Racial epithets, pragmatics, and semantics

« previous post | next post »

Those seriously interested in the meaning and the politics of racial epithets (as some of the commenters on Pakigate, Sootygate, Gollygate seem to be) should take a look at a paper called "The semantics of racial epithets", published by Christopher Hom in The Journal of Philosophy CV [= 105], no. 8 (August 2008), pp. 416-440. This is a technical paper in philosophical semantics (it's philosophy, not linguistics; and let me say that I do not necessarily endorse the view that it defends). Hom outlines its aim on his website thus:

Racial epithets are derogatory expressions, understood to convey contempt toward their targets. But what do they actually mean, if anything? While the prevailing view is that epithets are to be explained pragmatically, I argue that a careful consideration of the data strongly supports a particular semantic theory. I call this view Combinatorial Externalism (CE). CE holds that epithets express complex properties that are determined by the discriminatory practices and stereotypes of their corresponding racist institutions. Depending on the character of the institution, the complex semantic value can be composed of a variety of components. The account has significant implications on theoretical, as well as, practical dimensions, providing new arguments against radical contextualism, and for the exclusion of certain epithets from First Amendment speech protection.

Thus Hom is offering a reasoned case that it is best to see the denigratory character of racial epithets as built into their actual conventional meanings, and not just as a possible concomitant of some of their occasional uses. (Many of commenters seem to align with this view, though they tend to just assert it and call any other views absurd, rather than present arguments.)

At the beginning of the paper, Hom sets out his aims thus:

There are two competing strategies for explaining how epithets work, one semantic and the other pragmatic. According to the semantic strategy, their derogatory content is fundamentally part of their literal meaning, and thus gets expressed in every context of utterance. This strategy honors the intuition that epithets literally say bad things, regardless of how they are used. According to the pragmatic strategy, their derogatory content is fundamentally part of how they are used, and results from features of the individual contexts surrounding their utterance. This strategy honors the intuition that epithets can be used for a variety of purposes, and that this complexity surrounding epithets precludes a univocal, context-independent explanation for how they work. Neither view is without difficulty, although to many the pragmatic strategy is prima facie more attractive. I shall argue, however, that the semantic strategy actually fares better on a number of criteria. In doing so, I shall motivate a particular semantic account of epithets that I call combinatorial externalism. The account has significant implications on theoretical, as well as, practical dimensions.

In something of a departure for Language Log, which generally deals with material less technical than Hom's paper, I open the comments space below for use by people who have read the paper, please, not just for opinions stimulated by the above summary, or (pretty please!) random observations about racism or nasty words. I'll enforce the policy to the extent that time permits.


  1. Christopher Hom said,

    February 9, 2009 @ 12:35 pm

    I would also recommend the following:

    1. Potts, C., 2007, "The Expressive Dimension", Theoretical Linguistics, 33(2): 165-197.

    2. Williamson, T., forthcoming, "Reference, Inference and the Semantics of Pejoratives"

    3. Richard, M., 2008, When Truth Gives Out, Oxford University Press, Ch 1.

    4. Lepore, E., and Anderson, L, ms., "Slurring Words".

  2. Mark said,

    February 9, 2009 @ 12:36 pm

    Very useful paper. But I don't resonate with Hom's intuitions about his key examples of (supposedly) literally true sentences (these are numbered 13 to 27 in the paper). For instance:

    15. Chinese people are not chinks.
    17. There are no chinks; racists are wrong.

    Yes, these can readily be understood, but to me they at least to some degree have the feel of "I'm not hungry; I'm starving!"–what Larry Horn calls metalinguistic negation. And many of the other examples are in attitude-ascriptive contexts, where it's not clear that only truth-conditional content is at work. I'm not sure there is even a fact of the matter whether pejorative content in such epithets is truth-conditional (in contrast with terms like "racist" and "stud", where an account like Hom's seems exactly right).

  3. Mihai Pomarlan said,

    February 9, 2009 @ 12:46 pm

    After reading through the first 12 pages, here are some observations. Ok, I still have ~20 more pages to read, but I have objections to Hom's objections to the pragmatic approaches right now.

    One, why is so much emphasis put on truth-value of sentences, or their equivalence in simple truth-theory terms? While I am not a philosopher of language, I think I can ask why it seems easy to generate statements of undecided and even undecidable value (for a rigurous example, "the axiom of choice is true"; for a less rigurous example, "at noon tomorrow it will rain" or "the die will fall on six at next roll"), and in general situations when whether the sentence is true or false is less important than what can be implied about the speaker's beliefs/attitudes ("the Bible is literally true"), and all sorts of silliness thereafter ("there's a man called Herbert Pickington in Beijing" which even if true would be a lie because I don't know for a fact that the named person exists).

    Two, the dismissal of "radical contextualism" is incredibly casual ("[Randall] Kennedy presents many detailed examples, but fails to specify the determinate rules for calculating the contents for any given context of utterance"). This establishes that racial epithets are not exactly like indexical words, upon which I suspect grammar enforces clear rules, but does not, logically, rule out that a system of rules can exist, and it is Hom here that surrenders early.

    Three, I fail to see how you can get from

    "Timothy Williamson advocates pragmatic minimalism and holds that while derogatory content is non-semantic, it is specifically determined in each content as a result of conventional implicature"


    "the minimalist account generates the unintuitive result that certain racist claims are trivially true"

    because, once you allow different groups to endow the words they use with different meanings, what is said in one group by a sentence need not be conveyed by the same sentence to another group. To use his example, "Chinese are chinks" might mean "Chinese are chinese" if some group of people honestly uses "chink" as a neutral synonym. Language is a matter of convention after all, and it is not the case that all conventions are universally enforced among speakers of a language. This observation still holds for the "Indirect Quotation" test.

    Four, a lot of significance is attached to the derogatory, uhm(*), force with which the word "nigger" is associated. It is treated as a somehow special case, when in fact this need not be the case. I am pretty sure that there are third world countries with some competent English speakers (like my own for instance) where the word is nothing special as racial slurs go. We don't have the same history with it. Dare I scream Amerocentrism here?

    (*: If it's a force, how many Newtons? Sorry.)

  4. Vincent said,

    February 9, 2009 @ 1:21 pm

    What I find extraordinary about the paper is its non-awareness of differing times and places. It assumes that the time is now and the place is the United States.

    A reader from England old enough to remember a time when some of these “epithets” could be used innocently, either in everyday speech or in literature (e.g. Huckleberry Finn, The Nigger of the Narcissus); and who has learned rhymes like “Eeny Meeny Miny Mo, Catch a nigger by his toe”, or phrases like “nigger in the woodpile”; and has seldom heard “nigger” used in other contexts, except by hearsay through media whose aim is to inflame controversy; such a reader may discover that his use of language is to be constrained by hateful situations far away, and even made a criminal offence.

    When living in Jamaica, I visited a man in his seventies, my wife’s cousin, who had been trained in London for his career in the Jamaican constabulary. He referred to himself as a negro, neither humbly nor proudly, for that’s how he had been referred to in London and saw nothing wrong in it.

    As the husband of a Jamaican and as a volunteer who works with elderly people, I accept that many whites in the UK above a certain age have retained their prejudices. My aunt, in her eighties, was one, and refused to see me again after my marriage. It didn’t make her hateful, just unable to adjust.

    But the paper, in the options it presents, fails to distinguish use of language from hateful intent and genuine offence. The result is to create a climate of fear, as in the following example. My wife’s boss, in an NHS meeting which my wife attended, inadvertently used the expression “nigger in the woodpile” and it was only the stunned silence which alerted her to what she had said. She trembled, she shed tears, her stammered apology was embarrassingly fulsome. Her distress had nothing to do with possible hurt feelings (there were none, at least in my wife's case) and everything to do with fear of official complaint and its consequences.

    The paper by Hom seems to reflect similar political nuances of language use, ignoring the effect of time or place on those with long memories. In a paper on physics one might assume universality, whether the readers were American or British, regardless of the century they were born in, but it’s surely a sophistry, in linguistics, to simulate this by selecting a particular time and place as a universal benchmark on which all usage is to be judged.

  5. Timothy Martin said,

    February 9, 2009 @ 1:27 pm

    Hom seems to beg the question in his final objection to Fregean minimalism.

    …epithets and their NPC’s express the same sense. Thus, the following identity statements express the same thought:

    a. African-Americans are African American.
    b. African-Americans are niggers.

    however, while the first appears to be trivial and knowable a priori, the second does not. Competent English speakers are rationally compelled to accept (a) as trivial, while most would reject (b) as non-trivial, racist, and false.

