Crying “Crying Wolf”


Suppose that instead of one shepherd boy, there are a few dozen. They are tired of the villagers dismissing their complaints about less threatening creatures like stray dogs and coyotes. One of them proposes a plan: they will start using the word “wolf” to refer to all menacing animals. They agree and the new usage catches on. For a while, the villagers are indeed more responsive to their complaints. The plan backfires, however, when a real wolf arrives and cries of “Wolf!” fail to trigger the alarm they once did.

The above variation on the classic fable of the boy who cried wolf is by Spencer Case, who recently earned his PhD in philosophy from the University of Colorado. He uses the story in an article at Quillette to illustrate what he calls “concept inflation.”

Concept inflation occurs, Case says, “when speakers loosen the usage of an emotionally impactful word in order to manipulate an audience.” He thinks that certain uses of terms like “violence,” “gaslighting,” “racism,” and “sexism” are often deployed in concept-inflating ways:

Most dictionary definitions of “violence” mention physical harm or force. Academics, ignoring common usage, speak of “administrative violence,” “data violence,” “epistemic violence” and other heretofore unknown forms of violence. Philosopher Kristie Dotson defines the last of these as follows: “Epistemic violence in testimony is a refusal, intentional or unintentional, of an audience to communicatively reciprocate a linguistic exchange owing to pernicious ignorance.”

What Dotson calls “epistemic violence” isn’t violence according to ordinary usage or the dictionary. If intellectuals can commandeer the word “violence,” then presumably they can do the same with stronger words. So why not call epistemic violence “epistemic rape”? Indeed, why not “epistemic genocide”? After all, genocide is destroying a people in whole or in part, and part of destroying a people is destroying its voice. Maybe that can be done through subtle acts of silencing. This is absurd, of course, but there’s no principled way to stop moves like this if we accept coinages like “epistemic violence.”

What’s bad about concept inflation, according to Case? One thing is that it makes certain terms less effective:

When speakers expand the reference of a word in order to attach its associations to new things, they dilute the associations of the original word. Just as printing too much paper currency diminishes the value of the currency, concept inflation degrades the rhetorical effect of inflated words and phrases.

Another is that it’s like lying:

Immanuel Kant observed that lying couldn’t be effective in a world where everybody lied, since no one would be believed. Just as lying is parasitic on a truth norm, concept inflation is parasitic on norms of usage.

It’s also overly provocative: it’s “a strategy for activating disproportionate or unreasonable moral responses.”

Case thinks we shouldn’t engage in concept inflation, and should call people on it when they do it. If you see something, say something. Case writes:

We all have the responsibility to be good stewards of the languages we speak. We shape it when we decide to accept or reject new coinages or expressions… When we allow sloppy language to proliferate—for example, when we use the word “literally” to mean “metaphorically”—we degrade language and make it harder for everyone to communicate. This is analogous to polluting a common resource like the water or air. If some way of using a word seems fishy, then take your own reaction seriously and make your concern known. 

What should we make of concept inflation? Certainly the meanings of words change over time, and such changes may be good or bad in various ways. I’m not sure that concept inflation is bad in the ways Case identifies, though.

I’ve heard the phrase “epistemic violence” hundreds of times. Has this “degraded the rhetorical effect” of the word “violence” for me? I don’t think so. When I first heard it, I did have questions; like Case, I tend to associate “violence” with physical aggression. But then I was reminded of all of the ways in which we already use “violence” metaphorically, to describe coughing fits, brush strokes, turns, color clashes, etc. No one objects to using “violence” in these descriptions, even though these are not instances of physical aggression. The use of “violence” in these contexts does not seem to render it meaningless when it’s deployed in more traditional ways.

It may be worth observing that the metaphorical and hyperbolic uses of only certain terms end up coming in for criticism along these lines. Consider the phrase “attack on free speech.” It has been used to describe, among other things, requests to address people politely, criticisms of people’s speech, blog comment moderation, withdrawals of invitations to speak, and the like. These phenomenon do not exemplify the primary definitions of “attack,” which tend to include references to aggression, physical force, injury, even weapons. Yet I don’t recall anyone raising concerns about concept inflation when hearing about “attacks” on free speech. We seem to have been able to cope just fine with whatever loss of “rhetorical effect” the word “attack” has suffered through its metaphorical and hyperbolic use. Why worry about “violence” then?

(We’ve seen this kind of pattern before: structurally similar phenomena receiving either dismissive or approving labeling depending on whether it serves the interests of the less or more powerful.)

Let’s go back to Case’s fable. Why are the shepherd boys crying wolf? “They are tired of the villagers dismissing their complaints about less threatening creatures like stray dogs and coyotes.” Maybe dogs and coyotes are less dangerous than wolves, but they could pose serious problems for the boys nonetheless, and no one’s paying attention to these problems. The boys have been led to believe that the only way they are going to get help is by crying wolf.

This aspect of the story doesn’t get much attention from Case, but I think it’s crucial. Case thinks that concept inflation is bad and that the way to counter it is to take care with our language. How about taking care with people? If the other villagers had been listening to the shepherd boys and had been understanding of their problem, the boys would not have had to resort to crying wolf. But they were unheard, and crying wolf reasonably seemed like their best option. Likewise, those Case accuses of concept inflation might also be unheard: they don’t believe they are being listened to, or that their problems are being taken sufficiently seriously. And so perhaps they are reasonably drawn to bring attention to their problems with more dramatic turns of phrase.

This suggests that if one is opposed to concept inflation, a way of combating it is by listening to people and taking their concerns seriously.

If one cares about there being a culture of robust, quality, disagreement, an element of that is being vigilant about restrictions on and pressures against speaking up. But another that’s just as important is being careful to actually hear what others are saying.

To label an instance of speech “political correctness” is to brand it an ignorable complaint of the oversensitive. To label an instance of speech “virtue signaling” or “moral grandstanding” is to accuse it of being made in self-serving bad faith. To label an instance of speech “concept inflation” is to call it a kind of harmful lying. These are all labels that tell us to dismiss what is being said, rather than try to understand it or engage with its substance.

We already pretty good at not listening to others, to not taking seriously experiences different from our own, to not hearing what would disturb our complacency. If we care about ideas—and if we care about people—we should be wary of means that make it easier for us to dismiss them.

photo of “Head On” by Cai Guo Qiang


(Note: Dr. Case sent me a link to his Quillete article earlier this month and we discussed it in a series of emails which covered some of the aforementioned points. At the end of that exchange, he voiced his preference for a critical post about his article rather than none.)

guest
131 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
krell_154
krell_154
2 years ago

I guess the problem is that some people perceive it as villagers crying “Wolf” not only in response to coyotes or stray dogs, but rabbits, kittens and mice. And they don’t think it’s reasonable from villagers to demand help in dealing with rabbits, kittens and mice, because they do think the villagers’ fear is overblown.Report

Someone
Someone
2 years ago

I think concept inflation of the sort Case discusses is bad because a central practical effect is to motivate people fed up with this kind of thing to elect people like Trump and Bolsonaro to office. Neo-nazism and right-wing authoritarianism are spreading throughout the world, and one of the main rallying cries is “political correctness.” I fear that the more people on the left call things ‘violence’, the greater the chance the election of authoritarian monsters who will commit actual mass physical violence.Report

Mark Alfano
Reply to  Someone
2 years ago

So let me get this straight. Criticizing crypto- fascists makes fascists? Is that the argument?Report

Mark Alfano
Reply to  Someone
2 years ago

None of this is evidence. These are op-ed. If you have evidence, I’m willing to take a look. Otherwise, I’ll have to conclude that you think we should let the crypto- fascists win to keep the outright fascists from winning.Report

Someone
Someone
Reply to  Mark Alfano
2 years ago

Fine here you go

“Donald Trump as a Cultural Revolt Against Perceived Communication Restriction: Priming Political Correctness Norms Causes More Trump Support”, Journal of Social and Political Psychology, 2017.

https://jspp.psychopen.eu/article/view/732

“Abstract

Donald Trump has consistently performed better politically than his negative polling indicators suggested he would. Although there is a tendency to think of Trump support as reflecting ideological conservatism, we argue that part of his support during the election came from a non-ideological source: The preponderant salience of norms restricting communication (Political Correctness – or PC – norms). This perspective suggests that these norms, while successfully reducing the amount of negative communication in the short term, may produce more support for negative communication in the long term. In this framework, support for Donald Trump was in part the result of over-exposure to PC norms. Consistent with this, on a sample of largely politically moderate Americans taken during the General Election in the Fall of 2016, we show that temporarily priming PC norms significantly increased support for Donald Trump (but not Hillary Clinton). We further show that chronic emotional reactance towards restrictive communication norms positively predicted support for Trump (but not Clinton), and that this effect remains significant even when controlling for political ideology. In total, this work provides evidence that norms that are designed to increase the overall amount of positive communication can actually backfire by increasing support for a politician who uses extremely negative language that explicitly violates the norm.”

You want more? I can keep ‘em coming.Report

JDB
JDB
Reply to  Someone
2 years ago

One might also take a look at more general literature on the related phenomenon of “concept creep” (which Case distinguishes from his phenomenon in a footnote).
https://psycnet.apa.org/record/2016-08154-001Report

Mark Alfano
Reply to  Someone
2 years ago

Sorry, I should have mentioned that I want *good* evidence. This study is mediocre at best. E.g.:

(a) The Cronbach’s alpha for the restrictive communication concern variable, one of the *key* outcome variables was .55. That’s on the border between “unacceptable” and “poor.” Essentially what this means is that we don’t know whether this construct measures anything at all.

(b) The alpha for the other key outcome variable, informational contamination, was .71, which is slightly better but still quite poor, especially for a new and previously un-validated measure. So we should be highly skeptical that this study shows *anything*, let alone that it shows what you say it shows.

(c) The main outcome is correlational, but you are making a causal claim. In particular, “Persons who felt chronic
reactance at restrictive communication norms were significantly more likely to support Trump (overall support r =
.38) and significantly less likely to support Clinton (overall support r = -.40) – and both of those effects remained
significant even when controlling for political ideology (p’s feeling of constraint –> fascist politics. Neither causal arrow has been established here.

(e) The second study is a p-hacked priming study. Nuff said.

Sorry, try again (or don’t). This is nowhere near good enoughReport

Mark Alfano
Reply to  Someone
2 years ago

Somehow deleted part of my point (c) above. The problem is that with a correlational result like this, we don’t know which variable exerts a causal influence on the other. Does Trump support cause feeling of constraint, or the other way around (or does some third thing cause both of them)? In addition, the study establishes no connection between what we are allegedly talking about (criticizing far-right politics) and support for even-more-far-right politics. What you need is criticism –> feeling of constraint –> fascistic politics. But what if right-wingers are going to feel constrained no matter what liberals and leftists do? My contention is that they will because of Fox News, Breitbart, and the rest of the right-wing propaganda machine. (“War on Christmas,” anyone?) So neither causal arrow has been established and this study does not in any way support your claim.Report

Mark Alfano
Reply to  Someone
2 years ago

“You want more? I can keep ‘em coming.”

Your silence indicates that you realize that you were way out over your skis and are afraid of getting dog-walked again.Report

Carnap
Carnap
Reply to  Someone
2 years ago

I assume that Mark Alfano’s contribution to the (forthcoming) Routledge Handbook of Humility consists of an argument against it.Report

Mark Alfano
Reply to  Someone
2 years ago

Close. It’s about how true humility requires criticism of one’s ingroup. Which in my case = white Americans.Report

Oliver Traldi
Oliver Traldi
Reply to  Someone
2 years ago

You’ve got the right theory, just a wildly wrong application. An ingroup is a group with which one psychologically identifies, and hence which one finds *difficult* to criticize, for various reasons: it’s difficult epistemically, i.e. the group almost always seems correct to begin with; it’s difficult socially, i.e. there are interpersonal and maybe even professional costs to the criticism; it’s difficult psychologically, i.e. there can be something about doing it that just feels wrong and contrary to everything one is made of. In your case, criticism of the ingroup might mean criticism of politically progressive philosophers, for example, while in my case that would pose no difficulty at all.

