Race Talk

by Kevin Currie-Knight


Race talk – using racial categories as descriptors in conversation – is inevitably clunky. For the past several years I’ve noticed this, especially as the teacher of an Introduction to Diversity class where we discuss racial diversity, among other things. Race is a hard thing to talk about sensibly, not just because it is a charged “issue,” but because it inevitably produces awkward phrasing, things like “blacks, whites, and all the races,” which I often hear (and which makes me cringe every time). I suspect this awkwardness has much to do with the awkwardness of the idea of race itself. I want to articulate as best I can what I think makes race talk so awkward, and have come up with a number of reasons: race talk is shifting; sloppy; self-contradictory; and a subterfuge.

Here is my favorite example illustrating how awkward race talk can be. It’s taken from Rebecca Carroll’s memoir, Surviving the White Gaze. Carroll describes what it was like to be a black girl adopted and raised by a white family in a white area. In this passage, Carroll describes going back to school after a summer camp where, for the first time, she developed a crush on a black boy and befriended a group of black peers, even adopting their speech patterns.

In between classes at school our first day back, I stopped him. “Hey, what’s up?” “Nothing,” Connor said curtly. “I have to get to class.” “Wait, seriously, what’s going on?” I pressed. “I didn’t realize you liked that guy so much. Do you even know him?” I responded, channeling Jazmine, “Man, you buggin’.” Connor looked at me incredulously. “Oh, what? You’re black now? Just because you dance with some black dude? Whatever.”

What does ‘black’ refer to here? It’s likely that Connor knows that his sister’s physical appearance and family history “make” her black, her speech patterns or what boy she likes regardless. He probably also knows that adopting a certain speech pattern or crushing on a black person isn’t sufficient to make a person black (and likely wouldn’t suggest as much if one of his white sisters did either of those). But, here, ‘black’ does seem to refer to a vague and clunky cultural essence evidenced by how one talks and who one likes. And here’s the point: Connor can say all of this with full assurance that his sister will know what he means, as clunky as that is.

What makes race an idea rife for such awkward encounters? Cultural theorist Stuart Hall explained part of it when he famously called race a “floating [or sometimes, ‘sliding’] signifier.” The problem with race is that at the idea’s creation, racial categories were meant to refer to a biological and material essence that all members of a race shared in common. When it became clearer that science would find no such essence, the problem was that the race language that was now culturally commonplace became a signifier without anything to signify.

To put more nuance to this, race was once thought of as the underlying essences that explained physical, behavioral, and cultural differences between racial groups. Once we realized the fictive nature of this alleged cause, race language referred only to to the effects, but the way we talked proceeded as if we were talking about the cause.

I want to go further than Hall, though, because I think what makes race awkward cannot be fully explained by it’s being a floating signifier. We use floating signifiers – words for which there is no knowable referent – fairly often without such awkward results. We talk about minds, moral obligations, and authority – all floating signifiers – and the only folks who find talk about these things awkward is philosophers who have come to believe there is no there, there. We talk about general intelligence, despite the fact that even the most resolute psychometrician will admit that we’re having a bear of a time figuring out exactly what g is. We talk about all of these floating signifiers without it almost ever becoming as clunky and confused as some of our most common race talk can be. [1]

Here are four reasons why (in addition to race talk being a floating signifier) I think race talk is inevitably clunky.

Race talk is shifting

There are quite a few things race talk – calling a person “black” or mentioning that you were the only white person at the party – can mean, and the first problem is that what we mean often shifts from statement to statement. Philosopher Charles Mills generated a list of seven things that can be used to determine a person’s race: bodily appearance; ancestry; self-awareness of ancestry; public awareness of ancestry; culture, experience; and self-identification. Because race is a floating signifier, race talk is often used even by the same person in different conversations to reference different items on the list.