    How do we know that a. and b. don't have the same meaning, until we agree on a way of defining racial epithets? Isn't that kind of the whole point of this paper? Hom seems to appeal to the gut insticts of "competent English speakers" to convince us that a. and b. are not the same, but it's not gut instincts that we need – it's arguments.

    One last comment – and this isn't very important to the paper as a whole – but the sentence "derogatory expressions like ‘That’s gay’ seem antiquated, juvenile, and almost infelicitous." makes it seem like Hom isn't very familiar with how the younger generation uses English. Maybe there are regional differences, but in my neck of the woods, "that's gay" is very much alive and well, and only juvenile in the sense that it's young people who say it.

  6. Ivan said,

    February 9, 2009 @ 1:41 pm


    What I find extraordinary about the paper is its non-awareness of differing times and places. It assumes that the time is now and the place is the United States.

    Yes, this seems to be one of the main weaknesses of the paper. As far as I can see, the author says nothing about whether his remarks and conclusions are supposed to apply specifically to modern American English, to English in general, or perhaps more broadly. Throughout the paper, I read many claims that seem to be true for modern American English and for the mainstream Anglo-American society, but which are false for at least some other languages and cultures.

    I am from former Yugoslavia, and as you might imagine, South Slavic languages have their own rich inventories of ethnic and other slurs, so I have some relevant insight here. I can easily provide counterexamples from my own native language for most of the "uncontroversial features" listed in Section IV. In fact, some of them sound highly suspicious even for American English. For instance, the author's example used to illustrate the claim #6 seems to me like a counterexample to the claim #3 (I also have the impression that he gets the history and the present derogatory use of the word "gay" completely wrong, but then, I'm not a native English speaker, so I might be wrong).

  7. Joe said,

    February 9, 2009 @ 1:58 pm

    Hom's NDNA function of epithets reminds me of the "Word Association" SNL sketch starring Chevy Chase and Richard Pryor. Perhaps I'm missing something, but Hom's externalist explanation does not adequately address this complex function of epithets: NDNA usages are interpreted as non-NDNA (and vice-versa). For example, could you imagine someone asking the (presumably) NDNA-type questions (21-27) at a job interview? Perhaps you could if you were trying to get Richard Pryor to kick your ass ("DEAD Honky!")

  8. mollymooly said,

    February 9, 2009 @ 3:00 pm

    I think "non-derogatory, non-appropriated" NDNA seems to be the use-mention disctinction.

    On which basis #13 is missing quotes:
    – Yao Ming is Chinese, but he’s not a "chink".
    Otherwise it's like John Turturro's racist character in "Do the Right Thing":
    – Magic Johnson, Eddie Murphy, Prince are not niggers.

  9. Peter Howard said,

    February 9, 2009 @ 3:18 pm

    "[T]he derogatory force associated with ‘nigger’ does not vary across competent speakers of English."

    Er, yes it does. Quite a lot. And perhaps even more with 'paki' as was amply demonstrated in the comments on "Pakigate, Sootygate, Gollygate"

    It might be worth considering that it might not be possible to have a single successful theoretical strategy for racial epithets, if 'competent speakers of English' don't all have the same personal strategy. I mean, if some people think that you have to take into account who is making the utterance and when, and others that all uses of the word are racist. Period.

    My credentials for submitting this comment: the penultimate word of the paper is 'paradigmatic'.

  10. J. W. Brewer said,

    February 9, 2009 @ 3:35 pm

    I'm confused by example sentence #13. I take it's supposed to be "pedagogical" in the sense of a well-meaning, albeit patronizing, explanation that the epithet is not correctly applied to Yao Ming because it's not correctly applied to anyone. But it works equally well as a statement by a somewhat nuanced bigot who intends to convey that YM is indeed a member of the ethnic group in question but is not properly called by the epithet because he does not (in the speaker's judgment) personally exemplify the negative stereotypes about the group the epithet is intended to evoke. That's a pretty important type of usage (since many negative group stereotypes are, in fact, resilient enough to acknowledge and work around apparent counterexamples), and one which the paper doesn't seem to address at least at first quick read.

    Would a Fregean think the sentence "All horses are nags" is true by definition? Really? If so, this paper certainly doesn't make me second-guess my long-ago decision to major in linguistics instead of philosophy.

    The brief description on page 21 attributing bad human behavior primarily or exclusively to "institutions" and "ideology" sounds, without further explication, a little vacuous and NPR-like. Maybe it's developed more fully elsewhere, but I wonder whether the linguistic analysis, which I thought was at least interesting, is separable from a commitment to this particular non-linguistic schema for explaining the world.

  11. Yusef said,

    February 9, 2009 @ 3:51 pm

    All in all, I'm not impressed by this paper. I didn't have high hopes for it, but I thought I might as well give it a read to get an idea about what academic philosophy and linguistics is like. He labors the question, which is quite easily answered, seemingly in pursuit of his own socially authoritarian aim to have legislated what words people can use in conversation.

    The writer is putting across an incorrect meaning of racial slurs. The sentence "A said that Yao is Chinese" is not misreporting "Yao is a chink", because A can only have said that sentence to convey that A was Chinese, and the use of the word "chink" disparages Chinese people in general, not just Yao. If his audience knew that Yao was Chinese they'd wonder why he'd stated an obvious statement. The "derogatory content" /is/ detachable. Racial slurs do /not/ have null extensions – the semantic meaning is the same as a neutral racial term. Direct use of ethnic slurs are also often deliberately referring to the fact of the existence of the taboo.

    The writer is wrong to talk about the "truth value" of sentences like "Oprah believes MLK is a nigger", "Yao is a chink", "Institutions that treat Chinese as chinks are morally depraved." "Chink" means Chinese, not Chinese people with stereotypical characteristics, and the last sentence is a stupid thing to say, as are sentences 13-27, excluding 22 "Is Yao Ming a Chink?"

    All his talk about external semantics is just expressing the fact that to be understood, you must use words in a way that other people will understand. You don't need to talk specifically about "institutions" and "deontic prescriptions". He says incorrectly that epithets towards the English are much less insulting than epithets towards African Americans. The social consequences may be different, but these words could be used by the speaker with just as much contempt and cause just as much irritation in the listener.
    There is more anti-white racism and black ethnic identity than the other way round. Rape of white women by black men is thousands of times more common than the other way round. In today's society, it is white Americans who are the biggest victims of racial hatred, so maybe we should claim that the First Amendment doesn't cover the use of the word "limey," as use of the word encourages deontic prescriptions that are related to anti-white crime.

  12. S Hawkins said,

    February 9, 2009 @ 4:50 pm

    I'm having trouble accepting that the power of "nigger" lies in the word itself. He contends that "nigger" is particularly "explosively derogatory" and thus stands in contrast to "chink." Accepting this comparison for the moment (though I'm not actually completely convinced) I would suspect that this is as much an indicator of the relative positions of Chinese and African-Americans in contemporary US society as anything else. The explosive nature of the epithet is linked to the explosive nature of the relationship.

  13. Nick Lamb said,

    February 9, 2009 @ 5:19 pm

    This must surely be a hoax?

    I'm not even sure the author understood the joke in the "Marriage is gay" bumper sticker.

  14. acilius said,

    February 9, 2009 @ 5:40 pm

    Maybe I'm missing something, but all the discussion about context independence left me with a question about something that happened to a friend of mine when we were in school.

    This friend's name was not "Joshua Silverberg," but it was something else equally identifiable as Ashkenazi Jewish. So let's use the name "Joshua Silverberg" in the story. A local neo-Nazi took it upon himself to scrawl "Joshua Silverberg is Jewish" hundreds of times on a restroom wall in a pizzeria near the school.

    Joshua's friends interpreted this to mean "Because Joshua Silverberg is Jewish, neo-Nazis ought to target him." So we stuck close to him for some time afterward.

    My question is this. A name like "Joshua Silverberg" so strongly suggests that its bearer is Jewish that simply writing the name "Joshua Silverberg" hundreds of times on the restroom wall would have implied the propositional content of "Joshua Silverberg is Jewish." What did "is Jewish" add to the graffiti if not the force of a racial epithet? Would "Joshua Silverberg is a yid" have carried that force any more effectively?

  15. Christopher Hom said,

    February 9, 2009 @ 6:05 pm

    Mark, thanks for your comment. Your worry about metalinguistic negation is something that I’ve definitely been thinking about, and it’s something that Williamson raised to me in personal conversation a while back. I think I have three ways of responding to this worry:

    First, the relevant cases of metalinguistic negation raised by Horn get triggered by the literal contradiction of what is said (e.g. “I’m not hungry, I’m starving”). For there to be a contradiction to trigger metalinguistic negation in sentences like “I’m Chinese, not a chink”, there must be the assumption that “Chinese” and “chink” are synonymous. But that’s just to beg the question against my view, which holds that they are not synonymous.