It’s important to note that “ingroup” is a moving target. For instance, if you worked for a mainstream conservative publication in 2015, you would have been in lockstep with your ingroup in opposing Donald Trump’s presidential campaign. If you maintained that opposition in two years later, that would have been in far more tension with your ingroup. Well, say you decided around that point that you weren’t a conservative anymore, that you had abandoned the ideology because Donald Trump was such an abysmal nightmare. By now, 2019, you would have probably developed a new ingroup: this ex-Republican NeverTrump crowd. The psychological pull of conformity and groupthink would be coming from this new source, and you could no longer claim epistemic virtue in opposing the Trumpian crowd, with whom you, ex hypothesi, no longer identified.

There is also (I blogged about this a couple years ago) a very plausible and rational mechanism for increased ingroup humility, which is that we tend to see the outgroup as an undifferentiated mass, but the ingroup as a set of unique individuals with a lot of internal dissent and disagreement. In light of things like the independence condition on Condorcet’s Jury Theorem, a consensus among a disparate group with little internal correlation is far more reliable than a consensus among a groupthinking mob, with massive internal correlations. So epistemically speaking, the relevant ingroup is probably also the group one sees as being made up of free-thinking individuals, and the relevant outgroup is probably the group that one sees as being brainwashed by society, talking heads, propaganda, etc.Report

Mark Alfano
Reply to  Someone
2 years ago

Oliver: fair enough. My group has been investigating this sort of thing in, e.g., https://philpapers.org/rec/SULCRS.

I have a chapter about the relevant virtue for this context in my Nietzsche book (chapter 10), which will be out with CUP in a few months. (https://www.cambridge.org/core/books/nietzsches-moral-psychology/76A9BD1E87606D01CE8AA67293029614)Report

Kenny Easwaran
2 years ago

I like the analogy to currency inflation here.

In a toy economic model where we have laborers and corporations, there are two sorts of transactions – the corporations pay the laborers for their labor, and labors pay the corporations for goods. It doesn’t really matter what the specific prices are for these two types of transactions, as long as they match.

The problem is when there’s a persistent mismatch between the amounts demanded on the two sides – when wages are low but costs of goods are high workers run into trouble, and when wages are high but costs of goods are low companies go out of business (and both things cause problems for everyone). The usual fix is for the government to print extra money to give to one side, or for the government to raise taxes on one side, to help rebalance things.

Inflation occurs when both the costs of labor and the costs of goods increase. Contrary to popular opinion, it isn’t actually a problem when both costs are high, though the transition from low nominal wages/prices to high nominal wages/prices hurts people with high savings and helps debtors, and the transition in the opposite direction does the opposite. (usually at a different nominal price level than they currently are, either by deflating or inflating).

Semantically speaking, there’s a bit more complexity. With currency, one can switch from using $1 bills to $100 bills (or from $1 bills to 100 Yen notes), with the only cost being borne by people with savings. But it’s often harder to create new words that have higher denomination than old ones. Furthermore, the addition or subtraction of value from the semantic economy doesn’t need some third-party government – it can be done by either side on its own.

Still, none of this changes the fact that the fundamental problem is due to the mismatch between the two sides, and not because one side is objectively wrong with their demand being too high or too low.Report

Dale E Miller
2 years ago

Inflation doesn’t render currency valueless; it just diminishes its value. In the most extreme cases it may lose nearly all of its value, but those are unusual. So the question we should ask isn’t whether using some term more broadly than we have in the past renders it “meaningless.” It’s whether the term loses some of its force. Does it provoke less of a reaction from us? To what extent this has happened with terms like ‘violence’ is an empirical question, to which I don’t have an answer. Of course, this doesn’t gainsay Justin’s point that sometimes concept inflation might be forced upon people when their legitimate concerns aren’t getting appropriate attention otherwise. Nor his point that we can find examples of this from across the ideological spectrum, although to the best of my memory Spencer doesn’t deny this.Report

Mark Alfano
2 years ago

I for one will not rest content until we all switch back to proto Indo-European. Let’s undo all these corruptions of the discourse and use pristine words that mean what they meant when they first emanated from the mind of God.

Hey, maybe I should write for Quillette!Report

E
E
Reply to  Mark Alfano
2 years ago

What do you think you’re contributing to the exchange with this post?Report

Mark Alfano
Reply to  E
2 years ago

Prescriptivism. Is. False.

Now do you understand?Report

Justin Kalef
Justin Kalef
Reply to  Mark Alfano
2 years ago

Hi, Mark Alfano.

I think I understand. You’re saying, if I’m not mistaken, that no term that people actually use to refer to a group of other people, say, could be mistaken. Is that right?Report

Mark Alfano
Reply to  Justin Kalef
2 years ago

Yes, that is correct.Report

E
E
Reply to  Mark Alfano
2 years ago

And the ugly way in which you’ve said it has led ~98 people to tune out, and think less of you.

Some of us have been trying actively to combat the stereotype of philosophers being aggressive, mean spirited, condescending, and the like. You’re not helping.

But thanks for asserting that prescriptivism is false. That possibility surely hadn’t been on anyone’s mind.Report

Dale Miller
Reply to  Mark Alfano
2 years ago

But surely the claim that Justin Kalef attributed to you is false, and you meant to say something much weaker about the meaning of terms changing over time. I mean, surely groups of people are constantly being referred in ways that are mistaken, or at least that involve mistakes. A relatively uncontroversial example ought to be ways in which some people refer to Jews. The meaning of the language has not evolved enough for all of those terms to be appropriately applied.
More generally, while language evolves over time, we can still see some shifts in meaning as costly and worth trying to resist if it’s not too late. First we lost the original meaning of ‘begging the question’, and now we’re losing that of ‘refute’. Since the new meanings of these terms ones for which we already had handy phrases, and since we don’t have equally handy phrases for the old meanings, these are losses worth mourning.Report

Justin Kalef
Justin Kalef
Reply to  Mark Alfano
2 years ago

Hello, Mark Alfano.

I must admit to being very surprised by this. How can it be that left-wingers who are in favor of identity politics would oppose linguistic prescriptivism? It’s long been a mainstay of that view, as I understand it, that we need to be very careful in which words we select, and moreover that certain people’s pronouncements about which words we should and not use to refer to groups they belong to must be taken as sacrosanct. But how can that not be prescriptivism? Or do you reject that view?Report

Mark Alfano
Reply to  Mark Alfano
2 years ago

Semantic prescriptevism.Report

Justin Kalef
Justin Kalef
Reply to  Mark Alfano
2 years ago

Hi again, Mark. I don’t think this sharp line you seem to be drawing can do the work it’s meant to do.

To illustrate: suppose someone uses the following two techniques in an attempt to reduce racial discrimination:

a) She says ‘Jewish person’ rather than the word ‘Jew’, and sometimes tells others that ‘Jewish person’ is the correct word to use.
b) She only uses the phrase ‘white supremacist’ to describe people who could plausibly be part of the group headed up by the main character in _American History X_ before his imprisonment.

Her motivation for both these things is moral. She thinks that ‘Jewish person’ is much more respectful than ‘Jew’, which sounds to her like a slur (and she points to many instances in which people have used ‘Jew’ in a way that looks like a slur). She also thinks it’s important to preserve the emotional intensity of the phrase ‘white supremacist’ in order for it to have the desired social effect when it’s deployed, and she is wary of diluting that effect through what she sees as frivolous overuse.

Some criticize her by pointing to cases in which ‘Jew’ has been used with a clearly non-insulting aim, and in which ‘white supremacist’ has been used in a much broader sense. She acknowledges the existence of such cases, but says we still ought not use the words in those ways.

Am I right that you’d condemn her move to limit ‘white supremacist’ as based on a theoretical falsehood, but that you don’t think the same applies to her move to replace ‘Jew’ with ‘Jewish person’? How does this work, exactly?Report

Mark Alfano
Reply to  Mark Alfano
2 years ago

Look, I am not going to explain basic concepts in philosophy of language to you in the comments of a blog. If you don’t understand, maybe start with Wikipedia and then go to the SEP.Report

Andrew Sepielli
Reply to  Mark Alfano
2 years ago

Justin Kalef —

Out of the goodness of my heart, I have done the hard work for you. I looked up “semantic prescriptivism”, and got 38 results — the top two of which were from a Tumblr called “Terfs are Trash”, and a J.R.R. Tolkien subreddit. The latter provides dispositive support for Mark Alfano’s view:

“The idea that prescriptivism is arbitrary and silly is a pretty key part of linguistics, and as Tolkien was a linguist I think we should respect that.”

Just like the “Chewbacca was a wookie” defence, only nerdier.Report

Mark Alfano
Reply to  Mark Alfano
2 years ago

There’s a whole wikipedia page about it.

https://en.wikipedia.org/wiki/Linguistic_prescriptionReport

Justin Kalef
Justin Kalef
Reply to  Justin Kalef
2 years ago

Since semantic prescriptivism is passe and the Humpty Dumpty theory of language is in, I hope nobody will fault me for reading the words “I’m not going to bother explaining basic concepts to you” as meaning “I can’t think of any good reply to that.”Report

Mark Alfano
Reply to  Justin Kalef
2 years ago

Yes, it is true that there are only two options: revanchist prescriptivism and the Humpty Dumpty theory of language. That is the complete logical space. Thank you for your clever use of disjunctive syllogism.Report

Mark Alfano
Reply to  Justin Kalef
2 years ago

Holy wow, there’s a third option!

http://www.oxfordscholarship.com/view/10.1093/acprof:oso/9780198712053.001.0001/acprof-9780198712053

I guess the disjunctive syllogism isn’t valid in this case after all. What a pity.Report

Justin Kalef
Justin Kalef
Reply to  Justin Kalef
2 years ago

Holy wow, I wasn’t using disjunctive syllogism or implying there were only two options. “Since ~A and B” does not imply that B follows from ~A.

If you’d like to answer the question I posed at 8:06am yesterday, please do. If not, that’s fine: you can back away. But please pick one or the other.Report

Mark Alfano
Reply to  Justin Kalef
2 years ago

(1) You sure seemed to be using disjunctive syllogism. In particular, you wrote, “semantic prescriptivism is passe and the Humpty Dumpty theory of language is in.” The second conjunct is only something one would infer in this context if one accepted the disjunction of prescriptivism and HD. After all, the only thing I said above that was relevant was that prescriptivism is false. (My 8:45 AM comment was sarcastic, in case that wasn’t clear.)

(2) I’m not particularly interested in your 8:06 comment, but I’ll respond to it below anyway. My primary concern here is with Case’s prescriptivism. By my count, he refers to ‘original’ words and meanings five times in his Quillette piece, and he argues that we should be wary of changing the meanings of words and phrases from their “original” semantics. *That* is what I am criticizing. Sometimes change is good (morally, epistemically, aesthetically, perhaps in still other ways), sometimes bad, sometimes neutral, sometimes mixed. It all depends on the context. I reject the idea that we somehow owe it to the past to respect, as a default stance, alleged “original” meanings.

(3) Your 8:06 comment now seems to me to be a non sequitur, but anyway, here’s what I think: the hypothetical person you envisage — let’s call him ‘Ken Justfail’ — is a moral imbecile. By Ken’s lights, John C. Calhoun is not a white supremacist. After all, there’s no evidence that Calhoun ever got tatted up with swastikas or that he curb-stomped anyone. As far as I know, he never even tried to kill anyone. Instead, he wielded the power of elected office to institutionalize white supremacy. We have no reason to think that Calhoun would have dirtied his hands in the same way that Vinyard does in AHX. But anyone who is not willing to call Calhoun a white supremacist is either a monster or a moron. So I would accept Ken Justfail’s decision to use the phrase ‘Jewish person’ but criticize in the strongest terms his decision to refrain from calling people like Calhoun white supremacists.

What would you do?Report

Justin Kalef
Justin Kalef
Reply to  Justin Kalef
2 years ago

Hi, Mark.

I definitely wasn’t using, or implying, disjunctive syllogism. My comment was similar to “Bowties are out and golf shirts are in.” There is no inference there that not-bowties entails gold shirts.

From your fuller comment now, I have a clearer sense of what you are thinking of: what you had written before was really just hints toward an argument, and I was trying to connect the few scattered clues you had left us to go by.

So: you say that “we should be wary of changing the meanings of words and phrases from their “original” semantics. *That* is what I am criticizing. Sometimes change is good (morally, epistemically, aesthetically, perhaps in still other ways), sometimes bad, sometimes neutral, sometimes mixed. It all depends on the context.” That’s fine, and I think I agree. However, it doesn’t sound to me as though you really are rejecting semantic prescriptivism. In saying that change in meanings is sometimes good and sometimes bad, you’re giving room for evaluative statements about semantic change, and I’m not sure all that much more is needed to get us to saying that some ways of using words are better than others. A committed semantic descriptivist would say that the (fully legitimate) meaning of a word is whatever some part of the linguistic community takes it to mean. The view you’re expressing now seems much more sensible than that, fortunately.

