A recent Facebook post from a friend on the Dr. Laura Schlesinger “n-word” episode sparked an interesting discussion. Like me, he is a philosopher (but unlike me, he is paid to be one). He stated the thesis that no one should use the n-word, not even African-Americans. His stated reason for this view, which I think is reasonable and with which I agree, was as follows. The n-word fails to denote any object; it is in a class with nonsense words such as “toovy” and “teavy.” Given that the word is normally used in strongly pejorative contexts, as a way of degrading the individuals to whom it applies to something like sub-human status, it follows that, since African-Americans and persons of color are fully human, the word fails to pick them or anyone else out. Whatever a n—— is, there just is no such being.So, in actual use, paradigm protocol sentences such as “So-and-so is a n—–“ fail to have a truth value; they are not simply false, but senseless. My friend quoted the authority of Rudolf Carnap in support of this latter claim. So, since the word fails to denote, it cannot a fortiori have any connotation, negative or positive. So the word should not be used. It is senseless to use it.
This did, and still does, strike me as a mistaken analysis of the n-word. I was, and still am, puzzled by the claim that the word fails to denote. I take denotation here to be roughly the same thing as semantic reference: if a word, under some specified interpretation, picks out some entity or entities, the word refers. For formal languages, interpretations are usually stipulated; for natural languages, interpretations are read off typical language use. I take my friend’s claim to be equivalent to saying that the n-word does not, in typical language use, pick out any entity whatsoever. It is this claim which I believe to be wrong.
My friend’s analysis presupposes that in use the n-word is a count noun, like “chair” or “tree,” such that it could occupy the predicate position in a protocol sentence of the form “x is a y.” It is this presupposition, though, that I believe is incorrect. Well, not entirely incorrect; I think the word is also used as a count noun. I just don’t think that the paradigm cases that concern most people are uses of the word as a count noun. Instead, the uses that are the most paradigmatically bigoted and offensive are uses of the word as a slur. Slurs differ from count nouns in that slurs are used, not to describe a class or as a means of attributing class membership, but instead to make individual reference, like a name. This function of slurs is captured by the common English phrase “calling someone names” as a euphemism for addressing someone with slurs. Slurs are not proper names, like Brian Cubbage—a highly relevant fact, about which more below—which can make it easy to mistake them for count nouns. For that matter, slurs can be used as count nouns as well, muddying the waters even further.
So how can the n-word, considered as a slur, refer in a situation where the n-word, considered as a count noun, cannot? For that, we need to consider the reference of names, since it is in virtue of their semantic reference that slurs are like names. Perhaps the single most famous—and in my opinion most compelling—analysis of this subject is to be found in Saul Kripke’s (1972, 1980) Naming and Necessity. Kripke takes aim at that other most famous theory of the reference of names, Bertrand Russell’s theory of names in “On Denoting” (1905). Russell’s theory holds that proper names are disguised definite descriptions. On that theory, proper names are actually synonymous with definite descriptions, such that one can salva veritate replace the name with the appropriate definite description in sentences where names occur. Kripke famously contends through a series of arguments that this account of proper names doesn’t adequately capture the work that names do. His reason is that no single description, or even a cluster of descriptions, can be said to describe necessary properties of an object, such that no set of descriptions can be said to fix reference necessarily. Names, however, fix reference necessarily by stipulation: People use names to pick out an individual regardless of her properties, contingent or necessary. Using Richard Nixon, then the President of the United States, as an example, Kripke argues:
Suppose that someone said, pointing to Nixon, ‘That’s the guy who might have lost [the election]’. Someone else says, ‘Oh no, if you describe him as “Nixon”, then he might have lost; but describing him as the winner, then it is not true that he might have lost’. … The first man would say, and with great conviction, ‘Well, of course, the winner of the election might have been someone else. The actual winner, had the course of the campaign been different, might have been the loser, and someone else the winner; or there might have been no election at all. So, such terms as ‘the winner’ and ‘the loser’ don’t designate the same objects in all possible worlds. On the other hand, the term “Nixon” is just a name of this man’. (41)
To describe this particular reference-fixing function, Kripke introduces the concept of a rigid designator. “Let’s call something a rigid designator if in every possible world it designates the same object” (48). Kripke’s position is that a proper name like “Richard Nixon” rigidly designates an individual who actually was President in 1972, but might not have been; it even designates him regardless of how things might have been with him, even if his parents had never met and he never existed. It even designates him in a possible world in which he existed but wasn’t actually called ‘Richard Nixon.’