    Second, even if metalinguistic negation held in this case, what about the apparent truth-evaluable occurrences of slurs in other contexts such as in the antecedent of a material conditional, or in a question, or in an indirect context. To claim that all of those occurrences are also metalinguistic would seem to be ad hoc. The MacFarlane quote on (436) is appropriate to remark upon as well.

    Third, while I appreciate the value of linguistic intuitions, I believe that they are of limited value on this issue (see my pages 434-435), and we're left with the task of a careful evaluation of the theories.

  16. J. W. Brewer said,

    February 9, 2009 @ 6:39 pm

    Maybe part of the problem is that he's starting with "pure" epithets like "chink" and the n-word, which have always been epithets (or at best, in the woodpile/Narcissus examples invoked in a previous comment, rather vulgar slang) and then trying to generalize outward from them. Even in the bad old days when overt racism was socially acceptable, white supremacists had other words for the groups in question which they could and would employ when they wished to be polite or formal. Lots of other sometimes-offensive terms don't have that history, so you either need to look more to pragmatics and context or perhaps resort too quickly to ascribing contextual differences in usage to different "semantics" in different dialects/idiolects.

    For example, "Eskimo" has apparently evolved into a taboo epithet in at least some registers of Canadian English but generally not in American English. Is that best understood as a "semantic" difference between dialects (geographically, and in Canada historically) or something else? It seems to me quite different from e.g. the way in which "fanny" has different semantics (and different levels of offense/vulgarity) in American English v. British English, or even the way (to go back to the prior "Paki" thread) in which the c-word is deployed differently as an insult in American v. British usage. I'm not sure that Prof. Hom's approach accounts for this.

    There's also the phenomenon of words which have become somewhat archaic and now seem at best a little patronizing and perhaps at least mildly insulting, like "Negro" or "Chinaman." Are they "epithets"? It seems inaccurate to view these as primarily or conventionally conveying something as strong as hatred or contempt, but some sort of theoretical account is needed (since there is a meaningful risk that offense will be given, and they therefore ought to be avoided or used with care by those interested in avoiding giving offense).

    Finally, there are words where there's simply a lack of speech community consensus as to whether they are epithets to start with. I'm not sure how Prof. Hom proposes to think about the semantics of such words. One example I have encountered in American English is "chick" used to mean "human female." I have encountered some native speakers who seem to consider it intrinsically or primarily belittling or demeaning and thus at least a mild epithet, but others (probably a majority) who treat it as merely informal, roughly analogous to "guy" or "dude" for a male. For these speakers, using such an informal term might come across as disrespectful and thus offensive in a particular context where formality would be expected, but now we're back to context as determinative. (I don't think I've observed a clear geographical or social-class distribution that would enable one to line up chick-as-informal versus chick-as-epithet as tracking otherwise well-established boundaries between dialects/registers.) A parallel issue is whether there are informal-but-not-offensive slang terms that can be safely used by non-members of a particular racial/ethnic group to refer to members of that group. I take it that "spade" as used in "Some spade said 'rock and rollers, you're all the same'" (lyrics from "All the Way from Memphis," as recorded by Mott the Hoople in 1973) was intended to be such a hip-but-inoffensive usage rather than an "epithet," and I think I've seen other white-hipster uses of "spade" of similar vintage that seemed to evince similar intent, but I would tend to shy away from it myself because I don't have much confidence in my intuitions as to how it would or would not currently be taken by any given audience. I'm not sure that viewing my reticence about the word as driven by "semantics" rather than "pragmatics" is all that useful.

  17. David Eddyshaw said,

    February 9, 2009 @ 7:17 pm

    In the context of Pakigate, and how offended anybody is entitled to feel about it, it seems to me that the relevant part of this paper is "combinatorial externalism", which implies that the offensiveness of a slur is due to its connexion with racist institutions in the relevant society, and therefore is not mitigated by the speaker's own lack of intention to offend ("derogatory autonomy").

    The author explicitly allows that the circumstances may change with time, so that the derogatory force of an epithet may diminish or vanish; or the epithet may be as it were "quoted", (this seems to account for nearly all of his NDNA category); or appropriated by the target group and potentially denatured for at least some speakers.

    However despite this nuancing, he still seems to regard the relevant society and speech community as monolithic and the speech of all groups as standing in the same relation to the racist institutions which are the source of the word's offensiveness.

    There's actually a counterexample in his footnote 43 about "damn" among conservative Christians. You could well argue that damnation is not going to be an offensive concept to anyone who believes it's a fairytale, whereas racist practices are sadly real and unequivocal; but surely it's not correct that everybody in even the English speaking community is equally well informed or concerned about such problems, and I can easily conceive of whole social groups (such as British Army officers) whose understanding or lack of it is very different from say a typical (if there is such a thing) American liberal's. That may well be regrettable, but is nonetheless true; and in this context it seems quirky to insist on "derogatory autonomy".

    The magnitude of the rude gesture as it were spiritually associated with the rude word is surely different for different subgroups (whether one approves of this stateof affairs or not)

  18. Nathan Myers said,

    February 9, 2009 @ 7:28 pm

    I'm with Nick Lamb in his disbelief. I don't doubt that Prof. Hom meant the paper as seriously as he is capable of meaning, but I wonder if Prof. Pullum is pulling our collective leg in posting it. Is this sort of argument really taken seriously in linguistics circles, or is the posting meant to needle departments of philosophy for their inept and woolly forays into the modern science of linguistics? I mean the question seriously.

    That said, I'm not sure, myself, what the "Marriage is gay" bumper sticker means. It strikes me as a sort of Zen kōan, where engaging the effort to extract meaning achieves the purpose of the expression, and any meaning finally imputed is irrelevant or counterproductive. From context, Prof. Hom does not seem to take it that way.

  19. Christopher Hom said,

    February 9, 2009 @ 8:48 pm

    Don't hate the playa, hate the game.

  20. Mike Scanlon said,

    February 9, 2009 @ 10:10 pm

    If the insult is inherent in the word, how comes it that "niggardly," which has nothing whatever to do with negro, comes to be suspect simply because of a coincidence of sound. Does this not speak, at least in this case, for context and association being the source of the insult. Likewise, members of a group sometimes use toward on another, and affectionately, terms which are insulting from persons outside the group.

  21. Mihai Pomarlan said,

    February 10, 2009 @ 4:25 am

    Alright, having read the paper until the end I will add some other observations.

    Let's assume a special derogatory force for the word "nigger"; but then, combinatorial externalism has problems in explaining where it comes from. On the one hand, you can argue that the racist institution against African-Americans is a shadow of its former self, but then why isn't "nigger" going the way of "limey"? On the other hand, if you argue that African-Americans face significant discrimination and threats of violence today, how come "nigger" is making headlines in much the same way that "spic" and "wetback" don't? To a casual observer of the USA, it doesn't look like there's any less discrimination against Hispanics.

    Another objection is related to:

    "3. Derogatory autonomy: the derogatory force for any epithet is independent of the attitudes of any of its particular speakers".

    To Hom, this is a given. To me, this is pretty much the bone of contention and any reasoning starting from this point is circular. As a counter-example to this derogatory autonomy, consider the appropriated versions of racial slurs. When an African-American is saying to another

    "Yo, nigger"

    is he saying (to use Hom's method of interpreting racial slurs):

    "Yo, african and therefore stupid, lazy, subhuman and worthy of linching"


    "Yo, [fellow] member of the African community, a community that was/is being looked down on and threatened with violence"


    There's a world of difference between the two. If you say that "nigger" keeps its derogatory force in the second interpretation, then derogatory force is meaningless. It is one thing to actually insult, and another to refer to a history of abuse. It's like saying some text is offensive because the word "offensive" is in there somewhere.

  22. L. Morrissey said,

    February 10, 2009 @ 6:11 am

    "Contextualism is a premature surrender in the search for a principled analysis of epithets, and should be left as a last resort." I will have to disagree with this statement. In my experience, epithets are defined to a large extent by pragmatism, and, therefore, cannot be examined by purely semantic strategies.

    For example, here in Australia, the equivalent of the American "nigger" is "coon", a derogatory term for Indigenous Australians. If spoken in more rural areas, it means, "person who is Aboriginal, and despicable because of it", whereas if used in urbanised areas, it is more likely used as a mockery of the kind of people who use the word "coon", as if to say, "the foolish farmer with a base appreciation of cultural differences". As such, "coon" is, at once, a derogatory term for both Aboriginal and Caucasian people, depending on the context of the conversation, and the intent of the speaker.