Correct me if I’m wrong, but I don’t recall Case arguing anywhere that we ought to limit words to their original meanings _merely because_ those are the original meanings. There are many English words whose meanings have changed significantly over the centuries, and I don’t see anyone arguing that we should ignore that.

But it’s a very different thing if a word is _currently_ used in a much more limited sense by most people, so that that word is bestowed on people or institutions in a way that exposes them to severe criticism, and then some people begin using that same word in a much more expansive manner. In that case, there can be a great deal of confusion and mischief when people who don’t realize that the word is being used with a new meaning really think that the person addressed has done something he or she didn’t do. And even if the new usage becomes widely accepted, there can be a serious loss of clarity, with bad consequences, if the result is that nobody can distinguish the limited meaning from the very broad meaning any longer.

Is the trade-off worth it? As you say, we can look at these things on a case by case basis. But there doesn’t seem to be anything wrong with considering carefully whether the change in question is a good one.Report

Oliver Traldi
Oliver Traldi
Reply to  Justin Kalef
2 years ago

Some interesting ideas here, Dr. Kalef – thank you. I wonder if there is an analogy to the settlement function in law. There, the idea is that we respect precedent even in the case of at least some imperfect decisions because (1) it is exhausting to continuously relitigate the same issue, and (2) law is a tool, and must be settled to a certain extent in order to be useful. I wonder if there is a similar settlement-function argument available with regard to language.

I also think there are some fairly obvious ways in which constant language changes benefit some people, to the detriment of others. They confer status on those who can keep up with them – largely young, well-educated, neurotypical native speakers who have the time to read a lot of articles in magazines or academic journals. And they render evil-seeming those who can’t keep up with them – largely older people, the poorly-educated, those with certain kinds of disorders (autism or OCD, for instance, might be correlated with greater rigidity in language use), and those who don’t speak English as a first language. For this reason, it’s not surprising that successful academics and writers would be pushing constant linguistic change, although it is surprising that such pushes would be associated with social justice.Report

Justin Kalef
Justin Kalef
Reply to  Justin Kalef
2 years ago

Those are some very astute observations — so astute that I feel bad that most people will probably not read them because of how far down in this thread they appear.Report

Mark Alfano
Reply to  Justin Kalef
2 years ago

I see that, instead of answering my question, you have changed the subject. Perhaps you realized that your trap was so ill-conceived that you ended up suggesting that Calhoun was not a white supremacist.Report

Justin Kalef
Justin Kalef
Reply to  Justin Kalef
2 years ago

Not at all, Mark. I grant you the point about John Calhoun, as it’s merely incidental to the point I was making, as anyone who’s taking the time to sincerely follow the dialectic here has already seen very easily. The issue has to do with prescriptvism versus descriptivism as applied to these cases.

I’m satisfied already with the result of this exchange and think we’ve got some important things figured out. If you think that add more comments to this thread will cast better light on the matter, or enhance your reputation as a philosopher and a person, then that’s your perfect right, of course. However, II’m less than sure that I’ll see the need to add more to the discussion. Take care.Report

C.W.
C.W.
Reply to  Mark Alfano
2 years ago

The Finns would like to have a word!Report

Alastair Norcross
Reply to  C.W.
2 years ago

Revontulet. They can have that word. A gold star for the first person who can tell me what it means, without looking it up.Report

Andrew
Andrew
Reply to  Mark Alfano
2 years ago

This entire sub-thread has been a master class in how to divert a discussion with snide, condescending, and irrelevant comments. Really impressive stuff.Report

Ben Almassi
Ben Almassi
2 years ago

I guess I would like to hear more about the part where a real wolf arrives and cries of “wolf” fail to trigger the alarm they once did, where by hypothesis “wolf” has been extended to include all menacing animals. Why should this concept extension have such an effect, if these animals really are menacing?

I worry that the concept of concept inflation as Case presents it here conflates the mere extension of a concept into new application with the hyper-extension of a concept into an application where it doesn’t belong. I suppose these might seem the same if we take our concepts to be pretty static — a wolf is a wolf is a wolf. But Case’s example reminded me of Jeremy Saulnier’s recent film Hold the Dark, where a little conceptual flexibility about who or what counts as a wolf to be feared is quite valuable. Or just think about violence as we think about mass – Einstein and others were certainly engaged in concept extension when they (re)conceptualized mass as they did, but the mere fact that of concept extension does not yet tell us whether the concept is being extended *too* far or not. I can only report that I find relativistic mass and epistemic violence to be useful extensions of my hermeneutical resources.Report

IGS
IGS
Reply to  Ben Almassi
2 years ago

I believe you are missing the point of the fable with this comment. If the villagers *see* the wolf, then a wolf is a wolf is a wolf, but the point of the fable, as I understand it is that if the villagers wait until they see the wolf, then it will be too late. Proactively responding to the first alarm is the only way, and that is taken away by their disbelief in the warning.

As I understand it, the problem suggested is not that anyone would lose the ability to identify violence when they see someone punch another person. The problem suggested is that they will no longer envision such a thing when someone else uses the word violence.Report

Ben Almassi
Ben Almassi
Reply to  IGS
2 years ago

I think understand the fable well enough. My question was why we should accept the yadda-yadda that this concept extension will lead to disbelief in the warning, given that by hypothesis “wolf!” is being used as a warning for genuinely menacing animals.Report

IGS
IGS
Reply to  Ben Almassi
2 years ago

Presumably because the ideal response to a stray dog is different than the ideal response to a wolf; hence, responding as if there were a wolf is wasteful.Report

Ben Almassi
Ben Almassi
Reply to  IGS
2 years ago

With respect, a “stray dog” is not relevant here. If you and I are having a good-faith discussion of Case’s story, please recall that menacing animals was the extended reference class. And there it is not nearly so obvious what is “wasteful” about responding with alarm.Report

Jeff H
Jeff H
Reply to  Ben Almassi
2 years ago

Various snide replies to that suggest themselves, but I’ll confine myself to suggesting you re-read the first sentence of Case’s story before accusing others of bad faith for using *precisely his examples*.Report

Ben Almassi
Ben Almassi
Reply to  Ben Almassi
2 years ago

Jeff H, you are absolutely right. IGS, if you happen to read this, please accept my sincere apology, I was guilty of reading too quickly.Report

Paul
Paul
2 years ago

Where and when does Case find the gold standard for ‘violence’? Not in my lifetime.
Webster’s Third (1961) second definition is “injury in the form of revoking, repudiation, distortion, infringement, or irreverence to a thing, notion, or quality fitly valued or observed”. as in “no violence has been done to expert military opinion” — Winston Churchill

The real thought stoppers are gratuitous uses of ‘assault’, ‘rape’, ‘murder’, and ”genocide” in intellectual discussions best pursued without anyone feeling one is personally on trial for crimes against humanity. Such truly inflationary uses of emotionally-charged language are self defeating, as a reasonable response under such circumstances is silence and egress. . .Report

Paul
Paul
Reply to  Paul
2 years ago

Strike (editorially, not physically, please) ‘assault’ in my comment above — long ago defined as “a violent attack with nonphysical weapons (as words, arguments, appeals) – Webster’s Third, second definition. So let a thousand violent assaults bloom in academia.

But please don’t call one’s intellectual adversaries genocidal, or a rapist or murderer, without a grand jury indictment. That would be criminal,, or at least libellous.Report

Paul
Paul
Reply to  Paul
2 years ago

Great discussion. Now, assume we are in the neighborhood of more closely circling an apparent problem in charitably and fruitfully listening to and responding appropriately to an intellectual adversary, or at least one using what we perceive as weaponized language: what is the proposed solution?

I would very much welcome Spencer Case, or others, to model reasonable, realistic, and engaged rejoinders to typical examples of unwelcome concept inflation, say, in the context of a classroom of young adult students of philosophy, or any venue of public intellectuals..

Very short, dialogic examples would be ideal. As quite successfully modeled in four editions of Gerald Graff’s and Cathy Birkenstein’s bestselling primer, “They Say, I Say: The Moves That Matter in Academic Writing”
https://books.wwnorton.com/books/webad.aspx?id=4294994973Report

Avalonian
2 years ago

Thanks for this, this exchange is useful and thought-provoking. One thought I have is that dictionary definitions of “violence” are going to be very unhelpful here. Of course we say things like “that does violence to the idea”, when speaking of a mere misinterpretation. But language isn’t so a-contextual, the word “violence” (like all words) has different pragmatics in different contexts. When, for example, it is used to refer to a wrong that one person does to someone else, (and not merely something that is done to an idea or to a concept) it may well strongly suggest some kind of force or serious physical/emotional injury. And this is the sort of context in which the inflation is occurring. So Case’s (sharpened) point should be: in loosening the pragmatic implications in these contexts, we lose the ability to reliably pick out a specific kind of harm or wrong in those contexts. That said, Justin does seem to be right that we need evidence for assertions of this kind… has concept inflation had this kind of result in the past, and not just in imaginary communities featured in children’s stories?Report

Andrew Sepielli
2 years ago

See, I woulda called it “concept ooze”. Shows how weak my neologism game is. I’m considering hiring someone like an ex-Gawker writer to help me.Report

Oliver Traldi
Oliver Traldi
2 years ago

Great work, Spencer!Report

Oliver Traldi
Oliver Traldi
Reply to  Oliver Traldi
2 years ago

For what it’s worth, I agree with Professor Weinberg that calling something “concept inflation”, “moral grandstanding”, etc. should come *after* one takes oneself to have offered convincing countervailing reasons, and not *as* such a reason, and that people are often very bad at listening to disagreement. The same, I think, goes for “slurring”, “silencing”, and the idea of “epistemic violence” itself. So, in a sense, I take the apparent disagreement between the two of them to belie an underlying agreement.Report

Spencer Jay Case
Spencer Jay Case
Reply to  Oliver Traldi
2 years ago

I’m not sure that I understand. someone says “white silence is violence” am I supposed to prove that it *isn’t* violence before making the accusation of concept inflation? Or am I supposed to first refute the idea that white people should speak out more to about police violence? It doesn’t seem like I should have to do this in order to claim that the slogan is inflationary. It seems like the fact that it is inflationary could be a reason for rejecting the slogan here. I don’t think that we should reject any actual political positions because they’re described in inflationary terms. So could you say more?Report

Oliver Traldi
Oliver Traldi
Reply to  Spencer Jay Case
2 years ago

I take the point to be a bit more general: When an interlocutor expresses their belief that P, my giving a description of their speech act, rhetorical style, etc. under some theory does not serve the same role as my providing evidence that not-P. I take Dr. Weinberg’s contention to be that “white silence is violence” expresses some proposition that can’t be rebutted by saying, “How is it violence when I’m just standing here?” (although I myself believe that there is also another sense in which that’s a perfectly good response). We would similarly think someone has missed the point if they slam their copy of Romeo and Juliet shut in anger and muttering about fission reactions when they read the phrase “Juliet is the sun”.Report

Spencer Jay Case
Spencer Jay Case
Reply to  Oliver Traldi
2 years ago

Ok, but I could fairly ask them to express that proposition in different language, right? Because it’s not immediately clear what that means and the unusual use of the word violence triggers my suspicions, reasonably I think. The request for hearing the same thing expressed in different language seems like it’s generally ok. I don’t know how many philosophical conversations would have been terminated prematurely were it not for one party asking the other to put things in a different way.Report

Oliver Traldi
Oliver Traldi
Reply to  Spencer Jay Case
2 years ago

Yes, absolutely.Report

Smith&Jones
Smith&Jones
2 years ago

Does Case intend concept inflation to be a politically neutral description of a linguistic/rhetorical phenomenon? And is it a thin term or a thick term? Also, maybe it is worth keeping in mind that conservatives have long characterized non-conservative political and social beliefs or lifestyles as “attacks”, ”assaults”, “war” etc. dealing great “harm” to the family, men, the country, the American dream, religion, etc.Report

Spencer Jay Case
Spencer Jay Case
Reply to  Smith&Jones
2 years ago

I discuss “attack” in my reply to Justin below. The same goes for “assault.” Also calling every American soldier a hero is an example of concept inflation. With harm I think we’re talking about substantive disagreement most of the time, which is one reason it wasn’t an example I discussed. I think “war” falls under implicit hyperbole.Report

Smith&Jones
Smith&Jones
Reply to  Spencer Jay Case
2 years ago

Thanks for the reply Spencer.Report

Smith&Jones
Smith&Jones
Reply to  Spencer Jay Case
2 years ago

Also curious about something else: do you think we ever witness a corresponding phenomenon of concept deflation, i.e. where the extension of a thick concept is whittled down? If so, is it as bad of a thing as inflation?
It seems imaginable that we could slowly adopt (and in some cases maybe already have) overly demanding application conditions for ‘wolf’ so that we only cry wolf when we have spotted what is clearly a wolf and that wolf is giant or is too close to the village or has already caused some harm etc.Report

Oliver Traldi
Oliver Traldi
Reply to  Smith&Jones
2 years ago

Some might say that one and the same concept can undergo (attempted) inflation and deflation simultaneously. For instance, it narrows the scope of “racism” if only certain people can perpetrate it or be victimized by it, while it simultaneously broadens the scope of “racism” if everyday comments with no racial content, tastes in music and literature, etc. can count as racist.Report

Smith&Jones
Smith&Jones
Reply to  Oliver Traldi
2 years ago

Yeah. That’s interesting. If so, then we may not be able to define concept inflation in terms of an increase in the number of items falling within a term’s extension. Something like distance from a paradigm or degree of similarity to a prototype could be used instead— which might be how Spencer intends the idea to be understood anyway.Report

Spencer Jay Case
Spencer Jay Case
Reply to  Smith&Jones
2 years ago

I had an interesting discussion about this with Joshua Knobe a few days ago. I can’t think of any examples immediately, but it certainly could be a problem.