If on the day before the aforementioned conversation we had asked Connor whether one of his sisters is black, he’d likely have said yes, and if we’d asked why, he’d likely have referred either to her appearance (dark skin or a certain texture of hair), ancestry, or public awareness of her ancestry. But on the very next day, Connor could jibe his sister about “finally being black,” because she acquired a particular speech pattern and romantic interest. And he could make this shift in meaning, as the same person, on two different days, talking about the same thing: his sister’s race. [2]

Race is sloppy

If races had essences, categorizing someone by race would be a matter of checking what essence inheres in them. Since race didn’t pan out this way, all we are left with are sloppy – porous and insufficient – referents. Put simply, any criterion we want to use to racialize people can be easily shown to admit of problematic counterexamples.

Looking at Mills’ list, most folks today think race either has something to do with physical appearance, ancestry, or culture. But each of these is problematically porous. Not all black people and not all white people share the same physical characteristics. Skin colors fall on a spectrum, where some people we call “black” will be as white as some of the folks we call “white,” and vice versa. (Even hair texture runs along a spectrum that doesn’t break down easily by race.)

Ancestry won’t work either: first, because in the free world “mixed” ancestries have a long and accelerating history (one that actually proved problematic for governments attempting to enforce racist policy); second, migration patterns have been such that knowing someone’s current racial classification doesn’t tell me anything about how recently their “bloodline” has settled in any particular geographic area.

Culture is a horrible criterion for racial categorization for similar reasons. In liberal societies, pure cultures don’t exist, and people never adhere to pure cultural traditions. If we want to find out if Rebecca is/was black based on her cultural behavior, then that would mean that she was not black before summer camp and became black afterwards, a result few would accept.

Race is self-contradictory

Not only can race shift by referring to any number of criteria, none of which are sufficient on their own, but these criteria also can be internally contradictory, pointing in different directions. A light-skinned person with moderately curly black hair can be born to darker-skinned parents who descend from African slaves. This person can “pass” as white by enmeshing themselves into cultural patterns society calls “white” and keeping their family history a secret. If we were in the Jim Crow south, we might want to know whether the person is white or black, insisting that the question has an answer: either “yes” or “no.” But if you are keeping score, they would be white if we looked at culture, public awareness of ancestry, experience (since the world treats them as white) and self-identification. They’d be black if we looked at ancestry and self-awareness of ancestry. Is the person white or black? Since different criteria point in different directions, there is no clear answer unless we engage in the thankless task of choosing one criterion and ignoring all of its limitations.

Race is a subterfuge

Because of the three arguments above, I am inclined to think of race as most often a subterfuge: a vague idea we use to stand-in for other ideas we really have in mind. When we talk about the differences between black and non-black hair-care, we have in mind certain physical differences that affect how hair and skin should be treated. Do we need race language to do that job? Probably not. When we talk about diseases that different racial or even ethnic groups are more susceptible to, we are talking about genetic and physiological differences between people that show up in averages when we group people with similar genomes. Do we need racial language to do that job? Probably not. When we talk about the difference between Rebecca talking white and black, we are talking about culturally acquired speech patterns that have to do with a bunch of factors like where one was born or raised, what media one consumed, who one learned to talk from, etc. Do we need race talk to explain that? In this case, I’d argue that maybe we do, but only to the extent that these cultural differences have a lot to do with the effects of historical racism that segregated communities. That only means that we must racialize the past, not that we need to racialize the present or future.

If my arguments work, this means that today’s race talk is inevitably clunky. When we refer to someone in racial terms, there are a load of things we could be referring to, each of those things on its own is probably not up to its promised task, the things together are probably mutually contradictory, and there are probably better terms we could use than racial ones. What to do with this? Do we use this as a justification to stop using racial language or stop talking about race?

I’m not going to go that far. I sympathize with Ibram X. Kendi and others when they say that while race has always been a bad idea, it has had too much material effect on our current social reality to allow us to retire it. If people we call “black” have a history of unfair treatment that led to existing inequality – all stems from racial categorizations – we can’t likely remedy that by refusing to use race terms in our talk about solutions.