  23. elinar said,

    February 10, 2009 @ 6:35 am

    A truth-conditional account of racial and other epithets may indeed be simple and economical; unfortunately, real life and real language use are exceedingly messy, so if you want to provide a meaningful account of how real people communicate in the real social and political world, you need to go beyond such simplistic theoretical notions as ‘literal meaning’, ‘conventional implicatures’ and ‘truth values’.

    Whether a particular term is derogatory in a specific linguistic and political community is something that is negotiated, debated and endlessly contested by members of that community. There are no clear rules or determinate criteria for judging whether certain uses are derogatory. There couldn’t be such rules because this is ultimately a political and moral issue, not a linguistic one. There will never be total agreement on these issues or simple answers to such questions as “Is the use of ‘Paki’ a racist slur?; “Does this depend on the context?”; “How relevant is the speaker’s attitude?” , etc.

    Hom’s paper may be interpreted as a contribution to this ongoing moral and political debate; in other words, it makes such normative statements as “the derogatory force associated with ‘nigger’ does not vary across competent speakers of English”. What he means, of course, is that, in his view, it SHOULD not vary; that you shouldn’t be regarded as a competent speaker if you don’t understand the true derogatory force of this racial epithet. It is not a factually accurate statement since – as other commenters have pointed out – the derogatory force associated with various epithets DOES vary across (perfectly competent) speakers.

    There is nothing wrong with making normative statements as long as they are not confused with objective, factual statements. For instance, when Hom claims that “the derogatory force for any epithet is independent of the attitudes of any of its particular speakers’, what he is actually saying is that this is the way we OUGHT to think about epithets; it is certainly not an “uncontroversial feature” of how epithets function in ordinary language’ – it is a normative statement. And – as the Gollygate, Pakigate (etc) debates have revealed – quite a few people disagree with Hom’s view.

  24. Stephen Jones said,

    February 10, 2009 @ 7:52 am

    Hom's basic point, that the derogatory nature of the word is the result of its linkage to an attitude or institution seems promising, but I fail to see how that puts the explanation firmly on the semantic as opposed to the pragmatic side.

    It also doesn't seem to address the nigger in the woodpile, how the same word can be derogatory is some contexts but not in others ('paki' as used in 'Paki Shop' for example).

    And of course, being American, he totally fails to deal with 'gay', that can be neutral or highly derogatory.

  25. Stephen Jones said,

    February 10, 2009 @ 7:59 am

    And of course Hom doesn't seem to deal with the cases of words that are seen as derogatory as some but not by others.

    The case of the man in the USA with a Fijian wife who, having lost her at the shopping mall asked a security guard if he's seen her. Describing her he started with the most salient feature which was that she was black. The security guard recoiled in horror and said "We don't say that word in the USA, we say "Afro-American." The man didn't bother to say that as his wife was neither African nor American it wouldn't be very accurate.

  26. Ian Preston said,

    February 10, 2009 @ 8:38 am

    I have a couple of problems with the relationship posited by "combinatorial externalism" to hold between insulting epithets and social institutions. For one thing, I fail to see how use of pejorative epithets commits one to social practices which discriminate against the group described. It might be uncommon but seems to me in no way incoherent for someone to despise a particular racial group (and to use offensive epithets to express that attitude) without endorsing any discriminatory social institutions that harm their interests. How else could you understand a (rarely heard) sentence like: "Chinks have the same rights as anyone else"? It's not meaningless but seems to be under the account in the paper.
    The suggestion that insulting epithets gain their force from how "active, pernicious and wide-ranging" are the social institutions that support them seems to me also to fit badly with the requirement that "the account of derogatory force for epithets needs to generalize to similar, related language." There are strong epithets that can be used against individuals who would typically be thought of by those using the words as agents or collaborators of oppressive social institutions. Take, for example "scab" ("strikebreaker and despicable because of it") or "pig" ("police officer and despicable because of it"). I am sure there are other, possibly less dated, examples of similar force. Except when used ironically or when reporting the views of others, I would say that the insulting content in their meaning is forceful and ineliminable, in a way similar to the insulting content in racist or homophobic epithets.

  27. Bloix said,

    February 10, 2009 @ 3:36 pm

    Even putting aside the annoying jargon (how does a strategy differ from an account? Are they both the same as an explanation?) and the tendency to believe that whatever is in his own head qualifies as “data,” I find Horn unenlightening. I would have thought that it’s obvious that an epithet evokes a collection of stereotyped characteristics that are then applied to the target of the epithet. The user of the epithet expects that the hearer knows the stereotype. Is that all there is to this paper?

    To the extent that the paper contends that epithets are always used by the powerful for the purpose of the oppression of the powerless, it is clearly incorrect. Members of any group, powerful or powerless, can use epithets to refer to members of other groups or to members of sub-groups – including of course, members of a dominant group.

    And a counter-example which disproves the assertion that the derogatory meaning is built in to an epithet: there are many non-Jews, particularly in the American South and midwest, who never refer to “Jews” and always speak of “Jewish people.” This is because, in their usage, “Jew” is an epithet. I once had an interesting conversation with a co-worker who was surprised to learn that Jews refer to themselves as “Jews.” I would think that this one example is sufficient to disprove the claim that the derogatory content of an epithet is not dependent on context or convention.

  28. Dave K said,

    February 10, 2009 @ 11:43 pm

    Could somebody explain to me what "nigger in the woodpile" actually means? The only times I've ever seen it is in lists of expressions, or in metalinguistic discussions like the present one; I don't think I've ever heard it used in context, and am genuinely not sure exactly what it means or how one would use it. Is it a common expression in the UK, or was it until recently? Was it ever common in the US?

  29. Stephen Jones said,

    February 11, 2009 @ 2:45 am

    The SOED gives 'nigger in the woodpile' to be a concealed motivation or unknown factor that adversely affects the situation, and marks it as US English. To move from speaking agriculturally to speaking mechanically as Oscar Wilde would say the nigger in the woodpile puts a spanner in the works.

    Not too common a locution in my experience. No reference in either the BNC or COCA, though one can't rule out self-censorship as they both recent corpora. It appears three times in the Time Corpus, with the last date being 1943. The OED corpus has a reference from the NYT in 1990, though the reason it made the pages of the NYT appear to be that it was considered a tremedous gaffe. The Daily Telegraph has an example going back to 1960.

    The phrase that is common, is 'to work like a nigger', which has no other meaning that working very hard. The problem with set phrases is we often don't associate them with their constituent parts so any opprobium that is associated with the constituent epithet doesn't apply to the set phrase.

  30. Christopher Hom said,

    February 11, 2009 @ 3:59 am

    I wonder whether the Language Log actually enforces its own explicit policy on comments being 'brief', 'relevant', 'informed', and 'polite'. It should, if it wants to be taken seriously among professional academics. These are just a sample of the previous comments:

    'of course, being American, [Hom] totally fails to deal with 'gay', that can be neutral or highly derogatory'

    'Is this sort of argument really taken seriously in linguistics circles, or is the posting meant to needle departments of philosophy for their inept and woolly forays into the modern science of linguistics?'

    'in their usage, “Jew” is an epithet. … this one example is sufficient to disprove the claim that the derogatory content of an epithet is not dependent on context or convention.'

    'There is more anti-white racism and black ethnic identity than the other way round. Rape of white women by black men is thousands of times more common than the other way round. In today's society, it is white Americans who are the biggest victims of racial hatred…'

    [(myl) Comments are unmoderated, for the most part. The first comment from a given email address must be approved, but after than there is no "prior restraint". If we notice comments that are too long, too impolite, etc. we sometimes delete them; but we usually try to persuade people to act better rather than deleting the evidence that they've failed to do so.

    Most often, when comments come in large numbers, we don't even see them, at least for many hours. Speaking for myself, I haven't even read most of the many comments on this thread until now. I noticed yours on the top of the display of recent comments, which is why I'm responding here.

    Overall, our comments policy has roughly the force of traffic laws, which are often violated — in some cases, perhaps more often than they're obeyed — but still arguably have some mitigating effect on the empirical norms of behavior.