For example, if the word “genocide” were too restrictive — say, requiring more than 10 million deaths — then that could be as bad as the term being expanded too much and watered down. Some really, really awful atrocities would not be considered genocides, though they seemingly deserve to. The Armenian genocide wouldn’t even come close. If we set the bar at 100 million, then the term would cease to be practically important altogether because so far no dictator has managed to kill that many people.

Or suppose that we started defining “murder” so that it only referred to the most severe degree of murder. All other killings would be reclassified as some form of manslaughter. I could imagine that this could eventually result in murder seeming even worse to people, intuitively, than it now does. But it would also have the effect of downplaying the badness of many unjustified killings that deserve a very high degree of condemnation. So this would probably be a bad thing overall.Report

Smith&Jones
Smith&Jones
Reply to  Spencer Jay Case
2 years ago

Thanks for this. Yeah, those are good examples of what I’m thinking about. My thought was roughly: yes, false positives are bad and that’s similar to inflation, but false negatives can be bad too and that would be similar to deflation.

And you might actually see some cases like this when organizations and governments have to track something that reflects poorly on them when the counts are too high so they adopt more stringent conditions to save face in the short-term. A made-up example for illustration: crime rates in an area could be manipulated by reclassifying all non-fatal trivial altercations as mere misdemeanors. Then the area looks safer than it really is. That’s a cheap and easy way to cut down on crime.

As for real examples, I feel like the book The Tyranny of Metrics might be a good starting point. (The examples in there probably will not be good cases of organic semantic change though since the semantic shifts involved are often designed in a top-down way or are, well, insincere. )Report

Um
Um
2 years ago

This work by Case strikes me as philosophically unhelpful without some model of conceptual change in hand. Has Case presented (or endorsed) a particular model of conceptual change elsewhere?Report

Preston Stovall
Preston Stovall
2 years ago

Because conceptual change isn’t always a problem, whether the inflation of terms like ‘violence’ is a problem depends on the details.

I’m sure that having more ways of talking about the pain and wrong we think we experience at the hands of one another is a good thing, or will be in the long run. But when we examine the specific ways that terms like ‘violence’ have been inflated, there is reason to be concerned about some of them. A vocal segment of the academy has co-opted this language and put it to use in stifling criticism of its politics by castigating its critics as morally pernicious. That’s a real problem, both internally for the academy and for the people we’re educating.

Consider the case surrounding Rebecca Tuvel’s article in Hypatia. Some of the most outspoken and socially progressive philosophers came down strongly in opposition to the publishing of that article, and they did so using morally loaded language. Probably my favorite inflated use of ‘violence’ came from that debacle: a professor claimed that the publication of Tuvel’s article was an instance of “egregious levels of liberal white ignorance and discursive transmisogynistic violence”. Insofar as this is one kind of meaning ‘violence’ has been inflated to have, we need to start poking holes in that use of the term.

Again, this all turns on the particulars concerning which terms are being inflated and how they’re being used. And in general it’s to the good that we can remark on more of what we feel are injustices today. But we should also be willing to give those around us the leeway to respectfully disagree. There is cause for concern when a group of people who brook no disagreement over some of their ethical shibboleths start using inflated terminology to morally censure their critics.Report

Untenured Postdoc
Untenured Postdoc
2 years ago

“Let’s go back to Case’s fable. … Maybe dogs and coyotes are less dangerous than wolves, but they could pose serious problems for the boys nonetheless, and no one’s paying attention to these problems. They boys have been led to believe that the only way they are going to get help is by crying wolf. …If the other villagers had been listening to the shepherd boys and had been understanding of their problem, the boys would not have had to resort to crying wolf.”

But in the context of the fable, the boys are still getting eaten by the wolf in the end, aren’t they? Maybe that is the Case-Weinberg synthesis: the means by which the boys are calling for help are legitimate (Weinberg) but ultimately self-defeating (Case).

For what it’s worth, I don’t think Case is doing a good enough job at differentiating concept inflation from the usual kind of language change; the latter is supposedly innocuous. A famous example would be the word “awesome” which encompasses a whole lot of things these days than it used to.

But aside from this, there appears to be a tendency of imbuing some words (like various -isms, or violence, or attack, or…) with an “academic” meaning that is then taken to co-exist with the vernacular meaning. While this is not uncommon (scientists are allowed to define their own terms, even if these terms are borrowed from the vernacular), it seems to be disastrous in the political arena, due to frequent equivocations that could be avoided by being a bit more hygienic about language.Report

Gray
Gray
2 years ago

From the OP:

“It may be worth observing that the metaphorical and hyperbolic uses of only certain terms end up coming in for criticism along these lines. Consider the phrase “attack on free speech.” It has been used to describe, among other things, requests to address people politely, criticisms of people’s speech, blog comment moderation, withdrawals of invitations to speak, and the like. These phenomenon do not exemplify the primary definitions of “attack,” which tend to include references to aggression, physical force, injury, even weapons. Yet I don’t recall anyone raising concerns about concept inflation when hearing about “attacks” on free speech. We seem to have been able to cope just fine with whatever loss of “rhetorical effect” the word “attack” has suffered through its metaphorical and hyperbolic use. Why worry about “violence” then?”

I found this paragraph a bit odd. True, the objections some have raised in response to discourse about the “attack on free speech” have not, as far as I know, been put explicitly in terms of concept inflation. But the idea behind their objections, if I understand correctly, is that the concerns about free speech on campus are overblown. And this seems like, if not another way of saying “the label ‘attack on free speech’ is unjustifiably inflated,” then at least something very close. As I was reading the above paragraph, it struck me that you and Case might well agree on the general badness of concept inflation; it coud simply be that you apply the label to different things (he to use of the term “violence,” you to use of the term “attack” [in “attack on free speech”]).Report

Darryl E.
Darryl E.
2 years ago

I’d like a clear account of what “concept inflation” does for us that Sartori’s concept stretching and conceptual travelling don’t. It seems to me Sartori gives us the terminology we need to describe the phenomenon, with the advantage of not moralizing it. If we want to condemn or praise it, we have to actually do the work, not rely on a moralized conception of what’s going on to do it for us. He’s taken a useful way of thinking about conceptual change and made it less useful, but trading analytic clarity for easy judgement. Pass.Report

Spencer Jay Case
Spencer Jay Case
2 years ago

I want to reiterate that this isn’t some unexpected broadside against a younger member of the profession that anyone should consider inappropriate. We talked about this over email, and he has plenty of background knowledge to suggest that I’m not averse to this form of engagement. He even sent me an email saying that he would pull this post if I thought it was unfair or that it would harm me. I don’t. I do strongly disagree, however.

As for “violence,” the description of “a violent cough” and the other examples Justin gives are not concept inflation, nor are they metaphorical. They fit within the dictionary.com definitions of violence. So they are traditional. By contrast it’s unclear where we’re supposed to fit “epistemic violence” and other novel uses of the term. It’s unclear whether this term is metaphorical or literal. I don’t think that anyone coins new politically-salient terms that include the word “violence” with the thought of a violent cough in mind. It’s reasonable to assume that people who don’t know exactly what sense of “violence” is being used will think first of paradigmatic cases of violence, i.e., physical harm. I don’t think this is incidental to the rhetorical intentions of the people using the term.
Justin writes:

I’ve heard the phrase “epistemic violence” hundreds of times. Has this “degraded the rhetorical effect” of the word “violence” for me? I don’t think so… The use of “violence” in these contexts does not seem to render it meaningless when it’s deployed in more traditional ways.

I want to say, first, are you sure that you can reliably introspect about this? Maybe it has subtly reduced your responsiveness to concerns about violence (and you might not always know the context in which the “violence” is supposed to be occurring). I know that if I hear activists of a kind I suspect frequently use concept inflation as a rhetorical tactic warn of “anti-LGBTQ violence on campus” I am not likely to be as alarmed as I might have been several years ago because “violence” has been inflated. Even if you’re confident that your response to the word “violence” is exactly what it’s always been, it might change eventually. Finally, there are other words to consider like “racism” “sexism” and “harassment.” It does seem to me that the responsiveness to allegations involving these words is subdued because their extensions have gradually expanded, due partly to concept inflation.

I think it was Bill Maher who remarked that liberals didn’t have much they could say about Donald Trump that they didn’t say about Mitt Romney a few years earlier. The volume for political discourse had been cranked up to 11 for some time. Saying Trump is *even more* sexist, racist, and terrible doesn’t trigger any alarm bells. Some progressives seem worried that we are literally moving toward Nazism. My advice to them, if they are serious, is not to wear out the epithet “Nazi.” If an actual Nazi like David Duke ran for president in the future, you don’t want people to shrug and say “another day, another Nazi.” You might want to save your loudest alarm for when you really need it.

To reiterate a reductio ad absurdum I employed in the article, I assume that most of my interlocutors would object to parallel uses of the word “genocide” and “rape.” I don’t think many people would have the reaction, “oh, now I know that ‘genocide’ in such-and-such context refers to misgendering; in a different context it means ‘mass murder to destroy an entire people.’ So this seems just fine.” I invite you to reflect on why Rafael Lemkin put so much thought into coining the word “genocide” to refer to the superlatively awful things the Nazis did and that other horrible governments have done. He recognized that a distinct word is closely associated with a distinct mental category. Even if people can intellectually distinguish between different degrees of wrongdoing without having a distinct word, he thought that having a distinct word was important as a kind of moral-emotional-semantic marker.

I suspect that this thought was what lies behind the Biblical prohibition on taking the Lord’s name in vein. If ‘Jehovah’ is a word that gets thrown around indiscriminately, then Jehovah will eventually lose ‘His’ unique place in the hearts and minds of believers. The same is true of ‘genocide.’ If it’s thrown around indiscriminately, it will eventually no longer be “the G-word” and the way we feel about actual genocide will be affected. People might say “oh, I can make distinctions.” But in reality people are affected by language shifts. Actual Nazis know this. There’s an Alt-Right podcast called “The Daily Shoah.” This is probably part of a strategy to dull and dilute people’s appropriate emotional responses to Nazi evil by throwing around a serious word as if it were something casual.

I say that what goes for “genocide” goes for less severe words too, like “violence.”

I’m glad Justin mentioned “attack” to mean “criticize.” One literal meaning of the term is “to blame or abuse violently or bitterly.” And we also say things like “he attacked the project with renewed energy.” So this isn’t necessarily metaphorical. But because there are so many connotations with actual violence – most of the literal dictionary definitions mention violence – I think we’d do better to choose a milder term in most cases. “Weinberg Attacks Case on DN” is the kind of headline a sensationalist editor would use to rile people up about this exchange. I do think this widespread practice of subtle exaggeration by strong word choice undermines people’s trust in the media. So yes, this is bad.

