out is to do little more than remind us of one reason why we began the analysis of slurs in the first place. That is, it still offers little by way of explanation. For surely one will wonder why slurs were prohibited in the first place, if not by virtue of something that they literally mean or pragmatically or conventionally implicate. In other words, if Anderson and Lepore are to defend their claim that slurs are prohibited words, ‘‘not because of any content it or its uses communicate, or because of surrounding beliefs and intents, but because of a responsibility we all have in ensuring [that] certain violations [of prohibitions] are prevented’’ (forthcoming, p. 19) then they must now explain by virtue of what reasons the prohibitions against the use of (for instance) racial slurs were legitimately established in the first place. For surely we (non-racists) will want to maintain that there were and still are good reasons for slurs to be prohibited (at least by most, restricting their use to in-group members), over and above the brute fact that they have been marked ‘‘off limits.’’ And presumably these reasons for the prohibition against the use of slurs include at minimum that there is something derogatory about the literal meaning or the implications resulting from the use of the slurring words themselves (although this need not exhaust the reasons for their prohibition; that is, there may be additional sociological or historical factors motivating their prohibition as well). In short, since Anderson and Lepore’s account of slurs remains largely silent on what slurs mean, their account is at best incomplete. 6. The literal meaning of slurs: a new proposal Even among semantic theorists, little work has been done to explicate the literal meaning of slurs (Anderson and Lepore, forthcoming, p. 3). Intuitively, theorists that claim that slurs have literal meaning are tasked with explicating what it is. Typically, previous theorists have suggested that a slur such as kraut really means something like German and bad because of it (McCready, 2010) and that a slur such as nigger really means something like African American and despicable because of it (Saka, 2007) or African American and a fit object for derision because of it (Blackburn, 1984). As I will now point out, these suggestions are inadequate as analyses of the literal meaning of slurs. In this section, I will propose a more apt analysis of the literal meaning of the racial slur nigger. I will compare my analysis with the other competing proposals, and demonstrate in which ways my account is more reasonable. If it is correct, it may serve as a model for understanding other slurs of various kinds – like chink and slut – that target members associated with different classes. My view is the following. Slurs have mixed content in the sense that the use of racial slurs (such as nigger) can be analytically decomposed into both expressive and descriptive aspects. As a racial slur, by choosing to use the slur nigger instead of a neutrally descriptive term such as African American, the speaker intends to express (i) their endorsement of a (usually negative) attitude (ii) towards the descriptive properties possessed by the target of their utterance. For instance, consider the following felicitous utterance documented in Alex Haley’s The Autobiography of Malcom X: Now we all here like you, you know that. But you’ve got to be realistic about being a nigger. A lawyer – that’s no realistic goal for a nigger. You need to think about something you can be. (1964, pp. 38) This example is illuminating, for it suggests that (i) holds because the phrase ‘‘But you’ve got to be realistic about being a nigger’’ communicates the speaker’s endorsement of a negative attitude. It also suggests that (ii) holds because the phrase ‘‘Now we all here like you, you know that’’ suggests that what the negative attitude being expressed by the speaker is directed towards is not the agent, but rather some set of the agent’s properties. That is to say, the properties that the speaker endorses the expression of a negative attitude towards are properties that have been associated with members of a particular racial group, and as a result, the speaker does not directly express a negative attitude towards the agent him or herself. Indeed, in this example the speaker explicitly says that they like the target. However, an agent might indirectly express a negative attitude towards a target by expressing a negative attitude towards some set of properties that that target possesses. Slurs usually derogate, ridicule, or demean members of a certain class by targeting certain