Download Understanding the Semantics and Social Significance of Derogatory Words and more Study notes Religion in PDF only on Docsity!
Slurs Aren’t Special
Geoff Nunberg School of Information, UC Berkeley Last revised March 1, 2013 The surest sign that a group or society has entered into the self-conscious possession of a new concept is that a corresponding vocabulary will be developed, a vocabulary which can then be used to pick out and discuss the concept with consistency. Quentin Skinner, “The Idea of a Cultural Lexicon” Were there any perverts before the latter part of the nineteenth century? Ian Hacking, “Making Up People” The Invention of Slurs Were there any racial or ethnic slurs before 1940? That’s about when slur was first applied to a word that derogated someone on the basis of race or religion, rather than simply as “an expression or suggestion of disparagement or reproof”—still the only definition that the Oxford English Dictionary gives for the noun. 1 Even that date is misleadingly early: this use of the word of the appeared only in African American publications like Jet and the Chicago Defender until the early 1960’s, when it began to show up in mainstream publications like Time and the New York Times. Of course there have always been words that derogate people on the basis of their membership in a certain group, but before the mid-twentieth century there was no one English term that gathered such words as a class. And while some of the items we now describe as slurs (^1) The OED entry for the noun slur has not been revised since 1912, so includes no sense for the use of the word to refer to words like fag or wop. But other dictionaries compiled much later than that also miss the “offensive or insulting word” sense of the noun (the sense that figures in “he was fired for using a vile racial slur beginning with n”). The American Heritage defines it simply as “A disparaging remark; an aspersion” (though I’m assured that a new definition is in the works) and Merriam-Webster defines it as “an insulting or disparaging remark or innuendo.” Needless to say, fag and wop are not remarks, aspersions, or innuendos. Epithet, too , was first applied to explicitly racial and ethnic derogations in the 1940’s in African American writing (it shows up in the 1945 autobiography of Langston Hughes) and began to appear in the mainstream white press in the 1960’s.
were called out for obloquy in earlier eras, the moral tenor of the objections was different. Nineteenth-century condemnations by whites of the use of nigger , for example, were usually focused on the vulgarity of the speaker and insult to his (white) addressees rather than to the target—to use the word, an observer wrote in 1835, demonstrated “a scorn of those rules which every man who respects himself, and is unwilling to be classed with the lowest vulgar, observes.” And the offense, when one was discerned, wasn’t seen as the kind of incivility that undermined the conduct of public life, much less as a matter worthy of serious philosophical or philological scrutiny. Like some other modern social phenomena—the asshole, the (internet) troll, or date rape—the slur as such is a recent addition to our moral inventory. The new sense of slur belonged to a new vocabulary of race and social diversity—the clutch of new isms and morally charged labels (“color blind,” “hate speech,” “racial sensitivity”) that pervaded the culture wars in which slurs, real and alleged, played a central symbolic role. In the course of things, the slur was assigned a new moral etiology: it was condemned not for the ill-breeding or coarseness that the word betrayed in the speaker as for the insult and injury it inflicted on its targets, who were themselves no longer without a voice in the discussions. Indeed, as its genesis in the African American media suggests, the rise of the new meaning of slur coincided with the emergence of a doctrine of linguistic self-determination, which held that every group should have the right to determine what it should—and, even more important, should not—be called. It was inevitable—politically, institutionally, intellectually—that philosophers and linguists would be drawn to examine slurs, bringing insights of their disciplines to bear on an issue at the heart of many of the ambient cultural controversies. 2 And as it happens, slurs also seem to offer an excellent test bed for exploring some longstanding theoretical issues. How do words come by their capacity to express emotion or convey contempt? What can words signify over and above what they say about the world? The questions seem to bear directly on the (^2) Derogative terms had already made their way into the philosophical literature by 1973, when Dummett used them to exemplify the difference between the grounds for applying a concept and the consequences of its application. But his choice of the quaint epithet boche as an example suggested a disinclination to connect these questions to their sociopolitical significance, notwithstanding his personal engagement in these issues.