Justin writes that “one is opposed to concept inflation, a way of combating it is by listening to people and taking their concerns seriously.” Of course. But also if you want people to take you seriously long term, then don’t inflate concepts as a short-term strategy. He’s right that, depending on how you fill in the details of my wolf story, inflation might be justified. Sure. The analogy is with lying, and I think everyone but Kant thinks lying is sometimes justified. But I suspect that concept inflation is likely to backfire in the way that lying backfires in Aesop’s original story.

It’s very strange for Justin to worry that calling out concept inflation is a strategy for silencing people. One strategy for silencing people is to label your expression “violence,” thereby terminating the conversation. If you’re trying to express a conservative point of view and someone says your opinion is violence, you can’t expect to avoid the charge by re-wording what you’ve said. The content of your view is presumably what makes if violence. On the other hand, the alleged concept inflator can always reword her position so that she’s no longer using the word “violence” or whatever the inflated term is supposed to be. Then the objection no longer applies.

I agree that we should listen and understand one another and try to take one another seriously. It’s because I think concept inflation makes it harder for us to do this that I object to it so strenuously.Report

Joshua Blanchard
Reply to  Spencer Jay Case
2 years ago

Hi Spencer. I thought you and others might be interested in a real debate that happened just a couple of years ago regarding the word “genocide,” which you mention here and in your article. Here’s an interesting write-up by Emma Green in The Atlantic:
https://www.theatlantic.com/politics/archive/2016/08/why-did-black-american-activists-start-caring-about-palestine/496088/Report

Edward Teach
Edward Teach
2 years ago

“I was reminded of all of the ways in which we already use “violence” metaphorically… No one objects to using “violence” in these descriptions”

I think there’s a difference between these cases and the kinds that Case has in mind. The concept creep (inflation) occurring in these cases is much more metaphorical, and would be endorsed as such by most reasonable viewers. When the media says there is an attack on free speech, sure they mean we should defend this (defend seems to be another concept which has been successfully expanded in common usage), but they’re not calling for the military to get involved. But in the humanities, this creep used in such a way to argue that we ought to have a very similar response to the original concept. People routinely argue that certain movie tropes or casting choices are ‘harmful’ and therefore eligible for censorship. Pornography ‘silences’ and therefore is an infringement of women’s first amendment rights. Minorities are justified in literal physical violence against x because x is oppressive.Report

J.S.
J.S.
2 years ago

I think the notion of concept inflation needs to be precisified a bit before it hope to do the work Case wants it to do. Like others have mentioned, concept inflation doesn’t seem to differentiate between concepts that have semantically expanded in a way that is unproblematic (it really is prescriptivist pedantry to object, on grounds of clarity, to ‘literally’ being used as an intensifier in ordinary conversation) in contrast to concepts that have semantically expanded in a non-controversial sense of ‘problematic’. A more neutral phrase might be ‘concept expansion’. Legitimate concept expansion might be one that preserves enough similarities between the uncontroversial and controversial cases such that they can be understood using similar mechanisms, effects, or research paradigms. An illegitimate concept expansion is one that doesn’t preserve enough similarities to be usefully understood/studied as the same kind of phenomenon. I think ‘epistemic murder’ would be an obvious case of an illegitimate concept expansion: whatever epistemic murder might be, it likely won’t share very many of murder’s paradigm features, mechanisms, and effects. Anyone using ‘epistemic murder’ in the course of an argument would be abusing language (and their interlocutor) by using that term, unless a very persuasive accounting of similarities was supplied.

By this token, if the definition of ‘violence’ has expanded to include psychological violence and social violence and we take that expansion as unproblematic (and it is now commonly accepted that violence can be psychological), then whether or not epistemic violence ought to be considered violence is predicated on whether or not it shares enough features with paradigm cases of violence. Prima facie, the definition of ‘violence’ expanded to include certain forms of psychological maltreatment (e.g. abuse) not because it was rhetorically useful to do so, but rather because they share at least some cluster of similarities (severe emotional harm, drastically curtailed quality of life, etc.) with physical violence such that it can be usefully studied as an example of violence. If this can be proved of epistemic violence, then it would also qualify as violence. I’m not sure that it does, however, but it’s ultimately an empirical question, and one that is not ridiculous on its face. As it stands, these questions of legitimate vs. illegitimate concept inflation have to be judged on a case by case basis–recommending that we encase our language in amber for clarity’s sake is a cure worse than the disease.

That said, appealing to the polysemy of language isn’t an out for the defender of morally/politically-loaded concept inflation here. An “attack on free speech” isn’t meant to convey the sense of a physical attack–what one really means is a vituperative (and perhaps dangerous) verbal critique, which is really an alternate meaning of the word ‘attack’. Contrastingly, it seems to me that when people are talking about real-life concept expansion (i.e. not merely philosophical scenarios about crying wolf) they really mean (if they have thought at all about the issue and are making the claim in good faith) for, say, epistemic violence to be a *kind* of violence, not an alternate usage of the term like in the ‘attack’ case.

I I think the real danger here isn’t muddying the waters of clarity; it’s rather that these politically and morally loaded concept expansions can be possibly used as argument-enders by censuring those who disagree as morally repugnant. This should be avoided. Even if a case can be made for a plausible expansion of the concept of, say, violence to include epistemic violence, it’s still an open question–not at all uncontroversial–and those who object to said concept expansion shouldn’t be censured if they are objecting seriously and in good faith. And it seems like, at least in a few cases, this is exactly what happens (at least online, where our worst selves often roam free). This also cuts the other way, however: We shouldn’t take rigorous attempts to expand a concept into political or social territory as “political correctness run amok”. In the end what matters is the argument for or against a particular concept expansion, and that argument should be assessed strictly on its merits rather than on its perceived political affiliation.Report

JTD
JTD
Reply to  J.S.
2 years ago

I strongly concur with your point that those engaging in this debate should be using a neutral term like “concept expansion” and then debating on a case by case basis whether each particular concept expansion is helpful, harmful, or neither. I think the recent history of a similar debate in psychology is instructive here. One academic coined the term “concept creep” to start a debate about a particular phenomena. Another academic, who saw this phenomena in a more positive light, wrote a reply where they insisted on using the term “concept enhancement” instead. Both these terms were no doubt successful in attracting allies and riling up opponents. However, intellectually both were being irresponsible by using unfairly emotive terms in a debate that would be most productively conducted from a neutral starting point agreeable to all sides.Report

ajkreider
ajkreider
2 years ago

Though I’m not a fan of the methodology in the book generally, in “Down Girl”, Manne at least makes a conscious and reasoned case for expanding concepts (misogyny) beyond their dictionary definition or ordinary use.

She describes her activity as part of a broader feminist project of “semantic activism”. But at least she’s up front about what she’s doing (and doing it on a philosophically principled basis), so the charge of lying doesn’t apply. Manne doesn’t claim that all of the benefits are normative – some are descriptive. She also takes herself to be not alone in this project (Haslanger, etc.).Report

Oliver Traldi
Oliver Traldi
Reply to  ajkreider
2 years ago

I think if you think what ameliorative analysts and semantic activists are doing is lying, then the fact that they broadcast their strategy won’t help matters. Imagine a politician says on Monday, “In my speech tomorrow I’m just going to make some stuff up.” Then on Tuesday, in their speech, they just make some stuff up. This would remain lying; it wouldn’t make them a “make-stuff-up-ist” who is somehow immune to the charge of lying.Report

grymes
grymes
Reply to  Oliver Traldi
2 years ago

Yes it does: it makes them a storyteller, and storytellers who make it clear that they’re making stuff up are indeed somehow immune to the charge of lying. Fiction ain’t lies.

The politician would surely be blameworthy for this stunt, but not because they’re a liar. (Indeed, they’re being weirdly honest for a politician.) Whether the ameliorative metaphysician is similarly blameworthy depends on the goals and effects of ameliorative metaphysics, just as the politician’s blameworthiness depends on her goals and the effects of her storytelling.Report

YAAGS
YAAGS
2 years ago

Consider the following archaic definition of racism:

(Archaic) necessarily, x is racist iff (1) x possesses a sufficient number of attitudes towards a racial group* that are grounded in avoidable ignorance (where emotions and beliefs are taken to be attitudes), or (2) x is an act or object such that any reasonable person would interpret x as being explained by or expressing the relevant attitudes of an agent meeting condition (1).

This, or something close to it, is the traditional notion that most people have in mind when they say someone or something is racist. In contrast, ‘racist’ as it is used in the linguistic community that Case is complaining about means something entirely different. The notion of racism in this community has nothing in particular to do with psychological states. In particular, according to this community:

(~N) For x to be racist it is not necessary that x involve any attitudes, beliefs, or psychological states.
(~S) For x to be racist it is not sufficient that x involve any attitudes, beliefs, or psychological states.

To see why (~N) is true, consider the fact that the community takes things like government programs, economic policies, and algorithms to be racist simply in terms of their outcomes or how they function, regardless of whether any of the people who design or use them have any racist attitudes. To see why (~S) is true, consider the fact that the community is committed to the claim that African-Americans are incapable of being racist, including any African-American individuals that believe a large number of stereotypes concerning and possess seething hatred for other racial groups based in ignorance. This is because the new definition takes racism to be fully grounded in broadly social, economic, and historical facts. At a rough first approximation:

(Racism*) necessarily, x is racist iff (1) x occupies a position in a social structure that unjustly benefits those of x’s racial group above all others, or (2) x is an act or object such that is brought about by a social structure that unjustly benefits one racial group over another and x has an emotional/economic/physical cost for one of the unjustly affected racial groups.

However, (Racism*) is not quite right. The problem is that there are nested social structures. For instance, suppose an African-American family adopted a white child and systematically gave the white child fewer benefits than their other African-American children (e.g., less food, less attention, handed down clothing, demanded they do chores instead of doing their homework so that their grades suffer, etc.). A family is a social structure if anything is. But the African-Americans parents cannot be racist due to the fact that their family exists within an even larger social structure that is racist against African-Americans. Similarly, even if an American university in admissions policies and other practices (e.g., awarding of funding, offices, etc.) disfavored whites in virtually every respect in a way that could not even be justified by appeals to diversity, the university still cannot be racist because it exists within the larger American social structure which unjustly benefits whites. In other words, there is a nested hierarchy of social structures and it is clearly only the most dominant structure in this hierarchy that ever matters for the purposes of determining what is racist. Thus the real definition must run as follows:

(Racism) necessarily, x is racist iff (1) the most dominant social structure unjustly benefits those of x’s racial group above all others, or (2) … mutatis mutandis.

Since the most dominant social structure is simply the global economic/political system, and this system unjustly benefits whites, this gives us the correct result insofar as it implies that only whites can be racist. For instance, even though many Han Chinese in power in China may consider themselves as racially superior to other ethnic groups and thus engage in oppressive practices towards those groups (e.g., the cultural genocide of Tibet), they still cannot be racist because the overall global political/economic system benefits whites. The entire political world order would have to change in order for any Chinese person to be racist. Again, this is the correct result.

The benefits of the new definition are clear. First, it preserves the intuition that all white people are racist regardless of what they say or what attitudes they have or their socio-economic condition. Recall that it only requires that the most dominant social structure benefit x’s group in general, not that the structure benefit x. The latter would make it difficult to establish that hillbilly whites living off the grid in Appalachia are racist, simply because they are destitute and receive few benefits from larger society. This would clearly be intuitively incorrect. One could perhaps argue that they receive negative benefits insofar as the government generally doesn’t interfere with them because they are white. However, this would hold their obvious racism hostage to empirical fortune, since one might argue that the government doesn’t bother to interfere with them because they live off the grid in the middle of nowhere. Second, it allows one to correctly call out whites as racist whenever they do anything that negatively affects minorities, regardless of their supposed attitudes or intentions.

The new definition might also appear to have some costs. First of all, many people outside the relevant linguistic sub-community still subscribe to something like (Archaic). Thus when those in the linguistic sub-community point out that something is racist regardless of the attitudes involved this often leads to a terminological dispute. This can be somewhat counter-productive. Instead of discussing policy proposals like bodycameras and educational reform, many political discussions tend to get side-tracked by semantic disputes over the meaning of ‘racist’. However, this behavior is a form of racism, since it negatively affects minorities to not talk about these things. Since there is no obligation to accommodate racism, this is not really a cost after all.