properties or features associated with those members as a class. For instance, African Americans that have been derogated with the slur nigger have Please cite this article in press as: Croom, A.M. Slurs. Lang. Sci. (2010), doi:10.1016/j.langsci.2010.11.005 12 A.M. Croom / Language Sciences xxx (2010) xxx–xxx typically been derogated on the basis of being ‘‘emotionally shallow, simple-minded, sexually licentious, and prone to laziness’’ (Asim, 2007, pp. 27). It is in such contexts where a speaker intends to ascribe at least some such properties to a target that the slur nigger has typically been employed. Further, as Brown and Levinson have suggested, a speaker S who implicates through their use of language that they are of higher social status or more powerful than their hearer H is engaging in talk that ‘‘is risky, but if he [the speaker] gets away with it (H doesn’t retaliate, for whatever reason), S succeeds in actually altering the public definition of his relationship to H: that is, his successful exploitation becomes part of the history of interaction, and thereby alters the agreed values of D [social distance between S and H] or P [relative power between S and H]’’ (1978, pp. 228). Accordingly, since our social identities are in part determined by the way society perceives us, and so the way society comes to interact and continues to interact with us (Goffman, 1967; see also Brown and Levinson, 1978), the derogative use of slurs can be extremely destructive to the actual character of an individual that it attacks. By ridiculing or derogating a member based on certain negative properties or features, the speaker employing the slur can support, enforce, and contribute to a history of acts that negatively alter the social identity of targeted members.16 This is done, presumably, for the purpose of increasing the difference in asymmetrical power relations among the interlocutors in the specific conversational context, or among the groups to which they belong more generally. It has been noted, for instance, that ‘‘the British and their colonial counterparts relied on [derogatory] language to maximize the idea of difference between themselves and their African captives’’ (Asim, 2007, pp. 12) which amounted to ‘‘an articulate and aggressive racism which excluded the Negro from the society of competing equals without deporting him, by the simple and brutal mechanism of formally defining him as subhuman’’ (Fredrickson, 1971, pp. 41). Importantly, notice that since the properties that the speaker is endorsing the expression of a negative attitude towards are properties that have been typically associated with members of a particular racial group (for instance African Americans), it is possible for these same negative properties to be ascribed to members of a different racial group. In other words, it is surely possible that some of the properties that have been typically associated with African Americans can be properties that are also ascribed to someone that is not African American. For example, if stamina had been typically associated with African Americans, this does not by any means restrict us from applying the term stamina to, say, Chinese Americans. Therefore, my analysis of the slur nigger avoids the problems that Fregean minimalists, pragmatic minimalists, and linguists such as McCready (2010) inevitably face. That is, since McCready (2010), Williamson (2009), Saka (2007), Blackburn (1984), Fregeans, and others assume that nigger literally means the same as African American and that faggot literally means the same as homosexual, they are unable to account for cases where the slur nigger is in fact felicitously ascribed to non-African American individuals (MacDonald, 2000) and for cases where the slur faggot is in fact felicitously ascribed to non-homosexual individuals (Szekely, 2008). My account does not share this widely held assumption, for I have argued that the negative attitude typically expressed by the speaker using e.g. the slur nigger need not necessarily be directed towards an agent that is African American, but rather can be directed towards some set of properties associated with African Americans that the target agent (who may be African American or not) is thought to possess. My point is made clear below, where (20a) provides an example context, (20b) provides an example sentence, and (20c) and (20d) provide competing paraphrases. (20c) provides a standard paraphrase common of the aforementioned scholars, whereas (20d) provides an example of my own original paraphrase. In the original context, a comedian asks the crowd: (20a) (20b) (20c) (20d) ‘‘Is you a nigger?’’ All of them answer in the affirmative, including blacks, whites, Latinos, and a Sicilian, who proclaims, ‘‘I’m more of a nigger than any nigger in here.’’ (quoted in Asim, 2007, pp. 193) I’m more of a nigger than any nigger in here. [I’m more of a [nigger] than any [nigger]in here] = I’m more of a [person that is African American and despicable because of it] than any [other person that is African American and despicable because of it] in here.⁄ [I’m more of a [nigger] than any [nigger]in here] = I’m more of a [person with a set of negatively viewed properties that have been associated with African Americans] than any [person with a set of negatively viewed properties that have been associated with African Americans] in here. In this case, (20c) is clearly bizarre, because it seems infelicitous for people that are not African American to assert that they are African American and despicable because of it. On the other hand, although (20d) is somewhat cumbersome, it is perfectly adequate. For instance, even though it seems infelicitous for people that are not African American to assert that they are African American and despicable because of it, it is not infelicitous for people that are not African American to assert that they are people that possess a set of negatively viewed properties that have been associated with African Americans. In other words, because the speaker intends to communicate a negative attitude towards some set of properties associated with African Americans, his negative attitude need not be restricted to actual African Americans at all. People of other races may also share these same negatively viewed properties. However, unless otherwise indicated, it is generally assumed that those 16 As Jabari Asim (2007) points out, ‘‘For the slaveholding class, validation was intimately bound to – in fact, impossible without – the consistent degradation of those they chose to enslave. Such degradation, heard in the court of public opinion and reinforced through daily plantation protocols, led to what might be called the founding fictions of American slave society: (1) whites were superior beings destined to rule over their lesser counterparts, and (2) blacks were unworthy creatures whose very unworthiness made them perfectly suited to a lifetime of forced servitude. ‘‘The idea of the superiority of whites was etched into the slave’s consciousness by the lash and the ritual respect he was forced to give to every white man,’’ noted John W. Blassingame (1979) in The Slave Community. At the same time, masters pretended that slaves were simple-minded and childlike because it helped ‘‘to relieve themselves of the anxiety of thinking about slaves as men. In the centuries that followed – long after the official end of slavery – whites of all classes came to rely on language (and especially the use of pejoratives like the N word) in the pursuit of such relief’’ (pp. 14). Please cite this article in press as: Croom, A.M. Slurs. Lang. Sci. (2010), doi:10.1016/j.langsci.2010.11.005 A.M. Croom / Language Sciences xxx (2010) xxx–xxx 13 who possess the properties typically associated with African Americans are most likely to be actual African Americans. But notice that this still leaves open the possibility that, although speakers usually ascribe the term nigger to African Americans, it is still felicitous under certain circumstances to ascribe the term nigger to non-African Americans insofar as these non-African Americans are assumed to possess the negative properties typically associated with African Americans. Therefore, unlike the proposals offered by other scholars, my analysis is the first that can account for cases where the racial slur nigger is in fact felicitously ascribed to non-African American individuals (MacDonald, 2000) and for cases where the sexist slur faggot is in fact felicitously ascribed to non-homosexual individuals (Szekely, 2008). As these cases represent actual and widespread uses of slurs, accounting for them is a virtue of my analysis that is absent in others. Furthermore, I propose that my account of the literal meaning of slurs is more adequate than that previously proposed by Hom (2008) because his explication of the meaning of the racial slur chink is far too specific to help us understand what the slur chink systematically means across different conversational contexts (see Cappelen and Lepore, 2005). This was a problem of Hom’s that I pointed out earlier. My analysis, on the other hand, captures our intuitions about the literal meaning of the racial slur in different kinds of conversational contexts while still remaining sufficiently general so as to be