can’t always be neutralized by the quotation marks or other devices that we ordinarily use to hold the connotations of a expression at arm’s length. Still, I think these approaches are all fundamentally misconceived, for several related reasons. For one thing, people who abstract the slur to a universal type inevitably wind up building in what turn out to be culture-specific properties of the category—indeed, there doesn’t seem to be a lot of awareness of just how much the particularities of slurs owe to local conditions. That has several unfortunate consequences. It leads to a focus on a relatively small number of prototypical slurs, slighting other words which exploit the same principles to create effects that are communicatively parallel to those of slurs but which don’t necessarily raise the moral and social issues that surround the latter. It obscures some aspects of slurs that are marginal to their current sociopolitical interest but that figure centrally in their use. And those effects are compounded by an almost exclusive reliance on semantic mechanisms to account for the behavior and effects of these words. The theories vary, but almost all involve some kind of dedicated semantic machinery—slur-specific entailments, slur-specific conventional implicatures, or slur-specific expressive components—that has no obvious parallels with other descriptive terms. It’s in virtue of such mechanisms that the meaning of the word itself can be made the bearer of the attitudes speakers use it to convey. The result is not just a semantics that is often rococo and sui generis, but a neglect of the social grounding of the phenomenon, as functional explanation is subordinated to formal modeling. Generally speaking, this literature isn’t overburdened with philological detail. I’ll be arguing here not just that there’s no existing semantic theory of slurs that accounts for the basic facts of their use, but that there couldn’t be one. There are aspects of interpretation that can only be created by non-semantic means. One class of these involves implicatures that arise via the recognition of a transparent pretense, such as with most verbal irony—the implicature that rests on the hearer’s recognition that the speaker’s voice is displaced, that he’s speaking the words of a more innocent self or someone else who has seen the world differently from how it plainly is: “He turned out to be a fine friend, didn’t he?” I’ll argue here that an analogous ventriloquism underlies slurs, where the speaker assumes the voice the words of a
community hostile to the target. 4 But once you try to build such an effect into the meaning of the word, you occlude the reasoning that gives rise to its significance and hence are at a loss to understand its function. Baldly put, semantic accounts of these words don’t work, couldn't work, and wouldn’t explain anything if they did. This leaves me with the task of explaining, or explaining away, all the various features of slurs that have seemed to call for specialized semantic treatment. Most fundamentally, what makes them different from other descriptions—if redskin and Indian refer to the same group but differ in their connotations, what could account for the difference other than some arbitrary— which is to say, conventional—lexical feature? Here’s one answer: the words might be prescribed by different conventions, not of the same language-variety, but of different ones, defined over distinct speech-communities. Imagine a dish that Southerners adore under one name and Northerners execrate under another; the two names are perfect synonyms, but the choice of one or the other word carries very different connotations. I’ll be arguing here that this is basically what’s going on with slurs and other words like them. The failure to see this is symptomatic of an endemic problem, a failure to take seriously the nature of conventions as they operate in the rough-and-tumble of sociolinguistic life. The properties of these words that have led people to overcomplicate their internal semantics are really reflexes of the social heterogeneities that underlie the linguistic conventions over which we build our lexicons. In effect, semantic approaches to slurs displace the contours of the sociolinguistic landscape onto the meanings of words themselves, with the result that we discern all sorts of semantic corrugations that aren’t really there. Here’s my thesis in a nutshell: racists don’t use slurs because they’re derogative; slurs are derogative because they’re the words that racists use. Slurs, Derogatives, and Prejudicials The difficulties in characterizing slurs as a linguistic type emerge as soon as one tries to pin down the notion, and the terminological variation here can cloud things further. The definition offered by Mark Richard (2008) is as thoughtful as any: (^4) One seventeenth-century word-list defined ventriloquist as “one that hath an evil spirit speaking in his belly,” which is not far off as a description of the use of slurs. (OED)