Another potential downside is that when whites realize the new definition they might simply shrug. After all, according to (Racism) they are racist simply in virtue of being white and existing within the current global political and economic order. There is nothing whites can do the change the former (contra Tuvel), and very little they can do to change the latter. Consider the following moral principle:

(Don’t Be Racist) One is obligated to do everything in one’s power to not be racist.

(Don’t Be Racist) sounds extremely plausible. Given (Racism), this would imply that whites should give up most of their wealth and devote most of their lives to overthrowing the existing global social order (e.g., white professors should willingly give up their positions if there are any minority candidates and become baristas). Again, this seems like the correct result.
Unfortunately, whites might argue that (Don’t Be Racist) is only intuitively plausible when it is understood in terms of the archaic definition of racism, since one’s actions and attitudes are usually well within one’s powers to change and it usually doesn’t cost much to change them. But once again, this response is racist since it obviously negatively affects minorities. Since there is no obligation to accommodate racism, this is not really a cost after all.

*Here and throughout I use ‘racial group’ in a way that is compatible with them being socially constructed and without presupposing that they are biologically real.Report

JTD
JTD
Reply to  YAAGS
2 years ago

Thanks for this clear and informative comment. I have a few observations and thoughts.

First, what you describe here is not a case of “concept enhancement” (I prefer this neutral term to Case’s loaded alternative), but a case of “concept replacement”. The new definition does not only apply the term “racism” to a new phenomena, it also denies that it applies to many phenomena that the old term applies to (e.g., An African-American directing racial abuse at another minority, or a non-white ethnic majority that uses an ideology of racial supremacy to justify unjust discrimination against a non-white ethnic minority). Overall, these changes look radical enough to constitute a new concept (although I acknowledge that some theories of conceptual change will disagree).

Second, the central disagreement between those who favor the new definition and those who favor the old appears to be the following. All parties agree that “racism” should pick out a phenomena involving a type of injustice connected to people being racialized a certain way. All parties also agree that cases where a ‘white’ person unjustly discriminates against a ‘black’ person because of their race, or expresses negative attitudes about a ‘black’ person because of their race are cases of racism (let’s call this the “central case”). What the parties disagree about is what is the most salient, or morally useful, generalization from the central case. One side thinks it is the general phenomena of someone from one racial category doing the things described in the central case to someone from another racial category. The other side thinks that it is the general phenomena of someone belonging to a racial category that is unjustly benefited above all others by the most dominant social structure. Speaking for myself, it is not clear to me who is correct in this debate. There are many consideration that speak for or against the salience and moral usefulness of the two rival concepts. Furthermore, I find that those who think the case for their side is obvious and overwhelming tend to not fully confront some of the problems it has and to uncharitably present the case for the other side.

Third, even if the reformists are correct about what the most salient, or morally useful generalization is, it does seem to me that they are going to lose the cultural battle to have their version of the concept widely adopted. When one looks beyond the in-groups of the intellectual elites in several English-speaking Western countries one finds that the old concept of racism (which you misleading describe as archaic) is absolutely dominant. I doubt, for the following reasons, that the reformers will have much success in converting a critical mass of those who use to the old concept to adopt the reform.

(1) In large parts of the world an ideology of racial supremacy is used by a non-white ethnic majority to suppress non-white ethnic minorities. People from these minorities generally find the explaining this dynamic as “racism” is a powerful tool understanding their situation and fighting for justice. In my experience, they do not respond well when they are told that the US intellectual elite has a new understanding of “racism” which does not count their struggle as involving any of it.

(2) Many migrants from non-white ethnic minorities in countries like the US and UK experience racial abuse not only from the white majority but also from other non-white minorities. Most of these migrants are not part of the intellectual elite and most will insist that the racial abuse they get from other minorities is a form of racism.

(3) Consider the white majorities in countries like the US and UK. Even if we think it is their own short-sightedness, the fact of the matter is that most of them will refuse to accept that they are racist merely for being ‘white’ and will instead insist on the old definition.

(4) Consider an alternative strategy that many who share the concerns of the reformists are using to try to make progress on racial injustice. This strategy accepts the old definition of racism but then introduces new concepts such as “systematic racism” and “institutional racism” to supplement the basic concept. It is then explained that although every instance of basic racism is wrong on some level, the broader context within which the basic instance occurs makes a massive difference to whether that instance is a minor wrong or a major form of injustice. When the perpetrator is part of a social structure that unjustly benefits people from their racial category then the injustice is more significant. And, the higher that social structure is in the social hierarchy, the worse the injustice. Finally, it can also be said that supporting, endorsing, or upholding social structures that unjustly benefit people from your racial category is itself a form of “racism”, or at least a form of enabling “racism” which one bears moral responsibility for. This strategy addresses many of the concerns of the reformists and yet, is far more likely than the reformist strategy to win support from people in categories (1), (2) and (3).

So, it seems to me that (4) is the winning strategy as far as making progress on racial injustice goes. Finally, if this is correct then it also seems that the reformist strategy is actually counterproductive. When people in categories (1), (2), and (3), are confronted with the reformist proposal they are generally strongly resistant to it, feeling that it undermines their own personal experiences of injustice and treats them unfairly. This feeling often results in them becoming close-minded to intellectual elites telling them that they don’t fully understand what racism is and thus makes it harder for (4) to succeed with them when it otherwise might have.Report

JTD
JTD
Reply to  JTD
2 years ago

In the second paragraph above I obviously meant to say “concept expansion” and not the loaded “concept enhancement”.Report

YAAGS
YAAGS
Reply to  JTD
2 years ago

I generally agree with all of your points. I intended the comment to be a parody exhibiting how an extreme form of concept-creep (inflation, change, co-optation, what have you) can be obviously counterproductive, which you were then nice enough to spell out. I probably should have indicated that it was a parody, but I was curious whether anyone would actually take it seriously. (I thought the bit about white professors becoming baristas was sure to give it away, but Poe’s Law strikes again!) Sorry! That being said, while you can think of it as a straw man it’s also not clear to me how the people who use the term ‘racist’ in the novel ways mentioned can avoid something like the absurd view I present without changing at least some of what they say.Report

Michael Leza
Michael Leza
Reply to  YAAGS
2 years ago

The barista bit was funny, and your comment was perfect baby-eating level satire.Report

Michael Leza
Michael Leza
Reply to  YAAGS
2 years ago

This is a great example, thanks for posting it. I’ve been using the phrases “borrowing connotation” or “stealing connotation” to describe the behavior you satirize, depending on how extreme the behavior is, and I’m not sure that I like concept inflation better, but this really illustrates the problem being discussed in a detailed way. This kind of language is designed to invoke unearned emotions and to confuse bystanders into agreeing simply so they can avoid being associated with a word that their brain mostly translates simply as “bad.” It is dishonest and manipulative and it undermines real conversation about these issues (and thus actions to correct the problems).Report

Tom Hurka
Tom Hurka
2 years ago

Chapter 2 of Tony Coady’s book Morality and Political Violence (2008) contains an extended discussion of various “wide” uses of the word “violence,” e.g. in “structural violence,” which he generally argues against.Report

Tom Hurka
Tom Hurka
2 years ago

“To label an instance of speech “virtue signaling” or “moral grandstanding” is to accuse it of being made in self-serving bad faith. … These are all labels that tell us to dismiss what is being said, rather than try to understand it or engage with its substance.”

But some statements *are* made in self-serving bad faith and deserve to be dismissed. Is the OP suggesting that we should engage with and try to understand the substance of “Make America great again”?

“That’s bullshit” also tells us to dismiss what’s being said, but I thought we philosophers were grateful to Harry Frankfurt for explaining what bullshit is, so people can identify it and not be impressed by or waste time on it.

It’s of course wrong to dismiss something that shouldn’t be dismissed, but it’s also wrong to take seriously, and not dismiss, what deserves to be dismissed. Is the point of the OP that the first mistake is more common than the second? I would need some evidence for that. My sense is that there’s a lot of virtue-signalling/grandstanding/bullshit around these days.Report

Tom Hurka
Tom Hurka
Reply to  Justin Weinberg
2 years ago

Why assume that charges of e.g. virtue-signalling or grandstanding are always or typically hasty? There can be good reasons for making them, e.g. in the over-heated way a person speaks or, especially, if he gives no serious evidence for what he says and shows no real interest in finding any. That some of those who joined in the vitriolic criticism of Rebecca Tuvel’s article *hadn’t even read it* was at least some basis for suspecting that they were grandstanding. And what made that so — that they showed so little interest in evidence for what they were saying — is also a basis for dismissing what they said.

And doesn’t your worry about how our assessment of someone’s motives can depend on whether or not we agree with or identify with them apply just as much to the terminology you’re defending? Dotson’s definition of epistemic violence, quoted by Case, says it rests on “pernicious ignorance.” And our readiness to ascribe that to a speaker won’t be affected by our agreeing with or identifying with them but will be entirely dispassionate???

As so often, what’s sauce for the right-wing goose is sauce for the left-wing gander.Report

Oliver Traldi
Oliver Traldi
Reply to  Justin Weinberg
2 years ago

But, Dr. Weinberg, whether or not “motive” is at stake, don’t you agree that if A says P and B says “For you to say P is epistemic violence,” they have refused to engage with their interlocutor in precisely the way that you’re criticizing in your original post?Report

Ned Hall
Ned Hall
Reply to  Oliver Traldi
2 years ago

This is a good point, Oliver. At the very least, if Justin is right in saying that “the assertion that U is an instance of epistemic violence is shorthand for several claims about U” – namely, the claims he lists in his response to you – then we should expect that sincere, thoughtful assertions that something is an instance of epistemic violence should be very rare. (Just think how hard it is to be justified in believing (1) – (4).) I took your point to be that an accusation of “epistemic violence” can very easily be used as – and will typically be understood as – a conversation-stopper, in the way you indicate. Justin’s response just seems not to engage with that point.Report

Oliver Traldi
Oliver Traldi
Reply to  Oliver Traldi
2 years ago

Thank you, Dr. Hall. Dr. Weinberg, which of those seven points involve taking the assertion seriously – that is, trying to weigh evidence in favor of it versus evidence against it? Which of them is, as you put it, “try[ing] to understand it and engage with its substance”?

You say that the claims about the assertion that make up the accusation of epistemic violence are all “legitimately contestable by appeal to evidence that is in principle accessible to relevant parties”. But when you complained about moral grandstanding, concept inflation, etc., you didn’t complain that those accusations weren’t legitimately contestable by appeal to evidence that is in principle accessible to relevant parties. You complained that they were ways of redirecting the conversation away from the content of an interlocutor’s assertion. As far as I can tell, you’ve made my point for me.Report

Smith&Jones
Smith&Jones
Reply to  Oliver Traldi
2 years ago

I doubt the conversation-stopping potential of these terms is their most objectionable feature.

It seems like any phrase or term that is used to ascribe some negative property to an utterance has the potential to, in practice, be used as a conversation-stopper if the relevant audience finds the property to be sufficiently objectionable. Compare the use of ‘sloppy’. I hear people call essays and books sloppy fairly frequently and they often do so without much further ado. I often trust the speaker’s judgment and just proceed to ignore the supposedly sloppy work.

So ‘sloppy’ also seems to be deployed as a conversation-stopper and with some success. But, all the same, some things can be legitimately called ‘sloppy’ because they are demonstrably sloppy and that can be a serious flaw.Report

Oliver Traldi
Oliver Traldi
Reply to  Justin Weinberg
2 years ago

But Dr. Weinberg, people who write about moral grandstanding believe that it is both (a) an instance of a fairly common phenomenon owed to certain predictable patterns and (b) harmful. So what distinction are you trying to draw here?Report

Avalonian
Reply to  Justin Weinberg
2 years ago

Justin, while this is a super helpful conversation and thanks for hosting it, I did not find this reply to Oliver and Ned convincing. We can all agree that if people said the thing to each other that you quote in your first paragraph, then that would be great. But they don’t tend to. If your paraphrase is so accurate, then why do people use terms like “epistemic violence” instead? Well, because, as Spencer says, the pragmatics of the word “violence” are politically useful, they draw implicit associations to paradigm cases of interpersonal or political violence, which are not mere harms (as the last part of your paraphrase implies), but intentionally enacted harms. (Consider: “I’m so sorry I’ve been so violent towards you, it was unintentional”… doesn’t sound right, does it? Sounds like I’m denying that I’ve been violent.) And references to intentionally enacted harms do often function as conversation-stoppers, because they identify an interlocutor as a bad person, someone who intentionally and knowingly harms others, someone we need to objectively manage rather than talk to (in Strawson’s sense).