systematically applicable to other cases. We have covered a large amount of terrain in this essay, so I would now like to briefly restate our main findings: (1) Slurs are not purely descriptive terms, and so an analysis of slurs must account for the fact that slurs are expressive of attitudes or emotions. This I argued and accounted for in Section 3. (2) Slurs are not purely expressive terms, and so an analysis of slurs must account for the fact that slurs target descriptive features. This I argued and accounted for in Section 4. (3) Slurs typically carry derogatory content and force across different conversational contexts, and so an analysis of slurs must not be so radically contextual as to neglect this fact. This I argued and accounted for in Section 5. (4) Slurs can be used non-derogatorily, for instance in an appropriative manner, and so an analysis of slurs must be contextual enough so as to account for this fact. This I argued and accounted for in Section 5. (5) Slurs have non-derogatory correlates – or at least there are alternative lexical options available to speakers such that their use of the slur over a neutral counterpart typically expresses a strategic choice that can signal derogatory intent towards a target – and so the speaker’s choice to use the slur is often taken to indicate their approval of the slurring term and what it is typically used to convey. This I argued and accounted for in Sections 3 and 6. (6) Slurs can be felicitously applied towards individuals that do not belong to the group typically associated with the slur; for instance, it is an empirical fact that the racial slur nigger is used by some speakers towards individuals that are not African American, and that the sexist slur faggot is used by some speakers towards individuals that are not homosexual. This I argued and accounted for in Sections 5 and 6 (and continue to discuss in Section 7). I suggest that the previous points (1)–(6) provide a minimal set of desiderata to be met by any explanatorily successful account of slurs. 7. Criterial features/properties, family resemblances, and category membership While everyday speakers, such as Louis Szekely and Michael MacDonald, find it quite natural that the felicitous use of slurs need not be restricted by any one criterial feature such as the target’s racial identity, some traditional philosophers and linguists may find this result intuitively unappealing. Surely, the traditionalist may argue, since the slur nigger has most often been ascribed to African Americans and has been used to derogate them, the felicitous application of that term must be restricted to African Americans. The purpose of this final section is to suggest that this traditional assumption is too strong. Indeed, in this final section I’d like to suggest that my analysis of slurs can be combined with a family-resemblance conception of category membership, and that this offers an explanatorily powerful and true to the facts account of the various empirical uses of slurs (although my account of slurs does not require all concepts to be restricted to a family-resemblance account). The ‘‘traditional view’’ widely assumed in the linguistic and philosophical literature is that category membership is determined by the possession of some common, essential, criterial feature or property (Rosch and Mervis, 1975). However, empirical research on children has suggested that they first learn to classify items as members of the same category, not on the basis of those members possessing some common criterial feature (Rosch and Mervis, 1975; Bruner et al., 1966; Vygotsky, 1962), but rather by means of the formation of ‘‘complex structures’’ and groupings (Rosch and Mervis, 1975, p. 603; Bruner et al., 1966). Eleanor Rosch and Carolyn Mervis (1975), drawing on the philosophical work of Ludwig Wittgenstein (1953), call this form of categorical membership a ‘‘family resemblance relationship,’’ which is a relationship of items in a category where ‘‘each item has at least one, and probably several, elements in common with one or more other items, but no, or few, elements are common to all items’’ (p. 575). As Rosch and Mervis explain: The more prototypical a category member, the more attributes it has in common with other members of the category and the less attributes it has in common with contrasting categories. Thus, prototypes appear to be just those members of the category which most reflect the redundancy structure of the category as a whole. That is, categories form to maximize the information-rich clusters of attributes in the environment. (1975, p. 602) Please cite this article in press as: Croom, A.M. Slurs. Lang. Sci. (2010), doi:10.1016/j.langsci.2010.11.005 14 A.M. Croom / Language Sciences xxx (2010) xxx–xxx So on the family-resemblance conception of category membership, what makes some individual a member of the category DOG is not some criterial feature or property that all and every individual must have in order to be considered a dog. Rather, in line with the family-resemblance conception, I suggest that we view some categories of natural language as constellations of properties or features. In applying such categories in order to communicate some content to a hearer, speakers are able to choose among the categories available in their lexical inventory to pick out that which is (although not necessarily an ‘‘ideal’’ choice, at least) their most apt choice for their current conversational purpose. For instance, a paradigmatic or prototypical dog may have a tail and bark, but surely a speaker can still felicitously and informatively refer to an individual as a dog even if that individual does not have a tail and does not bark, provided that the category DOG is that which is strategically most apt among the lexical choices available for the speaker’s conversational purpose. For instance, assume that the family resemblance category (C) DOG consists in a network of properties (P) such as those displayed below:17 C. P1. P2. P3. P4. P5. P6. P7. P8. DOG Has four legs Has a tail Barks Has a keen sense of smell Urinates on trees Has shaggy hair Chases cats Likes to play catch These properties (P1)–(P8) could be ranked in the order to which their possession by an individual is taken as a salient indicator of category membership. For instance, properties (P1), (P2), and (P3) would be ranked relatively higher than properties (P6), (P7), and (P8), and so the former set of properties would be considered more salient indicators than the latter set of properties that the individual possessing them is a member of the category DOG. In other words, a prototypical dog would possess the highest ranking and highest number of properties in (P1)–(P8), although an individual can still informatively be considered as a dog while failing to possess some of the properties in (P1)–(P8) or even the highest ranking property (P1). However, for the speaker’s choice to refer to a target individual as a dog to be considered a strategically apt choice, it must be thought (or at least assumed for the purpose of the conversation) that the target individual possesses a sufficient set of properties (P1)–(P8) such that DOG is the most appropriate category under which to subsume the individual for the purposes of the speaker’s current conversational aim. That is, the category DOG must be a strategically better choice than other categories available in the speaker’s lexicon for the conversational purpose at hand. So in contrast with the traditional view where the possession of some criterial feature or property was thought to be essential for an individual’s being a member of some category, Rosch and Mervis suggest that e.g. although different dogs will share different sets of (P1)–(P8) with other dogs, it need not be the case that for each dog, it must share some essential criterial feature or property with every other dog. On a family-resemblance account of category membership, an individual may even fail to possess the highest ranking or most salient indicator of category membership while still being best considered a member of that category for the current conversational purpose at hand. I suggest that Rosch and Mervin’s family-resemblance analysis may be fruitfully extended to explain how speakers actually employ slurs in real life cases. For instance, assume that the family resemblance category (C) designated by the term nigger – call this category N – consists in a constellation of properties (P) such as those displayed below:18 C: P1. P2. P3. P4. P5. P6. P7. P8. N African American Prone to laziness Subservient Commonly the recipient of poor treatment Athletic Emotionally shallow Simple-minded Sexually licentious These properties (P1)–(P8) could be ranked in the order to which their possession by an individual is taken as a salient indicator of category membership. For instance, property (P1) would be ranked relatively higher than property (P6), and so (P1) would be considered a more salient indicator than (P6) that the individual possessing it is a member of the category N. In other words, although (racist or in-group) speakers may prototypically ascribe the term nigger to African Americans, (racist or in-group) speakers may still informatively or effectively ascribe that slur to someone that fails to be African American for their conversational purpose. However, for the (racist or in-group) speaker’s choice to refer to a target individual as a nigger 17 This is just an example used to illustrate the main point of this article, and is not presented as a rigorous analysis of the concept DOG. The following is merely an example of a list of properties that other scholars have already suggested that speaker’s using the racial slur nigger typically intend to predicate of their targets (for instance, see Asim, 2007). 