class—is not just the idea that they’re offensive, but the particular moral and political color of the offense they give. At this point we could go in either of two ways. We could think like lexicographers and try to frame a definition that does justice to the word slur as it’s actually used, which would preserve its immediate moral charge. But in that case we have to acknowledge that we’re not dealing with a universal linguistic type. Natural language might provide some special apparatus for conveying derogation, but not just for conveying unfair derogation of certain kinds of social groups; semantically speaking, redskin is not going to look different from Ratzi or stinkpotter. Slurs as such, then, have no independent existence as a linguistic category, and whatever moral judgments we make about them are independent of their purely linguistic properties. This point hasn’t always been appreciated; one sometimes sees people jumping from ethical to linguistic conclusions about the words. Hornsby (2001), for example, says that what she describes as derogatory words are “useless”: Derogatory words are ‘useless’ for us. Some people have a use for them. But there is nothing that we want to say with them. Since there are other words that suit us better, we lose nothing by imposing for ourselves a blanket selection restriction on them, as it were. If “derogatory word” here means roughly the same thing as slur , then Hornsby’s claim could be defended, depending of course on who “we” are supposed to be, but it couldn’t have any linguistic relevance. If “derogatory words” are simply words that disparage etc., then the claim is obviously false—surely one could find a use for a word like facho or clamhead, though in that case we’d probably demur from calling them slurs. And similarly for the claims that the assertions containing these words are essentially false (Hom 2008) or not truth-evaluable (Richard 2008), (Hedger 2012). There are no interesting linguistic conclusions to be drawn from any moral characterization of slurs, and vice-versa. The alternative approach to defining the class is to say, “Okay, let’s forget about slurs as such and focus on the specifically linguistic features of these words without reference to cultural specifics or subjective judgments.” We’d need another term, then, one that comprehends not just redskin but Ratzi and stinkpotter. For that purpose we could draft derogative , and I hereby do so, but with the understanding that “derogation” is just a convenient label for an assortment of negative, unflattering, or insensitive attitudes that shade off in all directions, from contempt to ridicule to superiority to patronage—derogatives, we’ll see, are no more a well-defined linguistic
class than slurs are. In fact there’s no reason to stop there; we’ll want to expand the class to cover all the words that convey an affective judgment of any tone or valence about the members of a category, subject to some other conditions that I’ll come to in a moment. I’m going to use the term prejudicial for these, since it implies only that the speaker has a preconceived opinion of a category, either positive or negative. 6 Like Richard and some others, I want to add another clause to the definition to stipulate that prejudicials require coextensive alternatives that don’t convey the same evaluation. Here’s how Richard puts this, only substituting prejudicial for his slur : Every [prejudicial], so far as I can tell, has or could have a “neutral counterpart” which co- classifies but is free of the [prejudicial’s] evaluative dimension Not everybody imposes this second requirement—it’s more often mentioned by philosophers like Richard and Hornsby than by linguists. That may be because it’s hard to think how one could work this stipulation into a semantic description, at least as linguists understand the notion—“possessing a neutral synonym” doesn’t feel like a possible lexical feature. In fact this is a necessary stipulation, though it requires some revision. Start with “neutral.” You wouldn’t describe pedophile and fascist as “neutral” words, since they trail so many negative and disapproving connotations, but those are the only words available if you want to say something positive about either category. In that sense, pedophile is not a slur in the sense that fag is (though someone who calls you a pedophile may very well be slurring you in another sense of the word). In fact this criterion excludes a number of words that are highly pejorative, such as Uncle Tom, Jewish American princess, and airhead , which come under the distinct category of hybrid words, as we’ll see.^7 Then too, the term “counterpart” seems to put the prejudicial and its denotational synonym at the same level, as if we were talking about the contrast between an evaluative term like weirdo and a more neutral item like eccentric , where there’s no default (^6) By the noun prejudicial I intend something more specific than a “prejudicial word” in the ordinary sense, which can be any word that characterizes something in a negative way, such as calling a subsidy to failing businesses a “bailout.” Note also that a prejudicial in my sense can convey a positive attitude, as well. (^7) I don’t think of these items as slurs, since they reproach their targets on the basis of their individual behavior, not (just) on the basis of their membership in a class; see below. But you may disagree, in which case you can argue that the account of the slur that I offer here isn’t lexicographically complete.