    Given the modest amount of time available to us for dealing with this blog, there's no real alternative. Some LL posters respond by keeping comments turned off; others turn them on and hope for the best. In general, I feel that there are enough interesting and informative comments to make up for the losers. ]

  31. Stephen Jones said,

    February 11, 2009 @ 6:32 am

    of course, being American, [Hom] totally fails to deal with 'gay', that can be neutral or highly derogatory

    Perhaps you could deign to explain what you find wrong with the statement. In your piece you gave 'gay' as an example of a word without derogatory overtones. This might be true in American English but it has become reborn as a general term of disapproval amongst many young Britons. I don't expect you to know this (I only know because of articles about it in the generalist press) and that was what I was referring to. It's quite interesting since there it is not clear that those using it as a general term of naffdom use it derogatorily when referring to homosexuals.

    'in their usage, “Jew” is an epithet. … this one example is sufficient to disprove the claim that the derogatory content of an epithet is not dependent on context or convention.'

    And what is wrong with that statement? It's a clear example of what I mentioned earlier with the word 'black'. That your paper doesn't appear to deal with words that are considered derogatory epithets by some members of a speech community and quite neutral by others. You can argue the point but to complain about it!?

  32. Chris said,

    February 11, 2009 @ 9:16 am

    In your piece you gave 'gay' as an example of a word without derogatory overtones. This might be true in American English but it has become reborn as a general term of disapproval amongst many young Britons.

    Speaking as an American, I don't think it's true in American English either. "Gay" as a term of disapproval is certainly frowned on in some circles, but in the informal, norm-defying discourse of youth it is used pretty often, I think (I don't have any quantitative data, but that's just my subjective impression).

    @Christopher Hom: I don't think any blog comment section will ever be taken seriously among professional academics. The principal posts, maybe, but even then, they are deliberately less formal than formal academic discourse (see, e.g., the posts on IE wheels and horses: interesting, informative, but not quite the form you would have put that information in to have it published in a professional journal).

  33. Bloix said,

    February 11, 2009 @ 10:02 am

    Christopher Hom –
    If a person can bring forward a black swan, the statement "all swans are white" is proven to be false. There need only be one black swan to demonstrate that the statement must be modified. Of course, if there is one black swan, there are likely to be more, but only one is necessary.

    As I understand your article, you contend that the derogatory content of epithets is "fundamentally part of their literal meaning." In Prof. Pullum's paraphrase, your contend that the "denigratory character of racial epithets [is] built into their actual conventional meanings, and not just as a possible concomitant of some of their occasional uses."

    I offered "Jew" as a counter-example. Among Jews and many others, the word "Jew" is not an epithet. "I am a Jew." "Many observant Jews live in this neighborhood." "Jews tend to vote Democratic."

    However, in some parts of the United States, the word "Jew" is an epithet, conveying a collection of stereotyped characteristics including clannishness, miserliness, dishonesty in business, physical weakness, cowardice, lack of patriotism. and other undesirable traits. In these parts of the country, there is a history of using "Jew" in what today we would consider a derogatory sense. "I bought it at the jew store." "Don't you try to jew me down." "I offered him a good price, but he's such a jew, he wants more." "Of course he didn't enlist, he's a Jew, what did you expect?"

    Today, many ordinary people avoid the word "Jew" in an effort not to use what they perceive to be an epithet. To be polite, they tend to substitute "Jewish person" or "Jewish people" in its place. "My accountant is a very nice Jewish man." "Most of the Jewish people in our town live near the temple." "The family that owns that store are Jewish people."

    To my mind, the perception by many that "Jew" is an epithet, while to many others it is not, is a black swan. It demonstrates the inaccuracy of your contention that the derogatory content of an epithet is fundamentally part of its meaning.

    Because one needs only one black swan to disprove a thesis, I consider your thesis disproven. You are certainly welcome to ignore me or to use this thread to correct any error I've made.

  34. Marc A. Pelletier said,

    February 11, 2009 @ 10:23 am

    I've read the paper. I'm usually one to edge critique, but eew! What unmitigated pap!

    Reciepe: (a) have a priori opinion (b) beg the question by cherry picking (or misinterpreting) supporting examples (c) handwave away obvious counterexamples (d) expunge all relevant context (e) reiterate a priori opinion in the form of a conclusion.

    As far as I can tell, this sort of "methodology" is usually reserved for theology.

    — MAP

  35. Christopher Hom said,

    February 11, 2009 @ 11:50 am

    There is a lot of misunderstanding in the comments. Here are four things to keep in mind:

    1. The view has the flexibility to allow for different language and idiolects to be causally responsive to different institutions.

    2. The view is about racial slurs, and not about racist expressions generally. "Jew" is no more a racial slur than "Chinese" is, although they can be used in sentences that have racist conversational implicatures, like "He's good at laundering; he's Chinese".

    3. Semantic externalism allows for competent speakers to be partially unaware of the meanings of their terms. This may seem unintuitive to some, but comports with foundational work by Kripke, Putnam, and Kaplan.

    4. This language is highly charged, both politically and emotionally. This taints the value of our initial (and often conflicting) intuitions. So I try to offer a careful consideration of the theoretic space, which might allow us to reconsider some of our intuitions (and vice versa) with the hope of achieving some kind of Rawlsian reflective equilibrium.

  36. Mihai Pomarlan said,

    February 11, 2009 @ 12:03 pm

    @Christopher Hom:

    "Jew" is not the original, Jewish name that the Jewish people gave themselves. It is the Anglicized version.

    Then, consider "nigger", or "negro". This, as far as I know, was originally the term used to describe the African slaves. By the reasoning in your post, "nigger"/"negro" would be equivalent to "Jew" or "Chinese", in that it represents the name given, or only names given, by some English speakers, to a population.

    So Bloix's point still stands. "Jew" is in fact a racial slur, is used by some people as a racial slur and avoided by others for the same reason, while used perfectly neutrally by Jews themselves.

    Semantic externalism may well be self-consistent, but not necessarily consistent with reality. As engineers know, one should never confuse a mathematical model with reality.

    As for the flexibility of combinatorial externalism, it is inflexible for not allowing cases like "Jew", it is predicated on its own conclusion (that epithets are always derogatory), and it still fails to account for the putative relative strength of "nigger" as opposed to "spic" or "wetback".

  37. Mihai Pomarlan said,

    February 11, 2009 @ 12:16 pm

    In fact, let's add "jew" to the comparison of "derogatory force". After all, there are racial stereotypes about Jews, many of them negative, and many threats of violence against people of Jewish descent. Just look around and examples of antisemitism abound. And the word that those people use is … "jews"! By the logic in the paper, "jew" should surely be a racial slur, because it is (sometimes) used to imply a set of nasty traits as well as threats.

    And yet, Hom is insisting the term is in all situations neutral. This is not consistent with his own model!

    It is however perfectly reasonable to hold that the meaning of "jew" is, in fact, dependent on the context of discourse.

  38. George said,

    February 11, 2009 @ 3:26 pm

    @ Dave K
    "Could somebody explain to me what 'nigger in the woodpile' actually means? The only times I've ever seen it is in lists of expressions, or in metalinguistic discussions like the present one; I don't think I've ever heard it used in context, and am genuinely not sure exactly what it means or how one would use it. Is it a common expression in the UK, or was it until recently? Was it ever common in the US?"

    Growing up in the southeastern US, I was surprised to learn about the "hidden motive" meaning for this phrase. I have always heard it used to imply (often facetiously) a mixed black ancestry.

    Thus "There was definitely a nigger in the woodpile" might be said of someone of obvious mixed descent. But "I think there must have been a nigger in the woodpile" could be used jokingly to suggest that a white person exhibiting some stereotypically black characteristic must have had some black genetic influence somewhere in the past (for example, a white girl with a 'ghetto booty').

    Folk etymologically speaking, for me the phrase always evoked the image of a clandestine tryst in the woodpile, the mixed couple needing to conceal their affair due to societal views on miscegenation.

  39. Mark Liberman said,

    February 11, 2009 @ 3:44 pm

    According to the OED,

    The phrase nigger in the woodpile … is said to derive from an incident in the U.S. in the time before the American Civil War when a group of escaped slaves who had been conveyed along the Underground Railroad to Pultneyville, New York State, with a view to crossing Lake Ontario into Canada were enabled to make the final stage from a warehouse in which they were hidden to a boat by means of woodpiles set up across the wharf through which a concealed passage had been constructed (see further N.Y. Folklore Q. (1958) 14 16-25).