18 Please cite this article in press as: Croom, A.M. Slurs. Lang. Sci. (2010), doi:10.1016/j.langsci.2010.11.005 A.M. Croom / Language Sciences xxx (2010) xxx–xxx 15 to be considered a strategically apt choice, it must be thought (or at least assumed for the purpose of the conversation) that the target individual possesses a sufficient set of properties in (P1)–(P8) such that N is the strategically most appropriate category under which to subsume the individual for the purposes of the speaker’s current conversational aim (which may be to derogate or build rapport with some hearer). For instance, if a speaker is intending to communicate that some target that they dislike possesses some subset of properties (P2)–(P8) – such as in the case MacDonald (2000) describes regarding the ‘‘white niggers’’ that ended up on D Street (pp. 60–61) – that speaker might strategically choose to employ N as the category that most efficiently and economically predicates the intended properties of their target and most forcefully expresses a negative attitude towards them, at least to an extent that is better than other categories available to the speaker in their lexical inventory. As we would all agree, if one wished to express their contempt for a target by linguistic means, hardly any words could serve this purpose better than slurs such as nigger. Additionally, if for instance an in-group speaker is intending to communicate that they are sufficiently similar to some hearer insofar as they both possess properties (P1), (P4), and possibly others, and if these interlocutors know each other well-enough or have established enough common ground to understand that the speaker does not dislike the target and does not intend to communicate that the target possesses most of the other (typically negative) properties belonging to N, an in-group speaker might strategically choose to employ N as the category that most efficiently and economically predicates the intended (shared) properties of their target such as e.g. (P1) African American, (P4) commonly the recipient of poor treatment, and possibly others, at least to an extent that N is better for this than other categories available in the speaker’s lexicon. So in contrast with traditional views of slurs, where the possession of some criterial feature or property (being African American) was assumed to be essential for an individual’s being a member of some category (the category N), I suggest that e.g. although different individuals that are referred to as niggers are likely to share different subsets of properties (P1)–(P8) with other individuals also referred to by this pejorative term, it need not be the case that for each slurred individual they must share some criterial feature or property (such as being African American) with every other slurred individual. In line with the family-resemblance account of category membership, I suggest that an individual referred to by the slur nigger may even fail to be African American, while the employment of that slur on part of the speaker may still be considered by that speaker to be a strategically apt choice for their current conversational purpose. In other words, speakers must be as efficient and economical as they can with the linguistic tools that are available to them for the purpose at hand, and when it is the speaker’s intention to derogate a hearer or build rapport with them, although the employment of slurs may seem like a less than ‘‘ideal’’ linguistic option for the speaker’s conversational aim,19 they are still strategically apt enough to be reliably used by racists and in-group members to achieve their respective communicative purposes. 