nationality and cruelty are “involved in the very meaning of the word; neither could be severed without altering its meaning.” This is a common and intuitive picture of things; as the psychologist Leo Rappaport (2005) wrote: Ethnic slurs serve as a kind of shorthand way of referring to the negative qualities associated with any particular group. They are quite specific. Hispanics might be called “spics” and Jews “kikes”; each term would stand for a specific c cluster of traits assumed to be typical of Hispanics and Jews. Christopher Hom (2010) has worked this picture up in some detail. As he tells it, a slur implies both a stereotype and its consequences, so that …the epithet ‘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. This leads to the satisfying conclusion that to describe someone as a chink is to make a false assertion, because there are no chinks; the word has an empty extension. But it’s a very strange sort of meaning for a word to have—with the possible exception of some specialized legal terminology, it’s hard to think of any other descriptive word whose meaning has the structure “Should be A because of being B because of being C.” And the fact is that when we call somebody a chink we don’t necessarily imply any of these things about him. Even if that farrago of negative traits were implicit every time chink was used, the individual referred to might be entirely exempted from the charge—“He’s a chink, yeah, but he’s terrible at math, he’s a careful driver and his collars are always dirty.” Moreover, as several critics of Hom’s view have observed, the derogation is present even when you say that so-and-so is not a chink or ask whether he’s a chink, where no predication is involved. If the predication of a stereotype figured in this at all, then, it would apply only to the class to which the referent belongs—to say that “Yao is Chinese and the Chinese are typically devious, slanty-eyed…” and so on. And when we use the word to make an assertion about Chinese in general, as in “That dorm has a lot of chinks,” we’ve said something like “That dorm has a lot of Chinese and the Chinese are typically devious, slanty-eyed….” But once again, that simply isn’t the way we interpret assertions like these. As various people have observed, when someone says something about chinks, redskins, flacks, or whatever, we can’t contest their remark by denying the aptness of the stereotype that the word putatively conveys, even when the assertion is made about the class itself:
- a. Juan is a kraut. b. That’s not true/That’s false. b. #German people are not bad. (McCready 2010) 2.Two states just voted to allow fags to marry. a. That’s wrong, three did. b. #That’s wrong, they voted to allow family-oriented same-sex couples to marry.^8
- Do I need to hire a flack? a. No, you need a better publisher. b. #No, you need a frank, independent-minded publicist.
- I hear the bill is supported by the libtards. a. No, most of the supporters are conservatives. b. #No, the supporters hold sound and compassionate progressive views.
- Does the regime encourage free enterprise? a. No, they support socialism. b. # No, they encourage dog-eat-dog capitalism. Potts (2005) describes this phenomenon by saying that the content of such words is “scopeless”; I prefer to say that it simply isn’t there. When we hear somebody described with a prejudicial, that is, we don’t interpret the speaker as having predicated anything about either him or the class he belongs to over and above his membership in it. (^8) I’m excluding here cases of what Horn (1985) describes as metalinguistic negation, where the speaker contests the appropriateness of someone’s use of a term. He gives examples like “I beg your pardon: Lee isn't an 'uppity nigger' — she's a strong, vibrant black woman.” The intuition that sentences like “He’s not a chink, he’s Chinese” involve metalinguistic negation is supported by other observations. For example, Horn observes that metalinguistic negation is disfavored when the negation is expressed by “it is not true that” or “it’s not the case that”: ?It's not the case that the dog SHAT on the carpet—he DEFECATED on it. Note that this holds as well for the evaluative import of derogatives, with the inclusion of “in fact” reinforcing the point: It’s not the case that John is a chink; (in fact) he’s Japanese. ??It’s not the case that John is a chink; (in fact) he’s terrible at math and his collars are always dirty. Similarly, McCawley (1991) observed that some “echoes” that express strong disagreement can’t be used metalinguistically: In a pig’s eye we had intercourse, we never so much as kissed/?we fucked! In a pig’s eye John’s a redskin! He’s from Finland/?*He’s teetotal, mild-tempered and articulate. (Can only mean John is not an American Indian)