What we can do might be something more akin to what we do when we talk about genes as “selfish replicators” or George Harrison’s guitar as “weeping.” We can talk in these ways and we can often assume that for reasons of shared language and culture, others will have some sense of what we mean. But we recognize that as with all metaphors, these have limited signifying value, especially if you try to take them literally. If someone doesn’t understand what I mean when I call genes “selfish,” I have no problem saying that it is just a metaphor. If someone wants to debate me about whether George Harrison made his guitar cry, I see if that person grasps the metaphorical nature of the statement before we debate the question. I can freely acknowledge that these are metaphors I use only because the language we have makes this a handier way – at least, sometimes – to express what I have in mind. Maybe we can start doing this more consciously with race talk.

Metaphors also obscure more literal ways of putting things. My printer didn’t “decide” to stop working. It is malfunctioning for reasons that I’d be wise to uncover if I want to solve the problem. How might we do this with race talk? Following the lead of my discussion partner Sheena Mason, I have taken to replacing some descriptive invocations of “race” (“She was stopped because of her race,”) with the more accurate “racism” (“…because of the cop’s racism.”). I’ve also taken to thinking about what other ideas I might be using ‘race’ as a stand-in for and using those terms rather than racial ones. (“Do I mean ‘black people,” “descendants of African slaves,” or something else?)

Sometimes metaphors are preferable to literal expressions, and sometimes, race talk might be best given how our language (and the cultural history it intersects with) works. But the clunkier the signifier, the worse it is at signifying in any given case.


[1] Partly, the reason why its existence as a floating signifier is more troubling for race than, say, talk of moral obligation is because most folks today have (rightly) come to think that race is an idea we should work to get beyond, where we are sort of “trapped” using it only because of our lamentable history. Not so with ideas of moral obligation, authority, or general intelligence.

[2] This concern – that race is shifting – is different from noting that the same word can have multiple meanings. When someone points out a break, they could mean a pause (in a meeting), a fragmentation (of a ceramic pot), a hurt (of a broken heart), a change of direction (in a river) or a few other things. The difference between break talk and race talk is that with the former, there are several established and settled meanings that are easy to disentangle with context clues, and everyone understands that there are multiple meanings of the word. When someone refers to a broken heart, we all understand that “break” can mean multiple things and to listen for the context clues to decipher the speaker’s meaning. Not so with many invocations of a person being black or white, where we often have no real sense that the labels can mean different things or which thing(s) the speaker has in mind.


  1. That little vignette about Carrol and her brother Connor really bugged me. I found it emotionally unsettling that two siblings, that I assume love each other dearly, had to engage on such a matter. Of course it all could have been done amiably and with familial humor and teasing, though I don’t think that was the intent of the author. Yet, probably all such adoptive and blended families or those with “mixed” parentage are eventually forced to confront the clunky question, who am I and to what culture do I owe most allegiance? A rite of passage that no child or young adult should have to make on what should be a nonexistent problem. Yet, as you quote Kendi, the cats been out of the bag for far to long and there is no way around it anytime soon. It cannot be wished away or ignored so we are left holding the bag and at the very least in the social and political sense acknowledge that we have to deal with, at least, the reality of grouping and ameliorating ethnic populations as representations of averages along a spectrum. There are a myriad of individual problems we are all subject to but there are certain shared problems peculiar to identifiable and self proclaimed groupings made up of people that share certain physical and cultural traits that are evident and in our current state of psychosocial evolution (perhaps genetically predisposed) if not imposed, self sorted.

    Yes, race talk is clunky in most respects pertaining to individuality, human rights and dignity but most people intuitively understand, most of the time, what is meant by the concept and, unfortunately, don’t always understand all the unnecessary baggage that comes along with it. Whether this is the mind using metaphor to express beliefs and or feelings seems moot, other than as an intellectual exercise.

    “When we talk about diseases that different racial or even ethnic groups are more susceptible to, we are talking about genetic and physiological differences between people that show up in averages when we group people with similar genomes. Do we need racial language to do that job?”