8. Concluding remarks Throughout the course of this paper we have established several findings. First, contrary to the early assumption of Potts (2008, 2007) and others that all linguistic phenomena could be neatly partitioned into either pure descriptive or pure expressive semantic types, and that these types are fundamentally distinct, I argued that slurs are not aptly characterized as either pure descriptives (in Section 3) or pure expressives (in Section 4). In agreement with McCready (2010) I argued that slurs possess a combination of both descriptive and expressive properties and so are best characterized as a mixed type. In Section 5 I reviewed the most influential accounts of slurs that have been suggested, including those derived from Gottlob Frege (1892), Jennifer Hornsby (2001), Randall Kennedy (2003), Christopher Hom (2008), Timothy Williamson (2009), Eric McCready (2010), and Luvell Anderson and Ernest Lepore (forthcoming). After explaining the many ways in which these analyses are inadequate, in Section 6 I then provided a new analysis of the literal meaning of slurs and demonstrated in what ways my analysis is superior to those of the scholars mentioned above. Finally, in Section 7 I showed how a family-resemblance conception of category membership can help us achieve a clearer understanding of the various ways in which slurs, for better or worse, are actually put to use in natural language discourse. Acknowledgements I would like to extend my sincerest thanks to Elisabeth Camp and Robin Clark for much discussion on this topic and for many helpful suggestions. Thanks also to Nigel Love and several anonymous reviewers at Language Sciences for their helpful comments on earlier drafts of this manuscript. I am also grateful to Sally Haslanger for discussing this topic with me while she was visiting the University of Pennsylvania to deliver the 2010 Seybert Lectures. And much thanks to Luvell Anderson for earlier discussion on this topic as well. Responsibility for any remaining errors should to be attributed solely to the author. References Anderson, L., Lepore, E., forthcoming. Slurring Words. Nous. Asim, J., 2007. The N Word. Houghton Mifflin Company, New York. 19 One might even question what ‘‘ideal’’ would mean in such cases, given the tradeoffs required to maximize on precision, economy of expression, or politeness strategies. Please cite this article in press as: Croom, A.M. Slurs. Lang. Sci. (2010), doi:10.1016/j.langsci.2010.11.005 16 A.M. Croom / Language Sciences xxx (2010) xxx–xxx Austin, J.L., 1962. How to Do Things with Words. Harvard University Press, Cambridge. Ayer, A.J., 1952. Language, Truth and Logic. Dover Publications, New York. Blackburn, S., 1984. Spreading the Word. Oxford University Press, Oxford. Blassingame, J., 1979. The Slave Community: Plantation Life in the Antebellum South. Oxford University Press, New York. Brandom, R., 2000. Making It Explicit: Reasoning, Representing, and Discursive Commitment. Harvard University Press, Cambridge. Brontsema, R., 2004. A queer revolution: reconceptualizing the debate over linguistic reclamation. Colorado Research in Linguistics 17, 1–17. Cappelen, H., Lepore, E., 2005. Insensitive Semantics. Oxford University Press, New York. Carlin, G., 1990. They’re only words. In: Parental Advisory: Explicit Lyrics. Atlantic Records, New Castle. Clark, R., forthcoming. Meaningful Games: Exploring Language with Game Theory. MIT Press, Cambridge. Cruse, A., 1986. Lexical Semantics. Cambridge University Press, New York. Culpeper, J., 1996. Towards an anatomy of impoliteness. Journal of Pragmatics 25, 349–367. Drew, P., 1987. Po-faced receipt of teases. Linguistics 25, 219–253. Dummett, M., 1973. Frege: Philosophy of Language. Duckworth, London. Fredrickson, G., 1971. The Black Image in the White Mind. Wesleyan University Press, Middletown. Frege, G., 1892. On sense and reference. In: Geach, P., Black, M. (Eds.), Translations from the Philosophical Writings of Gottlob Frege. Basil Blackwell, Oxford, pp. 56–78. Frege, G., 1956. The thought: a logical inquiry. Mind, New Series 65, 289–311. Goffman, E., 1967. Interaction Ritual: Essays on Face-to-Face Behavior. Doubleday Anchor, New York. Grice, H., 1961. The causal theory of perception. Proceedings of the Aristotelian Society, Supplementary 35, 121–153. Grice, H., 1975. Logic and conversation. In: Cole, P., Morgan, J. (Eds.), Syntax and Semantics, Speech Acts, vol. 3. Academic Press, New York, pp. 41–58. Haley, A., 1964. The Autobiography of Malcom X. Ballantine, New York. Hare, R., 1963. Freedom and Reason. Oxford University Press, New York. Haverkate, H., 1990. A speech act analysis of irony. Journal of Pragmatics 14, 77–109. Himma, K., 2002. On the definition of unconscionable racial and sexual slurs. Journal of Social Philosophy 33, 512–522. Hom, C., 2008. The semantics of racial epithets. Journal of Philosophy 105, 416–440. Hornsby, J., 2001. Meaning and uselessness: how to think about derogatory words. Midwest Studies in Philosophy 25, 128–141. Jackson, D., 14 December 2005. Epithet Stung, Even for Pryor. The Boston