that surround them? In fact I’ll be arguing here that we don’t have to treat these words as anything other than the descriptive terms that by all appearances they are. In any event, this account fails empirically. The stereotype associated with the category denoted by a prejudicial isn’t part of the word’s meaning in any guise. Here’s another way to make this point: instead of asking whether these ascriptions can be directly contested, let’s ask when they can be informatively predicated of the term. Suppose cruelty was actually inherent in the linguistic meaning of boche , whether as descriptive content, a presupposition, or a conventional implicature. Then the very fact of using the word would introduce the assumption that Germans are cruel into the conversational ground. But in that case, why doesn’t an assertion of “The boche are cruel” strike us as conversationally tautological? And similarly for “Commies are devious”—or “godless,” or “fanatical,” or “ruthless,” or whatever you take the stereotypically invidious traits of communists to be—which a militant anti-communist wouldn’t find so redundant as to not need saying. Someone who speaks about free enterprise presumably endorses the claim that market capitalism is the fairest and most productive economic system, and someone who refers to the estate tax as the death tax implies that he believes that it’s an unfair exaction. But we don’t sense a tautology when someone says, “Free enterprise is the most equitable economic system” or “The death tax is unfair.” (Indeed, Nexis turns up 1099 news articles in which “death tax” occurs within three words of “unfair,” virtually all of them uttered by people who believed they were saying something informative.) Some of those traits may well be hovering in the background when you use a prejudicial, but they aren’t part of the semantics of the word. 10 Prejudicials and Prejudice These arguments suffice to show that descriptive semantic accounts of prejudicials can’t explain their use, even at the expedient of introducing new and more intricate machinery to get them to come out right. By themselves, though, they don’t tell us why such accounts couldn’t be right. People do unquestionably make use of stereotypes about social groups, and while the word stereotype is generally used now only in a disparaging way, stereotypes are among the cognitive (^10) As we’ll see, in this way prejudicials are different from true hybrids, where we do evoke a Duh! reaction when we independently predicate their evaluative components of them, as in “Valor is commendable,” “Bullies aren’t nice,” “Fleecing someone is unfair.”
shortcuts that are necessary to the way people makes sense of the social world. So why couldn’t such stereotypes be part of the meanings of slurs and other prejudicials? Return to the descriptivist picture of these words. Earl sees a Chinese family moving in next door says to his wife, “The neighborhood’s filling up with chinks.” On the descriptive view, Earl has committed himself by that utterance to saying or conventionally implicating that the new residents share (or tend to share) a number of unwelcome traits in virtue of their ethnicity. As Hom tells the story, whether Earl knows it or not, his use of chink commits him to the proposition that his neighbors are increasingly people who “ought to be subject to higher college admissions standards, and ought to be subject to exclusion from advancement to managerial positions, and…, all because of being slanty-eyed, and devious, and good-at-laundering, and …, all because of being Chinese.” But while we can assume that Earl holds Chinese people in contempt, it’s unlikely that either he or his wife had in mind such an extensive and detailed bill of particulars about them, or was even aware of most of it. How could all of those attributes work themselves into the meaning of his utterance without passing through his consciousness on the way? To get all these properties into play, Hom appeals to what he calls “combinatorial externalism”: really, he says, the meaning of chink is determined not by Earl’s state of mind, but by the “social institutions of racism,” which pervade every utterance of the word with all the anti- Chinese prejudices that are in circulation. It’s hard to know what means—do we have to consult a historian of racism to find out what Earl really said, and would even he agree that Earl is semantically, and morally, responsible for endorsing a view on Chinese college admission standards that Earl has never given a thought to? What could possible count as evidence for that conclusion? So while some kind of externalist story is required here, it’s a mistake to draw a parallel with the familiar narratives about lemons and elms that mentions of externalism typically bring to mind. Those are stories about fixing the reference of a term: we collectively assign authority to various experts to tell us what our utterances of dogwood , sonata form , and quantitative easing refer to, and then we can succeed in referring with the terms whether or not we can individually pick out the things they denote. Ignorance of the meaning is no excuse. If your gardener promises you a rose garden and plants peonies instead, he can’t get off the hook by saying, “They all look the same to me.”