    This example is usually the one that is acknowledged as a legitimate concern that needs to be distinguished by some sort of population and geographic signifier. Does it make more sense to talk about peoples descendant from Sub Saharan Africa as being less treatable by certain drugs or regimens or more or less susceptible to certain diseases and conditions? Would this hold true for white South Africans? Or, do we need to specify non European or Asian Sub Saharan groups? Or is it more parsimonious to just say black people, as the term is generally understood to mean? After all, when race doesn’t socially matter the term black would be an excellent non derogatory shorthand or signifier for a certain group that share an “average” genome when considered in a specific context of scientific and medical concern.

    Excellent article KCK. Really insightful and thought provoking.

  2. Why is porousness problematic for the physical-appearance criterion of race? It seems that many of our categories are unproblematically porous, as well as unproblematically fuzzy, unanalyzable into necessary and sufficient conditions, and governed by a prototype-and-exemplar model. Seeing these as “problems” seemingly requires holding onto the inapt desiderata of identifying essences (and perhaps even an unrefined view of essences and natural kinds). In other words, once we fully drop implying an essence, these “problems” are no longer problems for the physical-appearance criterion.

    1. Good question. Normally, I agree; I woudlnt care about porousness because really, what counts is whether the signifier is servicable and points well enough to whatever is being signified. But here, that’s precisely the problem: there is enough ambiguity – if not an entire absence – of what is actually being signified race talk is very unuseful at particular times. Not all the time, of course. That’s why I don’t argue that this is any reason to reitre it. But in the instance of, say, transracial or interracial families or communities, there are quite a few times – “you are acting too white!” “I know you are multiracial, but what race do you want to check on the box?” or “you must appreciate your white/blackness” – where the proper response might be to point out that the emporer really is butt-ass naked.

  3. “In other words, once we fully drop implying an essence, these “problems” are no longer problems for the physical-appearance criterion.”

    If I understand you right, this has to be incorrect, and incorrect in a way that illustrates precisely the problem.

    In my Diversity class, one activity I love to play with the class is to put up pictures of “racially ambiguous” people and ask the class to racialize those people. They often get the answers wrong, and even when they get the answers right, I ask them to think about what their difficulty means for racialization generally.

    And maybe you think that this only seems a problem because the activity focused deliberately on ‘racially ambiguous’ people which I must have had to work to find. I’d just ask you to look a little bit into the large history of racial passing in the US – where “interracial” relations have been much more common than we like to think, often because it was quite legal for horny white slaveowners to rape black female slaves. The result is that there are a lot of really white looking people who – by the one-drop rule we in the US are sadly wedded to – turn out to be “black,” and a sizable number of unproblematically black-lookingn people who have more European (otherwise ‘white”) ancestry than they either think or would want to admit.

    We are a polyglot nation and the modern world is an increasingly polyglot world. I do not know how much longer even the purely physical idea of race will be tenable. I hope not long.

    1. Hi Kevin. Thanks for the reply.

      To respond, it seems that a full account of “race” has to address the physical appearances that enable different people to sort a random selection of DMV photos by “race” and get highly correlated results (especially if you refine the sorting procedure—e.g., you can put Blake Griffen in between black and white, instead of having to pick one or the other).

      To cut to the chase, we can take various types of differentiation and place them along a spectrum of extent of “realness” (or something of that sort). On the righthand side of the spectrum would be the natural kinds of fundamental physics. Somewhere in the middle are prototype/exemplar categories. Elsewhere, in no significant order, are the types of differentiation exemplified by music genres, planets/dwarf-planets/comets, biological species, sedans/SUVs/trucks, “toxic”/“non-toxic” plants, etc. You can take a pin, labeled “race,” put it on the spectrum closer to the righthand side and say, if biobehavioral racial essences were real, then “race” would be here. However, they aren’t real, so we need to slide the pin left. There can be various reasons to keep sliding further and further left; however, if it doesn’t stop at some point, then one cannot explain the sorting correlation.