    The citations include:

    1852 in Kansas Hist. Q. (1942) 11 235 No ‘nigger in the wood pile’ here..; white men are at the bottom of this speculation. 1876 Congress. Rec. 4 Aug. 5153/1 If some one should say..that there was some ‘nigger in the wood-pile’, some ‘cat in the bag’, some motive to actuate me. 1897 Congress. Rec. 18 Feb. App. 61/1 Like a great many others ignorant of facts, he finds ‘a nigger in the wood pile’ where there is neither wood pile nor nigger. 1911 W. WILSON in Outlook 11 Aug. 944 If you go through the schedules you will find some nigger in every wood pile. 1930 Cambr. Daily News 24 Sept. 7/6 Unless..there is a nigger in the wood pile,..the shares ought to be worth a mild flutter at round 8s. 6d. 1952 A. CHRISTIE They do it with Mirrors xii. 109 Well now, let's have your point of view. Who's the nigger in the woodpile? The G.I. husband? 1960 Daily Tel. 16 Jan. 8 This seems to be the nigger in the woodpile–the woodpile being an industrial recovery and activity remarkable by any standard.

  40. J. W. Brewer said,

    February 11, 2009 @ 4:24 pm

    If I may make a "metacomment," I think a lot of the "misunderstanding" that frustrates Prof. Hom arises from the fact that his paper was done according to what I take to be the conventions of contemporary academic philosophy, which seems to accept scholarship lacking the focus on an empirical, descriptive, scientific, and/or evidence-based approach to linguistic phenomena that tends to be rather self-consciously touted by many of the contributors/readers/commenters at Language Log. Presumably a lot of the commenters in this thread would have marked up the MS with marginal questions like "source?" or "support"? or "how do you know this?" if asked to vet it for publication, but we are apparently not the sort of people who work as editors or referees for the philosophy journal that published the piece. We shouldn't really fault Prof. Hom for doing the sort of scholarship that meets the standards of his own chosen discipline, but we may perhaps wonder whether or not scholarship conducted in that vein is likely to produce work that is of value to those of us interested in human language as an empirical phenomenon. Maybe, maybe not.

  41. elinar said,

    February 12, 2009 @ 3:42 am

    @Christopher Hom,
    It could be that people haven’t misunderstood you, but that they simply reject the basic premise of your argument and question the usefulness of your approach.

    My main objection is that you appear to be making political/moral points in the guise of conducting linguistic/semantic analysis. For instance, you say that “Semantic externalism allows for competent speakers to be partially unaware of the meanings of their terms”. So speakers who don’t understand the ‘real’ meaning of various terms are not necessarily incompetent; they just lack awareness! The problem with this is that it is possible to be completely aware of all the various meanings attached to a term, and still deny that it has the derogatory character you think it ought to have. This is not about literal meanings, idiolects or intuitions; it’s about taking a particular moral or political stance. It is one thing to object to the moral/political position of those who disagree with your argument; it is quite another thing to suggest that these speakers fundamentally misunderstand what words like ‘nigger’ and ‘chink’ mean.

    &Chris and J W Brewer,
    In my view, linguists, philosophers and other academics interested in language ought to take seriously blog comments and other lay metadiscourse. Because anyone who wants to be fully aware of the current use/meanings of words is likely to learn a good deal more from following Internet and other lay debates than from reading academic papers.

  42. Stephen Jones said,

    February 12, 2009 @ 4:39 am


    Your last posting accusing commentators of 'a lot of misunderstanding' is laughable.

    If you don't want to be misunderstood express yourself clearly. I can't find any reference to the get-outs you are now implying.

    It is nowhere clear in your paper that you are proposing a partial explanation, that only applies to a limited number of racial epithets, which term you seem to get out of describing rather well. You are now stating there is a clear dividing line between 'racial slurs' and 'racist expressions' generally. That is precisely what many are denying.

    Still it's nice to be reminded that in certain fields of academia 'peer reviewed' is a synonym for 'sloppy rubbish'.

  43. Chris said,

    February 12, 2009 @ 9:16 am

    Still it's nice to be reminded that in certain fields of academia 'peer reviewed' is a synonym for 'sloppy rubbish'.

    I think that's going a bit far, even for something like Social Text. It's sufficient to state that the intersection of "peer-reviewed" and "sloppy rubbish" is nonempty.

    I'm not impressed by what I've seen of semantic externalism, although I admit that my study of it is superficial. It seems to be founded on the dubious premise that there *is* a well-defined meaning for an utterance, and then "proves" that the speaker is not necessarily aware of it. But who says that the "true meaning" even exists at all? (In natural languages, as opposed to formal ones where the meaning of the formal language's terms is defined by the formal definitions and rules of the language.) Any given utterance has a variety of possible interpretations (and when that variety is wide, the utterance can be described as "vague" or "ambiguous" or similar terms).

    P.S. Maybe the best resolution to the question of whether or not "Jew" is an epithet is to consider it polysemous, with both insulting and non-insulting meanings? But, on reflection, it seems that Christopher Hom may be aiming at a theory of meaning that requires that polysemy cannot exist. Unless I'm simply misunderstanding him, or he is saying something other than he intends to say.

  44. Kris Rhodes said,

    February 12, 2009 @ 3:20 pm

    It is true that Hom made difficulties for himself when he made empirical claims about what particular epithets mean for all competent speakers. As a Philosophy student with an interest in Philosophy of Language myself, I recognize the kind of claims he made in this vein as being pretty standard fare in PhiloLang, and I have of late come to recognize the problems with making these kinds of claims.

    I think much of Hom's point can be salvaged, though, if we understand him to be asking the question, "[i]When[/i] a word [i]is[/i] an epithet in some language (or ideolect), is its derogatory value part of its conventional meaning in that language, or rather does it come from other conventions concerning its use?"

    Perhaps Jew is an epithet in some communities, and not in others. (I'd never have called it an epithet myself, but apparently some would.) The question is, when it [i]is[/i] an epithet in some dialect, does its derogatory force come from its conventional meaning or from some other source?

    Also, some people in this thread seem to think that when a black person calls his black friend a "nigger," "nigger" is not an epithet in such a case. This doesn't seem immediately clear to me. The term might still have derogatory force, and in this case, the use of the term relies on an ironic distancing of the two people in the conversation from that derogatory force. If my friend makes a very clever move in a chess game, and I say "You idiot!" and point out something about the move which is actually in his favor, as though it were not in his favor, I'm making a joke by calling him an idiot, but the force of the joke requires that the term "idiot" [i]does[/i] have derogatory force (at least so it seems to me). It's because of the derogatory force that the joke can be understood. And it may be the same with a term like "nigger" used in a friendly way amongst black people. So, as I said, it's not clear that "nigger" isn't always an epithet just because some people use it in a friendly way with their peers.

    Still, the point should be well taken that sweeping claims like "Nigger is an epithet for all competent speakers of English" should be avoided. This is an empirical claim that requires documented support. Most philosophers of language miss this, because the usual procedure is simply for one philosopher of language to make a claim about "English," and then for another philosopher to say "Yes, that matches my intuition about English as well" and then the two take it as an established fact about English. This is unfortunate but true. It can be explained (but not jutified) by the fact that Philosophy of Langauge has tended traditionally to be concerned with little toy languages, and with finding an understanding about how meaning works for such formal models of language. I think some people have thought that it would be useful approach to tackle these toy models first, then apply the lessons learned to the more messy world of natural language later. But somewhere along the way, the distinction got lost, and now we have philosophers of language trying to use the techniques learned while dealing with toy languages directly on natural language. The techniques don't fit, but it can be hard to realize this if you're training has habituated you against this realization.

    I've rambled on long enough.


  45. Kris Rhodes said,

    February 12, 2009 @ 3:24 pm

    Also, [preaching] it is important that academics engaged in academic discussion treat each other and their views with respect, and try to learn from each other. Almost everyone is almost always onto something. The trick (and the joy and the productive value) of it is to figure out what the other guy's onto, even when he's wrong. But you can't do this if you are not treating him with respect, both in your own mind and in the words you utter. The insults get in the way of honest pursuit of truth. I suspect most of you agree with what I just said, but reminders can be valuable.[/preaching]

  46. Nathan Myers said,

    February 12, 2009 @ 3:40 pm

    A consensus seems to be developing that this paper is rubbish, from a scientific linguistic standpoint. However, Prof. Hom's cryptic remark, "Don't hate the playa, hate the game", raises what seems an interesting question: is it obligate rubbish? That is, are the conventions of this particular literature such that to get published one has no choice but to write rubbish? We might compare, for example, the psychology literature of much of the last century, in which research proposals and results had to be couched in behaviorist mumbo-jumbo in order to get past the various gatekeepers.

    The remaining question, if so, is whether there remains some kernel of sense after one filters out the obligate rubbish.

    As a side question, is there a more conventional term for what I have called obligate rubbish?