that the petitioners shouldn’t need to produce any evidence for their claim, since the argument could be made on prima facie grounds. Inasmuch as redskin had been used by white Americans from the time of Fenimore Cooper up to the mid-twentieth century, what attitudes could it have conveyed other than disparaging or condescending ones? What else was out there? So suppose we took a middle ground, and say only that Earl’s utterance conveys a set of traits that most people commonly associate with the Chinese—think of this as the cluster theory of stereotypes. Calling someone a chink, on that account, might not convey anything about college admissions standards, but would imply something about deviousness or clannishness. Even in that case, though, all we would require of the term is that it somehow get us to a certain cluster of traits; it doesn’t have to entail them all by itself. But semantic approaches wind up doing just that. Independently existing stereotypes would have to somehow replicate themselves, redundantly, in the semantic gene plasm of a derogative name for the group. And in that case, we’d be left with the puzzle of why derogatives are always derogative. After all, racial stereotypes may be generally deplorable, but they’re rarely categorically negative—they’re typically compounded of contrasting or inconsistent traits. As the sociologist Ali Rattansi (2007) observes: Stereotypes… reveal contradiction and ambivalence rather than completely invariable hostility or admiration toward other groups… Attitudes toward Asians in Europe and the US, for instance, reveal admiration for supposed community unity, thrift, ambition, hard work, respect for education, and ‘family values,’ but also hostility for insularity, suspicion regarding their loyalties, to the Western nation-states in which they have come to live, and a sense of superiority toward their more ‘backward’ cultures…” The positive traits of these stereotypes are the more genial manifestations of the conflicted racial attitudes that also shape the negative ones. (That point eluded the management of the Washington Redskins when they insisted with Fenimorian condescension that their mark was a respectful tribute to American Indians’ courage and tenacity.) But chink doesn’t convey any ambivalence about the Chinese; as Jeshion (forthcoming) notes, the word is “unequivocally and exclusively contemptuous.” One might argue that the slur semantically selects only the negative features of the stereotype. But just about any feature of a stereotype can be regarded as positive or negative on a given occasion. On Monday it’s “You have to give it to the chinks; they work hard”; on Tuesday it’s “No wonder those damn chinks all get A’s—they don’t do anything but study until two in the morning.” Whatever ethnic traits a given utterance of the word chink brings
to mind, if any, are just those that are contextually consistent with an antecedent attitude of condescension or contempt. The attitude comes first. To suggest that invidious stereotypes are the source of bigotry is credit the bigot with a weirdly misplaced rationality, as if his antipathies were sound logical conclusions drawn from what happen to be false premises. But racism, anti-Semitism, homophobia, and xenophobia are fundamentally rooted in the basic fact of alterity rather than the stereotypes that people cite to justify or rationalize the attitude—of contempt, loathing, fear, or condescension, as the case may be. 12 As Walter Lippmann wrote in Public Opinion , the 1922 book that introduced the notion of the stereotype into American intellectual discourse, “[T]he perfect stereotype…precedes the use of reason; is a form of perception, imposes a certain character on the data of our senses before the data reach the intelligence.” That’s why those stereotypes can vary from one person to another or change over time without dramatically shifting the significance of a slur. At one time or another kike has conjured up images of Jews as Christ-killers, as money-grubbing tradesmen, as clannish and superior, as conspiratorial international bankers, as deviates and radicals, as Stalin’s “rootless cosmopolitans,” and even, in the 1930’s, as possessing a duplicitous genius for basketball. But it isn’t as if the word has had seven or nine or however many meanings, or as if there’s any semantic misunderstanding between the speaker who denounces kikes with Abbie Hoffman in mind and the listener who agrees with him while thinking of George Soros. The animus transcends the various pretexts we give for it. Hornsby (2001) puts this point nicely: Racist speakers no more convey specific bits of ideology whenever they use racist words than our imagined British speaker imputes barbarity to every German person she described with ‘Boche’. If speakers’ involvement with the ideology went as deep as it would need to in order to be implicit in their very use of words, then common understandings would be difficult to preserve. True, certain derogatives may bring specific stereotypes to mind. Sometimes the stereotype is suggested by the name itself, particularly when it’s derived from a description— (^12) As often as not, of course, group stereotypes have some grounding in fact. On occasion, indeed, they may provide a valid justification for conceiving a negative attitude toward the group and possibly for a prejudicial term that expresses it, though in that case it won’t be considered a slur (think facho , for example). Even then, though, the stereotype figures only as the perhaps post-hoc justification for the attitude the term expresses, not as part of what it conveys.