I tend to think that this is what is wrong with so much concept inflation; it’s not that current linguistic practice must be treated as sacrosanct, it’s that interlocutors are (knowingly) pragmatically implying things that are either outright false or deeply contestable: Dotson’s own definition of “epistemic violence” allows that in lots of cases, harming via a failure to listen will be both unintentional and due to excusable ignorance (“Tracking Epistemic Violence” 238). But this just means that actual uses of such terms will often imply pernicious and harmful falsehoods of their own.Report

Matt
Reply to  Avalonian
2 years ago

Replying to Justin above (can’t seem to do it directly there):

…and this phenomenon is harmful
This seems to me to be a bit of the argument where a lot of confusion comes in, because of the tendency to conflate harms and wrongs. To harm someone is, most basically, to set back their interests. Now, the process you describe above might well set back someone’s interests. But: 1) you can set back a person’s interests without that being, in any reasonable way, “violent”, and you can harm someone without wronging them. (If I run faster than you in a race, and therefore get the prize, I have set back your interests. Or, if I open up a shop next to yours and take away some of your business, I have set back your interests, and in both cases I have harmed you, but have not wronged you.) (You may also wrong people w/o harming them, but that’s less important here.) But _violently_ harming people _typically_ involves wronging them. (The exceptions are usually set interactions, like contact sports.) So, by invoking “violence” here, there is a move to claim that the person is not only setting back the other’s interests, but also wronging them. That part, though, seems in these cases to often be left undefended, or at least significantly under-defended, by those using terms like “epistemic violence” and the like, and to often be a rhetorical slight of hand, designed to slip from harm/set back to interest to wronging.

(Because I’ clumsy, I accidentally hit “report” instead of “reply” at first on Avalonian’s comment above. Please ignore that.)Report

Skef
Skef
2 years ago

“it really is prescriptivist pedantry to object, on grounds of clarity, to ‘literally’ being used as an intensifier in ordinary conversation”

I think this is too simple.

If the issue was just about the drift in this *individual* concept, when others could effectively play the same role, then I agree the objection would be just pedantic. However, it appears that a) whatever concept is typically used for this purpose (at least in English) at a given time suffers this drift, b) the rate of the drift is accelerating, and c) there aren’t many candidates left.

Gathering definitive evidence for these claims is “beyond the scope” but English is littered with the past examples. As concepts go, “real” should do the job but “really” is now such an intensifier that the “real” in it only comes out when squinting. “Actually” is better, but not much. “Truly”? Nope.

This is an interesting phenomenon because it is evidence that some drift is closely tied to semantic use. If forced to propose a theory I would guess that the meaning is altered by the combination of a) the process of language acquisition by hearing the use of words and b) whatever metaphors are “dying” at the time.

However, the explanation is separate from the problem, which is that it really can be important to distinguish that something is happening non-metaphorically, non-figuratively, etc., and the supply of concepts that can do so is ebbing. (This is more easily — and metaphorically — seen in terms of Kenny’s economic analysis.)Report

Skef
Skef
2 years ago

(How could I have left out “very”?)Report

Oliver Traldi
Oliver Traldi
2 years ago

Do the anti-prescriptivists here similarly think it would be pedantic or traditionalistic to object to someone calling affirmative action “racist”? What about if someone said that taxing people to pay for Obamacare is “violence”?Report

Philodorus
Philodorus
Reply to  Oliver Traldi
2 years ago

Hi Oliver,

I’m an anti-prescriptivist, in that I think Spencer’s claims like “we all have the responsibility to be good stewards of the languages we speak” or “when we allow sloppy language to proliferate—for example, when we use the word “literally” to mean “metaphorically”—we degrade language and make it harder for everyone to communicate” are just false. Nobody has to be a steward of the language they speak, because speakers of the language will devise ways to communicate the things they would like to communicate. There is no risk of our language getting “degraded” and it becoming harder for us to communicate. I’m pretty sure this is just a consensus view among linguists.

That doesn’t mean it would be “pedantic” or “traditionalistic” to object to people calling affirmative action racist. The reason to object to that labeling is not that it will eventually dilute the meaning of the term “racist,” so that we have no effective way to communicate our thoughts about racists. Rather, it’s that affirmative action is not racist. Or it’s that the connotations and normative significance of “racist” ought not be associated with affirmative action.

So I think this part of Spencer’s criticism is sort of getting things backwards. It’s not that people like Fricker and Dotson are going to dilute the meaning of every negative-sounding word they keep sticking after “epistemic” — speakers will find ways to communicate their thoughts about oppression, violence, injustice, and the like whatever philosophers do. It’s just that they say false things and are trying to stir up emotions and reactions that they are not in fact entitled to (as I think Spencer also says). They aren’t going to make ordinary language weaker and less effective with their neologisms; rather, they’re just trying to draw on the power of certain words that they are not entitled to.Report

Oliver Traldi
Oliver Traldi
Reply to  Philodorus
2 years ago

Ok, thank you – that’s very well put.Report

Urstoff
Urstoff
Reply to  Oliver Traldi
2 years ago

“socialist” would also make for a similar case study.Report

history
history
2 years ago

You might well know this, Spencer, but Larry Blum (2002, 2003) used “conceptual inflation” when discussing analogous issues arising from uses of “racist”. The following debate involving philosophers such as Tommie Shelby is also relevant and covers ground roughly charted in the YAAGS comment above. Just thought you might want to cite Blum.

Blum: https://philpapers.org/rec/BLURWIReport

Justin Kalef
Justin Kalef
2 years ago

I can only speak about my own experience with these words. Here goes.

When I was young and first heard about racism, I associated it with people of very bad character. But the word ‘Nazi’ was even more extreme. I had my bar mitzvah at a synagogue that was burned down when someone — I don’t think we ever knew who it was — threw a firebomb in the windowlate one night. It was terrifying. I remember the day our rabbi walked us through the charred remains of the place where we had once worshiped, and told us that this was evidence that racism and neo-Nazism were alive and well, decades after the end of the Nazi holocaust. I could never hear the word ‘Nazi’ without a chill running down my spine.

But ‘white supremacist’ was even worse. I was aware that there were bands of racist skinheads going around beating people up. I was utterly terrified that I would run into them someday. The phrase ‘white supremacist’ was reserved for the most violent and extreme sorts of racists one could imagine. One did not use that word lightly, and it never failed to have an effect.

Nowadays, when I hear that someone is a ‘white supremacist’, I wonder whether it’s some vicious criminal who walks around like the characters in Romper Stomper, beating racial minorities to death, or just someone who teaches an insufficiently diverse curriculum or something like that. I never thought the day would come when the phrase didn’t immediately generate a visceral reaction, but it has.Report

ag
ag
Reply to  Justin Kalef
2 years ago

I think this says a lot, but what do you take the upshot to be? I’m guessing it’s that these words have been so weakened that we should all be concerned, i.e. that this has made it harder to recognize the ‘really bad’ people and actions. Another upshot someone might see in what you say though is that the varieties and mechanisms of racism are far more subtle and systemic than we used to recognize (or than many recognize to this day) and that a great advance in the understanding of injustice has been to see how those horrible racist events and people you have described are continuous with much more small-scale often-ignored daily happenings, and that to root out the larger ones you must recognize and address the smaller ones.

Part of the means for making plain this continuity would obviously be to use the words that apply to the more apparent horrors when pointing out the smaller-scale ones. And so ‘inflation’ of this kind would really be a mark of the successful evolution of our understanding of the continuity of injustice, from the small-scale to the large-scale.

Now perhaps you and many others here don’t think that injustice is continuous in this way, and don’t think the growing belief that it is represents an advance in the understanding of injustice. That is fine. But then the disagreement is a substantive one, having to do with the way the world really is, rather than one about whether concept inflation in and of itself is bad. In other words, just showing that ‘white supremacist’ does not mean what it used to, and is used for much less apparently harmful cases, does not and cannot in and of itself show that this change is bad. So why debate whether it is, instead of debating the more important point of how best to understand injustice? And why go treating those you disagree with like ‘people crying wolf’ instead of like people with substantive disagreements with you about the nature of injustice? (I mean to ask all this in good faith, not to rhetorically make a point.)Report

Justin Kalef
Justin Kalef
Reply to  ag
2 years ago

Hi, ag.

You suggest, as I understand it, that the radical expansion of the term ‘white supremacist’ to mean not just violent Nazi skinheads but also liberal-minded people who oppose racism, etc. might be good because it could lead us to think of the ways that all the minuscule things that fit into the huge category form the causal basis of the things that fit into the old category. But it’s far from clear to me that that’s true. It seems at least as likely that they don’t and that picking away at people for slight or nonexistent violations of some code and using one of these terms to refer to them annoys those people and alienates them from taking the problem as seriously as they once did.

One thing’s for sure, though: I no longer have any clear way of knowing, when I hear that someone is a white supremacist, what that’s supposed to mean, unless I already know something about the speaker. And it’s getting hard even to articulate what the word originally does mean. I’d like to say that a white supremacist is someone who is committed to the principles of pro-white racism and is prepared to use or advocate violence to further the ends of a racial contest. But even that’s less clear now because the words ‘racism’, ‘violent’, etc. have been stretched far beyond what most people mean by them.

We need words that make clear what we oppose, and that don’t just send the volume of every little thing up to the top of the register. If we don’t, then we’ll have no way of distinguishing things we need to distinguish.Report

ag
ag
Reply to  Justin Kalef
2 years ago

Thank you for the response. But of course, to call them “slight or nonexistent violations of some code” is just to deny that they are bound up with the most serious harms of racism in the ways proposed. Maybe you are right to deny this, but again that is just a substantive disagreement about the nature of injustice, not a problem merely of people misusing words.

And notice that a person can’t help but be aware of the view that small-scale racists acts are continuous with large-scale ones these days. Even those whose new-found awareness of this view find it “alienating” still have been made aware of it, and need to think about it because of how it has been pushed into the national consciousness. Perhaps the risks of alienating potential allies are not worth this greater level of awareness. Perhaps the loss of certainty that when someone is called a ‘white supremacist’ they are overtly, indisputably evil is not worth this greater awareness. I don’t know the answers to these questions.

My point is simple though: it seems like many people here bemoan these losses with such force precisely because they disagree on the substantive issues about the nature of injustice. Yet they make it out to be a problem about language use that can be addressed in a way that is more or less neutral with respect to different understandings of injustice. So they hide the core of the issue. The issue is not so much these losses, as that you think what are considered by some a deep-rooted part of injustice to be “slight or nonexistent violations of some code”. And so why not debate them about injustice, rather than language?

Another way to put the point is that the people who are expanding this vocabulary themselves think they have words that “make clear what they oppose”. And the reason you don’t agree is less because you don’t like the way they use the words, and more because you don’t want to oppose the same things as them. But then why not just debate whether you should oppose those or not, i.e. about the nature of injustice?Report

Oliver Traldi
Oliver Traldi
Reply to  ag
2 years ago

This is a really weird comment, ag. Philosophers make substantive claims about “ameliorative analysis”, “semantic activism”, “conceptual engineering”, etc. There’s no reason those should be somehow off the table. Further, even if there are disputes here that are partially verbal (which seems to be what you’re suggesting), I’ve never seen anybody suggest that the right way to treat a partially verbal dispute is simply to let the other side of the dispute determine what words to use. Rather, the right strategy is to find language that both sides find neutral and then see if the dispute remains. If it doesn’t, then the dispute was merely verbal. If it does, then the dispute should be conducted in that neutral language. It’s hard to actually know whether there even *is* a dispute about injustice until the linguistic issue is resolved.Report

Chris
Chris
Reply to  Oliver Traldi
2 years ago

I disagree with one of your points here: Another way to deal with verbal disputes (and to find out whether they’re merely verbal) is to grant the other side their language, and reformulate your argument using their terminology. I thought this was often standard philosophical practice? An example: in Marquis famous paper “Why Abortion is immoral” he uses the terms “pro-choice” to describe the opposition and avoids using “pro-life” for his own position. That is, he grants the opponent their word choice, so that the focus will be elsewhere (in this case, on his account of why killing is wrong).