  47. Ian Preston said,

    February 12, 2009 @ 4:32 pm

    A consensus seems to be developing that this paper is rubbish, from a scientific linguistic standpoint. However, Prof. Hom's cryptic remark, "Don't hate the playa, hate the game", raises what seems an interesting question: is it obligate rubbish?

    If that's the consensus then count me out, perhaps because I am not sure the point of his exercise was ever one of scientific linguistics. I read Hom less as attempting an analysis of the empirical usage of any particular racial epithets and more as attempting to understand from the perspective of a philosopher how meaning attaches in general to racial epithets in "speech communities" where usage is accepted to have the broad features that he describes. He may well make some ill-judged generalisations about usage along the way but, within the limits of what I take him to be doing, his paper seems to me considerably more subtle than many comments above give him credit for. That doesn't stop me thinking he gets it wrong and saying so above, for what little my amateur views are worth, but I don't understand the level of dismissive intellectual assuredness that leads someone to think discussion is advanced by disparaging him and his entire discipline. If anyone ever chooses to extend his research to a scientific linguistic study of incivility in general then they could do worse than begin their research by looking at some of the condescension on this thread.

  48. Harrison said,

    February 12, 2009 @ 4:33 pm

    This comment board is a joke; it's nothing more than a bunch of amateur linguists bashing someone who works in a field they know nothing about because he's published a paper they don't understand. [That many of you are amateurs is evident from the fact that you fail to see how philosophy of language underlies the entire discipline of linguistics – a fact that an informed professional linguist would see quite easily]

    Nathan Myers has just wondered whether in order to get published in philosophy of language, "one has no choice but to write rubbish." Apparently, knowing nothing about a discipline is no obstacle to knocking it.

    Here's a general question directed at all of you: Having done no serious work in a particular discipline and, in general, knowing nothing about it, how are you in a position to critique the methodology employed by those who work in that field?

  49. Peter Howard said,

    February 12, 2009 @ 5:14 pm

    Explicate it then, Harrison. I wouldn't regard myself as even an amateur linguist, but I do think I can spot an example of begging the question when I see one, or an unsupported assertion. I've been wondering, throughout this discussion, whether Hom's essay is seriously flawed, or whether all the commenters, including me, are missing the point. If you can throw any light on this, I'd be really interested to hear what you have to say.

  50. Nathan Myers said,

    February 12, 2009 @ 5:40 pm

    Harrison: "Apparently, knowing nothing about a discipline is no obstacle to knocking it." I wonder, idly, whether you can tell the difference between a question and a statement. I have asked three questions:

    1. Is this sort of argument really taken seriously in linguistics circles?

    2. Are the conventions of this particular literature such that to get published one has no choice but to write rubbish?

    3. If so, [does] some kernel of sense [remain] after one filters out the obligate rubbish?

    None of these pass judgment, which I am not credentialed to perform in any case. They merely consider the consequences of accepting the verdicts passed by other posters here. Nobody has essayed to answer any of them, that I have seen. My assumption is that Prof. Hom is no more inclined to waste his time than any of us. Shovel away, there ought to be a pony in there somewhere. Perhaps here, freed of the conventions of his field's literature, Prof. Hom can express his kernel directly.

  51. Harrison said,

    February 12, 2009 @ 7:29 pm

    Peter: None of the fallacies you mentioned are present in Hom's paper. I suggest you take a look at a philosophy of language textbook. You'll learn something about the methodology involved.

    Nathan: So, apparently, the purpose of this comment board is for the blind to lead the blind. Knowing nothing about philosophy of language, why would you ever take seriously the uninformed ravings of most of the posters here? Don't you think Prof. Hom would be glad to engage you in discussion if the majority of the posts here didn't display so much hatred and ignorance? Further, why should Prof. Hom make his point while "freed of the conventions of his field's literature"? Why not simply take the time to learn something about philosophy of language? If you're not willing to do that, you have no business asking such questions in the first place.

  52. Nathan Myers said,

    February 12, 2009 @ 9:33 pm

    Does philosophy of language, as practiced, really underlie the entire discipline of linguistics? Do working linguisticists really refer to works of language philosophers, rely on their terminology, distinctions, conventions, argument modes? More to the point, do the competent ones? (I omit Chomskyites, for example.) Once, biologists and historians alike were obliged to treat theology as such an underpinning, but they have graduated themselves.

    Professor Hom, if he cares to have his ideas understood outside his insular field, is obliged, like any academic, to express himself in the perhaps less convenient, less concise idiom of the larger world. Historians do it, biologists do it, physicists do it, linguisticists do it (here) — often superbly. Why should philosophers of language get a free pass? I don't know if Prof. Hom does so care, and I don't think he has any obligation to care. Care or no, he, like perhaps most academics, might not actually be equipped. Still, Prof. Pullum must have posted the article for a reason. He asked us to read the paper, but not the foundational texts of Ph. of L, and we did.

    I have preferred to attribute the paper's woolliness to conventions of his field. You seem to be denying its woolliness outright, or sticking it to Prof. Hom (it's not clear which), but insisting that what appear to be flaws in the logical reasoning would wash away given exposure to those foundational texts. Perhaps so, but biologists and historians in my experience seem to welcome questions, and to be eager to explain. If your response really means "go away", I am happy to do so. However, if somebody is willing to explain, I'm happy to listen.

    The purpose of this comment board is manifestly not for the "blind to lead the blind". It's to communicate. If you have something substantial to say, by all means say it. Answering questions asked is a good place to start.

  53. Harrison said,

    February 12, 2009 @ 11:52 pm

    Nathan, your insincerity is completely transparent.

    Here’s what’s wrong with your latest post:

    1) You’ve given a terrible counterexample to the view that phil. Lang. underlies linguistics. The moribund notion that theology underlies biology and history has nothing to do with what I’ve claimed.

    2) No philosopher of language is asking for a “free pass.” Generosity goes both ways; if you want Prof. Hom to explain his position to you, you’ve got to be willing to do a little work informing yourself about phil. lang. There’s a reason serious academics don’t appear on the O’Reilly Factor – they don’t wish to waste their time refuting someone who has no interest in learning anything about their research.

    3) No one’s asking you to get a PhD in philosophy of language, let alone read the foundational texts of the discipline. I have one simple and reasonable demand: if you’re going to criticize the methodology used by philosophers of language, you had better well understand it.

    4) I never suggested that “flaws in logical reasoning would wash away given exposure to foundational states”. The problem is that you and most other posters on this blog have the irritating habit of constructing flaws out of thin air (not that I mean to suggest that Prof. Hom’s paper is flawless –it doesn’t purport to be). Of course simply reading a text won’t get rid of flaws in a journal article, but that’s beside the point – the flaws you’ve pointed out were never there in the first place.

  54. Harrison said,

    February 12, 2009 @ 11:53 pm

    Addendum: I’ve got serious work to do, so this will be my last post.

  55. Nathan Myers said,

    February 13, 2009 @ 12:24 am

    Harrison: The expression "what appear to be flaws" is not the same as "flaws".

  56. Dave K said,

    February 13, 2009 @ 12:55 am

    Stephen Jones wrote:

    Perhaps you could deign to explain what you find wrong with the statement. In your piece you gave 'gay' as an example of a word without derogatory overtones. This might be true in American English but it has become reborn as a general term of disapproval amongst many young Britons. I don't expect you to know this (I only know because of articles about it in the generalist press) and that was what I was referring to. It's quite interesting since there it is not clear that those using it as a general term of naffdom use it derogatorily when referring to homosexuals.

    Oh, it's definitely true that "gay" is used as a general term of disapproval among American youths, too; in fact, I'm fairly certain that this usage originated in the US, and only later migrated to the UK, given that the meaning of "gay" as "homosexual" pretty clearly originated in the US. When I was growing up outside Chicago in the 1970s and early 1980s, "That's (so) gay" was extremely common among my peers as a general expression of disapproval, and it has continued to the present day, as far as I can tell.

  57. Stephen Jones said,

    February 13, 2009 @ 1:20 am

    An excellent post by Kris Rhodes.

    The question of whether much can be extrapolated from the study of formal languages to the study of natural languages is the matter of some debate.

    Another problem, I feel, is that Hom has produced some valuable musings, but that the bureaucratic necessity of translating them to the inappropriate format of academic peer review has weakened their communicative power. Darwin didn't have to produce a dozen peer-reviewed articles a year, and we should be celebrating the fact.

  58. Stephen Jones said,

    February 13, 2009 @ 1:37 am

    Hom seems to ignore the secondary use of 'gay' as a term of general disapproval. It is peculiar in British English since it doesn't seem attached to general dislike or persecution of gays amongst those who use the word, or any of the socially repressive organizations combinatorial externalism would claim to be behind it.