focused on the use of the words to disparage and injure their targets, making allowances only for their reclaimed use by those they target. 14 But only the tiniest proportion of the utterances of these words are addressed to members of the group they target or are used as direct insults. They’re far more commonly exchanged between people who share, or make as if to share, the attitudes they convey—to create solidarity in a shared sense of resentment or superiority, to underscore the normative values of the group, to distinguish themselves from genteel speakers who are too fastidious to use the words, or simply out of a schoolyard pleasure in complicit naughtiness. (In this regard, too, slurs resemble vulgar descriptions like asshole .) In fact the tonal differences among slurs for the same group very often have as much or more to do with the distinct self-conceptions they presuppose than with what they convey about their targets. The fact that someone chooses to refer to blacks as spades rather than as jungle bunnies, spooks, niggers or coons tells us less about what traits he ascribes to blacks than about how he sees himself and his listeners. But to say that derogatives convey attitudes rather than descriptions doesn’t necessarily mean their expressive force is built into their conventional meanings, as expressivists usually assume. In fact, many of the same arguments that make a semantic treatment problematic for the descriptive view create problems for expressivist views, as well. The attitudes evoked by prejudicials, like the stereotypes they can bring to mind, have to be independently out there, and for that reason it’s unnecessary to reproduce them in the semantics of the term. A speaker’s use of a slur can’t convey contempt unless the group it targets is already regarded contemptuously. In some contexts, in fact, describing someone with the default or neutral name for the group he belongs to will be sufficient to evoke an ambient derogatory attitude, particularly when his membership in the group has no independent conversational relevance: “She’s been seeing a (^14) An exception is Miščević (2011), writing in response to Richard’s definition of a slur as “a device made to denigrate, abuse, intimidate, and show contempt.” Miščević agrees that that definition fits the “second- person uses” of slurs, but adds, “nowadays, 3rd person uses [of slurs] seem at least as common as 2nd person ones. People have become more cautious, the use pejoratives behind the back of the targets, so the intended perlocutionary effect cannot be directly to offend.… [the] perlocutory intentions” in this case have to do with eliciting solidarity against the target.” Actually, third-person uses of slurs are probably hundreds of times more frequent than their uses in direct address, and the chief purpose of these is not always to elicit “solidarity against the target,” as we will see.
black guy from another high school”; “I had a gay kid come in to whine about his grade the other day.” Slurs add to that, of course, but not by contributing any novel attitudes. What they do, rather, is to make the speaker’s attitude explicit and specific, carving out a particular feeling-tone from among the various tunes in circulation. Moreover, the feeling-tones that these words convey—the affective attitude toward their targets and its connection to the social identity of their users—are generally much too complex and socially embedded to be rendered in meaning-formulas like “despicable because Indian” or “Jewish and used to refer to Jews with hatred or hostility,” which reflect the simplistic picture of the word’s affect that’s implicit in the word slur itself. How should one describe the affect associated with the hipster’s spade , for example? It implies the hipster’s self-conception as the “White Negro,” the title of Norman Mailer’s famous essay, and the attachment to jazz, marijuana, and “cool,” which the hipsters can take credit for introducing into the mainstream cultural discourse. The word certainly wasn’t used contemptuously—Ken Kesey described it as a “term of endearment,” and Chester Anderson, as I noted, went so far as to describe spades as “our spiritual fathers.” But the term was also dehumanizing and obtuse; as one black critic described it, “those fay cats… don’t want us to be Uncle Toms, but they still want us to be spooks. They don’t really dig us as a people; they just dig us for our music and our pot.” (Foreman 1998) Those passages suggest the range of complex and conflicted attitudes at work here. However you describe the evaluation implicit in spade —as derogative, condescending, romanticized, or simply a slur—the label won’t begin to capture the word’s implications or affective force—and here, as often, the intended affect (“a term of endearment”) isn’t necessarily what the word conveys to a neutral listener. Like most of the words that do the heavy lifting in the social world, spade is what Empson called a “compacted doctrine,” which stands in for a dense and brindled chunk of lived reality. That doesn’t mean that the attitude expressed by spade is descriptively ineffable, but like the other complex keywords that social historians and ethnographers wrestle with, it isn’t easy to eff. The most we can expect the lexical meaning of the word itself to do is to index the sociohistorical context in which the attitudes and identity of the word’s speakers are formed, and whose point of view the speaker assumes in using the word. The externalism of prejudicials isn’t grounded in deference but identification.