      Alternatively, you could jump far to the left to the point of solely non-discrete variation in skin color and then, when you find correlations with eye color, hair texture, etc., start moving right.

      Rhetorically, at least, if the only moves one countenances are pushes to the left then one’s position is not differentiated from sliding all the way to saying that Ewan McGregor and Samuel L. Jackson cannot be differentiated in any racial sense, which seems like a reductio, both scientifically and politically.

  4. Great post. Found it thought provoking. It’s strange how many different things get grouped in race. There is black and white. But then also sometimes religion or culture. Also landmass (as in Asians). Then brown people which can include people from Mexico, the Middle East and Asia. On top of it there is mixedness.

    But, seconding rgressis’ comment, if race is a murky concept, how is racism any clearer? In the post it seems like you are looking clear eyed (in a cool ordinary language philosophy way) at the different uses of ‘race’, but then pull back from what this means for ‘racism’. I agree it seems wrong to conclude from murkiness of ‘race’ that there are no historical and cultural power dynamics to grapple with, as if all is ok. But the challenge then seems to be (to continue the clear eyed descriptive project) to articulate the different power dynamics at play without assuming a general explanation like racism illuminates the issues. It’s like showing how there is no one thing games have in common, but then saying what motivates playing the variety of games is winning. But ‘winning’ itself is meant in different ways, and is as multifaceted as ‘games’.

    If the cop stopped her because of his racism, that might mean he stopped her because he sees her through certain stereotypes, which reflect assumptions of who is seen as in the in group and who isn’t, etc. But this merges with issues of class, education, dress style, etc. Maybe he is well off and she is poor. Or he is poor and she is rich and driving a Benz; then is his stopping her motivated by racism or envy or insecurity or all of these? Often there is no deeper, “real” cause in these socially charged cases, anymore than there is a “real”, if implicit decision behind most actions.

    People invent essences when they want to do something and not be questioned. Psychologists speak of the mind because it is supposed to be their province of expertise. In the 1700s and 1800s people started with race as a way to push some people around. Nowadays racism is used to push back and push others around.

    1. The way I see the difference between talk about race and talk about racism is this:

      Talk about race is talk about something that, as I said, is terribly hard to pin down.
      Talk about racism is talk about someone’s belief in that thing that is terribly hard to pin down. Identifying that folks have this belief is, however, not hard to pin down.

      Let’s say that ‘god’ is a fuzzy concept where we are not ever sure what someone means by it. Surely, even if the term itself is hard to pin down, it is not har to tell whether someone has a belif in god. The fuzziness of god doesn’t translate to unsureness about identifying who is and isn’t a believer.

      In the same way, race can be entirely fuzzy, but it is not at all hard to tell when someone believes in race and holds it to be significant (the classic definition of racism). I maya not know what the KKK means when they talk about race pride or how the inferiority of black people owes to their race. But that doesn’t mean it is hard for me to tell whether they are racist, whether they hold some idea of race to be significant.

      1. Interesting response. I hadn’t thought of that way of seeing the difference. Have some reservations though.

        In the KKK case, I wonder if the distinction you are drawing is because you are looking at it from a third-person perspective. Not sure the distinction makes sense from a first-person perspective. Can a KKK person say “It’s unclear what race means, but I believe my race is good and theirs is inferior”? If he concedes race doesn’t have a clear meaning, what is his belief tracking? Seems to me both race and racism have fallen out of the picture, and what is really doing the work is demonstrative concepts like “those people” and my people”.

        Maybe the KKK person and his opponent would understand each other better and be clearer about their disagreement if they were forced to not use racism. So the KKK would have to say things like, “The group I belong to is better than yours,” and the opponent would say, “Your group is backward and unsophisticated, so my group is better.” These seem to me more what they are really thinking. Strange as it may seem, a focus on racism becomes a way to not express the really judgmental thoughts both sides have of the other. Better they express the clear thoughts even if they are raw than murky thoughts which give a vaneer of debate about race and racism.

Leave a Reply