Kripke’s full argument for his theory that names are rigid designators is wide-ranging, and I won’t address its full extent here, much less its subsequent development into a “two-dimensional” theory of reference in later philosophy of language. Suffice it to say that I find Kripke’s picture convincing. On it, names are more akin to demonstratives like “this” and “that” than they are to count nouns. Unlike demonstratives, though, whose reference is completely context-bound, a name undergoes an “initial baptism” that influences but does not entirely predetermine its subsequent referential history; a thread, however tenuously stretched, ties the name to its referent across multiple contexts. On this theory, it is not particularly difficult for a name to have a reference; in fact, it’s almost too easy. As long as I am able to baptize it and its subsequent use does not break the direct referential tie, I can use pretty much any word as a name, even if the word had some other meaning at the time of initial baptism or (especially) if it had no other meaning at all.
What does this theory have to teach us about slurs? Slurs like the n-word, like names, have a referential history. In fact, it is precisely their history as terms of direct reference to people that deserves attention, understanding, and condemnation. It is precisely because of the fact that slurs hit their referential targets, and that they have the targets that they do as a result of a history of specific actions of hatred, oppression, and violence, that they are able to hurt anyone. The n-word is baptized in blood, in a history of lynchings, cross-burnings, and institutional discrimination from which we have not distanced ourselves to the extent we think we have. Slurs can, with work, be reappropriated by their customary targets and, their referential history broken, rebaptized anew, as has recently been done with the word “queer.” The fact remains, though, that such acts of reappropriation take work, and they are not always successful.
Yet, as I mentioned before, slurs are not quite proper names. In ordinary use, if someone says “Brian Cubbage,” it is abundantly clear without knowing anything about her state of mind or other elements of her context that she means to refer to a single person—the name functions as a “rigid designator” in Kripkean parlance. Not so with slurs. If I were to overhear someone on the phone, for example, say to the other party “Hey, n—–,” I wouldn’t be able to infer the specific identity of the person on the other end of the conversation without more information. Still less would it make sense to tell such a person that she had mistakenly applied the slur in the same way that it would, for instance, make sense to tell someone that the person she is calling Brian Cubbage is really Brian Boitano. It seems perverse to tell someone that she is mistaken in calling someone a n—– because, well, there are n—–s, but this person isn’t one of them. The perversity isn’t because the n-word lacks reference, but instead because it misses the speaker’s referential impact, the point of why the speaker is using the n-word in the first place.
I offer that slurs like the n-word have an intermediate referential status somewhere between proper names, count nouns, and demonstratives. Like names, they refer directly to individuals, but unlike names, they do not do so rigidly. Instead, they do so in a way that is meant to reduce and deny the individuality of the person designated by them, even while they depend for their effect on their hearers knowing to whom they refer in any particular context of use. In fact, they only enjoy relative context-independence; they are linguistic actions par excellence, a way of doing something with a word. They are like demonstratives in this way. Slurs are reductive: They pick out an individual, but they do so only to reduce her to just another member of a group to which an indefinite number of others belong. In this way, they are like count nouns. Slurs occupy a unique, complicated position in the language.
It is no wonder, then, how slurs can gain so much power and can cause so much damage.
If Dr. Laura’s senselessness has done anything besides give Sarah Palin an opportunity to demonstrate that among the numerous things she doesn’t understand is the First Amendment, it’s given us an occasion to reflect on why it is that, in our putatively “post-racial” times, racial slurs still have the power to hurt and pollute.