    I suspect the people that use it as a derogatory term in the US are 'anti-faggot'. The British use is much more interesting and perplexing.

  59. Stephen Jones said,

    February 13, 2009 @ 1:53 am

    Looking through Hom's post again, in order to answer DaveK, has meant I have read the end, and it is clear that Hom has a definite (and in my opinion nasty political agenda).

    He wishes to have the words he and presumably a whole collection of commissars decide on stripped of their First Amendment Protection. That is to say they are to be considered 'fighting words', 'literal threats', independent of the existence of any threat, the desire to issue it, or even any reasonable assumption they could be considered to be so.

    An important practical implication is that, when the practices are sufficiently threatening, the use of an epithet may count as a literal threat, and hence no longer merit freedom of speech protection under the First Amendment. p.p. 33-34

    The lack of determinate criteria for judging whether certain uses of epithets are derogatory also has significant legal implications for First Amendment speech issues. A detailed consideration of how the meanings of epithets impacts these issues is beyond the scope of this paper. Footnote p.5

    For example, because the meanings of some epithets entail their potential uses as literal threats, CE provides new grounds for ruling that some uses of epithets ought to be excluded from First Amendment speech protection. p. 27

    The complicated details of the legal consequences of CE are beyond the scope of this paper, but I plan to address them in my book, Hating and Necessity (in progress). Footnote p. 27

  60. elinar said,

    February 13, 2009 @ 4:18 am

    I am not a ‘practising’ academic, but I’m (sort of) aware of what is going on in linguistics. It is obviously not a monolithic discipline; not all linguists subscribe to the same philosophy of language. For instance, there are ‘proper’ professional linguists who reject the truth-conditional approach to semantics, so it is not just amateurs who are highly critical of the kind of analysis/approach proposed by Hom.

    Of course, we could say that academics are just playing a game and us simple folk should keep out of it. Unfortunately, they sometimes enter the real world, getting involved in legal and other practical matters that concern all of us. As Stephen Jones has just pointed out, Prof Hom himself has a political agenda, claiming that his theory has new grounds for ruling that “some uses of epithets ought to be excluded from First Amendment speech protection”.

    As citizens, we have every right to be critical, and wary, of experts who believe that they know better than anyone else what the ‘normal’ use is, or what the ‘true’ connotations of an epithet are.

    I recently came across a disturbing example of someone abusing their academic position. (See ‘Expert Witness for petitioners in Halpern vs. Canada on the meaning of the word ‘marriage’ http://www.ub.edu/grc_logos/people/amercier/proof3.htm)
    It relates to a Canadian court case concerning the amendment of the common law definition of ‘marriage’. This professor of philosophy was called as an expert witness to oppose the extension of the definition to include lesbian and gay couples. Here are some of the arguments he put forward:
    1. Altering the very meaning of the term is in conflict with the normal use and development of language.
    2. It is part of the present meaning of the word 'marriage' in our common tongue that it applies only to male-female conjugal unions…. it is a necessary truth that same-sex couples cannot marry.
    3. English does not allow us to say 'I now declare you husband and husband’.

    It may be tough on gays and lesbians, but what can we do: English doesn’t allow gay marriages; we cannot tamper with its nature and normal development! (Of course, the same tactic is used by religious fundamentalists when they tell us that God doesn’t allow gay marriages; that it is against human nature.) This is supposed to be a neutral description of how language works, but what the professor is actually doing is expressing his own political biases while hiding behind necessary truths and other theoretical constructs.

    This is one reason why I’m so suspicious of Hom's theory and of any semantic analysis based on externalist, literalist or conventionalist thinking.

  61. Mihai Pomarlan said,

    February 13, 2009 @ 4:37 am

    @Kris Rhodes:

    About the "you idiot" at chess example. Maybe the strength of the joke comes from the fact that, though the meaning of intellectual disability is expected to be contained by "idiot", it actually isn't in this case?

  62. Kris Rhodes said,

    February 13, 2009 @ 7:32 am

    Stephen Jones:

    You say Hom wants "commisars" to exclude use of epithets regardless of whether the user intends to threaten by use of the epithet. But the four quotes you gave do not support this characterization. There are too many uses of terms like "may," "some," "potentially," and statements that the issue needs to be looked at in detail (i.e. no sweeping claims about all uses of all epithets will do) for a characterization such as yours to be accurate.

  63. Kris Rhodes said,

    February 13, 2009 @ 7:33 am

    @Stephen Jones

    Also, I appreciate your appreciation of my previous post. :)

  64. Mihai Pomarlan said,

    February 13, 2009 @ 8:22 am

    @Kris Rhodes:


    I said, it can also be held that context does trump conventional expectation when meaning is concerned. Sometimes, hilarity ensues (if we try to "define" humor as playing with, and deceiving, expectations). And then, it would be true that the derogatory meaning of epithets varies with their context of use, regardless of what meanings they are "usually" associated with.

  65. Stephen Jones said,

    February 13, 2009 @ 8:23 am

    Kris Rhodes

    Hom is I think being more prescient than you are. Whilst, no doubt, Hom is hoping to have a more elevated position than mere commissar, to decide that certain epithets are intrinsically threatening, as he proposes, is obviously going to require somebody to decide which ones they are.

    At present the legal position in the US is decidely in favor of pragmatics. 'Fuc the Draft' is protected speech because 'fuck' is targeted at the draft and not an individual and is thus not a 'fighting word' (actual US jurisprudence according to Wikipedia).

  66. Kris Rhodes said,

    February 13, 2009 @ 8:26 am


    It's strange if that professor testified in that case that marriage necessarily involves a man and a woman. On her homepage–the page you linked to–she has a paper in which she "[a]nalyzes and rejects a suggestion that marriage analytically involves a male and a female and hence of necessity excludes gays and lesbians." Maybe she has learned her lesson since the trial?

  67. Kris Rhodes said,

    February 13, 2009 @ 8:31 am

    @Stephen Jones

    "Hom is I think being more prescient than you are. Whilst, no doubt, Hom is hoping to have a more elevated position than mere commissar, to decide that certain epithets are intrinsically threatening, as he proposes, is obviously going to require somebody to decide which ones they are."

    Stephen, I'm sorry but I can't see where you're getting this from Hom's paper. His paper–including the quotes you gave in a previous post–includes language that explicitly avoids the kind of scenario you've envisioned.

  68. Kris Rhodes said,

    February 13, 2009 @ 8:34 am


    Ah, I see, it wasn't the professor whose page you linked to but another one referred to in one of the articles at that page–as you said.

    Carry on!

  69. Stephen Jones said,

    February 13, 2009 @ 9:24 am


    What Hom is saying is that if it can be objectively proved that certain epithets are inherently threatening, independent of the intention of the speaker in uttering them, then those utterances don't have First Amendment Protection.

    Hom uses the phrase 'literal threats'. So if I say in the office 'Shh, I've been working like a nigger all week" then I am threatening any person in the room who can claim to pass off as a 'nigger', even if there was no threat intended, and I wasn't even remotely thinking of black people when I uttered it.

  70. Kris Rhodes said,

    February 13, 2009 @ 10:08 am


    From pages 33 and 34 of the paper, part of which you quoted, I see that he makes a distinction between "straightforwardly racist contexts" and "non-racist contexts." It is only in the context of straightforward racism that he makes the claims about first amendment protections that you've been talking about. He then goes on immediately to refer to "non-racist contexts" and says that "the account offers the requisite flexibility" to deal with usages in these contexts. So then, Hom does indicate explicitly that he does not intend his views about first amendment rights to apply straightforwardly to "non-racist contexts." And I'd say the situation you described constitutes a "non-racist context."

  71. Stephen Jones said,

    February 13, 2009 @ 12:52 pm

    So then, Hom does indicate explicitly that he does not intend his views about first amendment rights to apply straightforwardly to "non-racist contexts." And I'd say the situation you described constitutes a "non-racist contextBut he's not making it remotely clear what are racist or non-racist contexts. It could be even that non-racist contexts are those when a non-racist epithet such as 'fag',bitch or 'philo-linguist' are used.

  72. TJ is a Nigger said,

    April 24, 2014 @ 6:41 pm

    I think the argument to retire the word "chink" is not without precedent. We hardly ever hear the word "niggardly" anymore. It predated the racist "N-word" by centuries, and dates back to Old Norse and its verb "nigla," meaning to haggle over little things. Language is ever evolving and contextual. The times have turned to where the contexts for using the word "chink" make it more problematic than useful.

RSS feed for comments on this post