Globe. <http://www.boston.com/news/globe/editorial_opinion/oped/articles/ 2005/12/14/epithet_stun g_even_for_pryor/>. Jay, T., 1992. Cursing in America. John Benjamins, Philadelphia. Jay, T., 2000. Why We Curse: a Neuro-Psycho-Social Theory of Speech. John Benjamins, Philadelphia. Jay, T., Danks, J., 1977. Ordering of taboo adjectives. Bulletin of the Psychonomic Society 9, 405–408. Jay, T., Janschewitz, K., 2008. The pragmatics of swearing. Journal of Politeness Research 4, 267–288. Jorgensen, J., 1996. The functions of sarcastic irony in speech. Journal of Pragmatics 26, 613–634. Kaplan, D., 1989. Demonstratives: an essay on the semantics, logic, metaphysics, and epistemology of demonstratives and other indexicals. In: Almog, J., Perry, J., Wettstein, H. (Eds.), Themes from Kaplan. Oxford University Press, New York, pp. 481–614. Kaplan, D., 1999. What is Meaning? Explorations in the Theory of Meaning as Use. Brief Version – Draft 1. Manuscript, UCLA. Kennedy, R., 2003. Nigger: The Strange Career of a Troublesome Word. Vintage Books, New York. Kratzer, A., 1999. Beyond ouch and oops: how descriptive and expressive meaning interact. In: Handout from the Cornell Conference on Theories of Context Dependence, Semanticsarchive.net. Kreuz, R., Kassler, M., Coppenrath, L., Allen, B., 1999. Tag questions and common ground effects in the perception of verbal irony. Journal of Pragmatics 31, 1685–1700. Labov, W., 1972. Language in the Inner City: Studies in the Black English Vernacular. Blackwell, Oxford. Lasersohn, P., 2005. Context dependence, disagreement, and predicates of personal taste. Linguistics and Philosophy 28, 643–686. Leech, G., 1983. Principles of Pragmatics. Longman, London. Leung, R., 29 December 2004. Chappelle: ‘An Act of Freedom’: Comedy Central Comedian Talks about Using the ‘N’ Word in Jokes. 60 Minutes, CBS News. <http://www.cbsnews.com/stories/2004/10/19/60II/main650149.shtml>. MacDonald, M., 2000. All Souls: A Family Story from Southie. Ballantine, New York. McCready, E., 2010. Varieties of conventional implicature. Semantics and Pragmatics 3 (8), 1–57. McLaughlin, E. 16 July 2008. Rapper has Defiant Words for New Album. CNN Entertainment, <http://www.cnn.com/2008/SHOWBIZ/Music/07/16/ music.nas/index.html.> Pfister, J., 2010. Is there a need for a maxim of politeness? Journal of Pragmatics 42, 1266–1282. Potts, C., 2003. Expressive content as conventional implicature. In: Kadowaki, M., Kawahara, S. (Eds.), Proceedings of the North East Linguistics Society, vol. 33. GLSA, Amherst, pp. 303–322. Potts, C., 2005. The Logic of Conventional Implicatures. Oxford University Press, New York. Potts, C., 2007. The expressive dimension. Theoretical Linguistics 33, 165–197. Potts, C., 2008. The pragmatics of conventional implicature and expressive content. In: Maienborn, C., Portner, P. (Eds.), Semantics: An International Handbook of Natural Language Meaning. Mouton de Gruyter, Berlin. Quang, P., 1971. English sentences without overt grammatical subject. In: Zwicky, A., Salus, P., Binnick, R., Vanek, A. (Eds.), Studies out in Left Field: Defamatory Essays Presented to James D. McCawley on the Occasion of his 33rd or 34th birthday. John Benjamins, Philadelphia, pp. 3–10. Richard, M., 2008. When Truth Gives Out. Oxford University Press, New York. Rosch, E., Mervis, C., 1975. Family resemblances: studies in the internal structure of categories. Cognitive Psychology 7, 573–605. Saka, P., 2007. How to Think about Meaning. Springer, Netherlands. Sally, D., 2003. Risky speech: behavioral game theory and pragmatics. Journal of Pragmatics 35, 1223–1245. Slugoski, B., Turnbull, W., 1988. Cruel to be kind and kind to be cruel: sarcasm, banter and social relations. Journal of Language and Social Psychology 7, 101– 121. Soames, S., 2002. Beyond Rigidity. Oxford University Press, New York. Szekely, L., 2008. Offensive words. In: Chewed Up. Image Entertainment, Los Angeles. Williamson, T., 2009. Reference, inference and the semantics of pejoratives. In: Almog, J., Leonardi, P. (Eds.), The Philosophy of David Kaplan. Oxford University Press, New York, pp. 137–159. Wittgenstein, L., 1953. Philosophical Investigations. Anscombe, GEM (Trans.). Blackwell Publishers, Oxford. Zimmermann, T., 1991. Kontextabha Ngigkeit. In: Stechow, A., Wunderlich, D. (Eds.), Semantik/Semantics: Ein Internationales Handbuch der Zeitgeno Ssischen Forschung. Walter de Gruyter, Berlin, pp. 156–229. Please cite this article in press as: Croom, A.M. Slurs. Lang. Sci. (2010), doi:10.1016/j.langsci.2010.11.005

RELATED PAPERS

RELATED TOPICS

Log In

Slurs

Slurs

Slurs

Slurs

Slurs

Related Papers

RELATED PAPERS

RELATED TOPICS