This is also (I thought) a common (though perhaps not the most common) maneuver in metaphysics, etc. when trying to avoid quibbling over terminology – I can think of other examples, too (in the traditional problem of induction when Black claims that the inductive justification of induction isn’t circular, you could quibble about his overly narrow use of “circular” or – you could do what e.g, Skyrms and Salmon do and just grant him his use of “circular” and still show there’s a problem. It is a common (and often rhetorically effective, I think) technique in philosophy.Report

YAAGS
YAAGS
Reply to  Oliver Traldi
2 years ago

I’m with Oliver. If the dispute is substantive then it can be cashed out in any number of ways. The problem with a lot of this stuff is precisely that it creates verbal disputes where there would otherwise be either agreement or an entirely different dialectic. You can talk about differential tax and funding systems for school districtricts as leading to unjust inequalities where poor predominantly African-American communities have access to only bad education because they can’t afford to live in the rich school districts. It’s hard for conservatives to defend that sort of thing on libertarian grounds because it affects equality of opportunity. Or you could say that white people who vote for those tax systems are white supremacists and become embroiled in a verbal dispute because conservatives think you have to actually hate minorities and believe in white supremacist ideology to be a white supremacist. This will completely derail the conversation and the left will come out looking like a bunch of loonies and ideological extremists to anyone outside academia. That’s why people like Tucker Carlson absolutely love critical race studies people. In fact, the first thing conservatives do when you confront them with the first sort of argument is accuse you of accusing ordinary middle-class white people of all being racist, precisely because they know that this is probably the only way for them to win rhetorically. If the inflationists are helping anyone, it’s conservatives.Report

Justin Kalef
Justin Kalef
Reply to  ag
2 years ago

Many utterances are important because they press us to have a certain reaction when they are used. They function as alarm bells. “Wolf!” is one such utterance. Another is “That person is a racist!”

Perhaps my concerns, and I think those of others, could be more clearly expressed if we consider those words as actual alarm bells, and think about how their effectiveness would change. The point I’m making is really not that difficult, I think.

Right now, we have conventions that limit the proper use of fire alarms. We don’t have those conventions in place because we don’t take fires seriously: they’re in place because we _do_ take them seriously and want to rally people into effective action whenever there’s a fire.

in order to preserve that function, some of those conventions limit the use of fire alarms. People can get in real trouble if they pull a false fire alarm. It’s not hard, I think, to see why that is.

Suppose that some people (who, to be clear, are prescriptivists even if they claim not to be) argue that we ought to expand the category of things that are alarm-worthy. Sure, we can pull a fire alarm in the case of a fire. But what if we see gasoline or old newspapers being stored improperly? Addressing that problem would, arguably, reduce the incidence of fires in the first place. And what about rooms that are crowded beyond the Fire Marshall’s posted capacity? Shouldn’t they count as alarm-worthy? Not on the old conventions, but why not just throw those out?

We also needn’t stop there. Perhaps we _mustn’t_ stop anywhere, because we can always do better. If we’ve got to the point where nobody overcrowds a room or leaves a fire hazard around, why not focus our attention on more subtle things that, we think, are more likely to cause fires? Maybe it should be okay to pull a fire alarm if someone who is known to be a smoker enters the building, The fire fighters might not have a fire to put out when they arrive, but they can give the smoker some good lectures on smoking responsibly. Yes, all the other people in the building will have to go stand outside when the fire alarm is pulled, but it would help raise awareness about fires and unite us in solidarity against them.

I take it you see the problems here. Can you see why similar reasoning seems compelling in the case of calling racism or calling violence? If you think the apparent similarity is superficial, then could you please explain why?Report

Maja Sidzinska
Maja Sidzinska
2 years ago

This issue is rendered more complicated by the fact that in this metaphor the villagers ARE the wolf (or the stray dogs or coyotes), which surely creates an additional incentive not to hear the complaints of the shepherd boys. (Over and above, that is, of not wanting to hear complaints in general.)

I am possibly sympathetic to conservatism when it comes to the definitions of terms (not sure what I think of this yet), but full-on supporters of “terms-conservatism” should give a defense of why they don’t think they should have to listen about the stray dogs and coyotes, as was implicit in Justin’s comments.Report

C.W.
C.W.
2 years ago

I think it pays to think about why all this started in the first place. There has been a powerful malaise among American political subjects for a long time now (probably as long as I can remember) that has been combined with a knowledge of a lot of terrible things in the world and a strong desire to change them and not just be all talk. People felt like they were shouting into the void in trying to flag all the terrible things going on, but no one was listening, and nothing would change. They would be criticized in making social media posts as just resorting to slacktivism without actually doing anything. And so, feeling this kind of pressure, and feeling both powerless and yet feeling the demand to do something to change things, these folks, and in particular academics, have started to key off on the idea that words make a difference, and so they have been busy trying to redefine things in such a way that the way we talk will slowly and subtly shape and change our society. They feel as if they are doing something by issuing these calls, that these shifts in language will make a difference, but it comes at the price of using a words in an inflated way outside of their natural homes. The desire is a practical effect to change things through discourse. It was born from a kind of pressing anxiety to try to do something to change things for the better, though. This is the key thing. It is an attempt to change things just through speech. Make of it what you will.Report

Michelle Ciurria
Michelle Ciurria
2 years ago

I was going to say something similar to ajkreider’s point (above). Those who favour ameliorative analysis – like Kate Manne – are going to be more congenial to concept modification (in whatever form) than those who favour conceptual analysis. Notably, concept modification doesn’t necessarily amount to concept inflation. Does the expansion of the concept “rape” to include spousal rape and date rape in the 1980s amount to “concept inflation” or “concept amelioration”? (Obviously the latter). The starting premise of ameliorative analysis is that common usage is rooted in patriarchal, colonial systems of meaning, which are fundamentally oppressive. Concept modification is urgently needed. That said, the boundaries of concept modification can be legitimately negotiated, and are currently being negotiated by feminist philosophers, critical race theorists, transgender theorists, etc. If anything, I would argue that philosophers tend to be semantic conservatives, whereas semantic activism is the path less taken. In any case, the battleground for the most ameliorative uses of language is WITHiN emancipatory subdisciplines, not within mainstream discourse.Report

JTD
JTD
Reply to  Michelle Ciurria
2 years ago

Regarding your example, was the concept really modified, or did society change certain background beliefs that led us to correct our views about the extension of the concept even though the intension stayed the same? For example, maybe our concept (qua ordinary language concept and not stipulative legal concept) is “non-consensual sex that is morally wrong” and what happened in the 1980’s is that a sufficient number of people changed their views about the wrongness of spousal and date rape. Here is one piece of evidence that supports this kind of story. Imagine two people in the 1950’s arguing over whether spousal rape is rape. What substantive things are they likely to dispute? Probably not the question of whether people commonly regard it as rape–they will both agree this is uncommon. However, we would expect them to be disputing its moral status.Report

Daniel Greco
2 years ago

I’ve recently been participating in a reading group on Herman Capellen’s “Fixing Language”, and it seems to me that there’s a certain degree of overlap between Case’s position and some of Cappellen’s discussion. Where Case talks about “inflation” Capellen talks about “exploitation”, by which he means, roughly, attempting to shift the usage of a word so that it has a new extension (i.e., applies to new things) while preserving its old connotations and associations. Here’s a representative sample:

“Exploiters undermine rational discourse by encouraging verbal disputes and in so doing undermine continuity of inquiry. They treat speech as a medium of manipulation, not as a medium for communication (i.e., as a medium for the exchange of thoughts and ideas). There are of course Exploiters with good intentions, but the overall effect of their exploitation is to contribute to and encourage a use of language that undermines what we should treasure the most about it: the continuous exchange of ideas. Exploiters are in effect anti-intellectualist opportunists that contribute to a destruction of genuine communication.”

(Not sure if html markup worked, but if it did, emphasis in original.) The book is available on Oxford Scholarship Online, for those who are interested.

For what it’s worth, I’d be inclined to take a much less hard/line stance than Cappellen: I think “exploitation” in his sense is sometimes all things considered defensible. But it depends on the case. To take two of Case’s examples, the original meaning of “gaslighting” was pretty specific, and referred to a pretty rare phenomenon that I don’t much value in being able to unambiguously refer to. Not so with “violence”. I want to be able to express a strong default norm against using violence except as a response to violence. I think a widespread internalization of that norm is (part of) the story of why there’s a lot less murder now than there was 200 years ago. If the extension of “violence” shifts, but the word preserves its connotations, I worry that the norm (along with associated ones) will be degraded. Responding to verbal provocations with physical attacks- may come to seem more acceptable—it’s meeting violence (new sense) with violence (old sense). Expelling students for participating in violent protests can sound fine, but when “violence” is used in an expanded sense, perhaps it shouldn’t.

And fwiw, this is all compatible with thinking that there’s nothing especially principled about the traditional meaning of “violence”. To take an analogy, consider chemical weapons bans. It’s really hard to say what’s so special about chemical weapons, as it’s certainly not the case that all chemical weapons are worse in all respects than all conventional weapons (let alone all non-chemical, non-conventional weapons). But I’m still happy we have robust norms against the use of chemical weapons, and I’d rather not degrade them. That’s how I feel about violence–I agree it’s not that principled a category to begin with, but I’m still happy that we’ve managed to coordinate on reasonably robust norms limiting it, and I’d rather not degrade those norms by degrading the (rough) consensus around what amounts to violence.Report

Oliver Traldi
Oliver Traldi
Reply to  Daniel Greco
2 years ago

Yes, I was thinking of bringing this in. To be clear, what’s being “exploited” for Cappelen are the “lexical effects” of a term. If Cappelen has his druthers, he writes, lexical exploitation “without an effort to make the case that the revision is topic preserving” ought to be “banned,” because it “undermine[s] rational discourse by encouraging verbal disputes and in so doing undermine[s] continuity of inquiry.” The topic of merely verbal disputes is important here. Note that many of the contentions of the concept inflaters become far less interesting when detached from the inflation. The innovation is precisely that X *is* a species of violence, that Y *is* a species of misogyny, etc.Report

Kate Abramson
Kate Abramson
Reply to  Daniel Greco
2 years ago

Side issue, but I’m honestly curious, Daniel– why do you think gaslighting is a pretty rare phenomenon? (I take it that I use the term in something at least pretty close to what you’re thinking of here as the ‘original meaning’, though perhaps that’s not right) And why do you not value being able to unambiguously refer to that phenomenon?
[honest curiosity about an ongoing arena of interest for me; not looking to get into an argument]
All bestReport

Daniel Greco
Reply to  Kate Abramson
2 years ago

Kate, I was assuming, as I take Case to do, that the original meaning is one where what happened in the movie “Gaslight” is a paradigm case. So gaslighting is the sort of thing that only goes on over time, in relationships where the gaslighter has a substantial amount of power over the victim. If that’s right, then I take it the thought was that you can’t have gaslighting in a one-off interaction, and probably can’t have it produced by entirely innocent motives.

I’m not sure about any of that as a story of how the word was used before the last decade or so. But suppose it is. In that case, then I think it would be right to see the kinds of usage we see in contemporary feminist discourse as a shift. But unlike Case, I’d be inclined to welcome that shift, for the same sorts of reasons that Manne encourages us to shift our usage of “misogyny”. That is, maybe the original usage doesn’t allow for collective gaslighting based on entirely innocent motives. But it still might be more useful to have a word that covers that too, if it’s much more common, and important to talk about. Especially because, unlike in the case of violence, it’s not really as if we’ve got a whole bunch of norms that involve coordinating on a shared understanding of what “gaslighting” is.

I want to emphasize how tentative all this is. I just meant to be using the example as one where, even if we are in the midst of an attempted shift in meaning, that doesn’t really bother me. But I’m far too ignorant about the history of usage to be confident that we are in the midst of an attempted shift in meaning.Report

Kate Abramson
Kate Abramson
Reply to  Daniel Greco
2 years ago

Thanks, Daniel. [I’m restraining myself from getting into an actual philosophical discussion about this here– not the right format– but I appreciate you taking the time to let me know what you had in mind]Report

TheStez
TheStez
2 years ago

“[‘political correctness,’ ‘virtue signaling,’ ‘moral grandstanding,’ and ‘concept inflation,’] are all labels that tell us to dismiss what is being said.”

Sometimes. But “mansplaining,” “white-tears, “white/male fragility,” “bad faith,” “bad actor,” and “stay in your line” are almost always dismissals—or accompanied by dismissals.Report

philosopher
2 years ago

Two broadly relevant papers
Maitra, Ishani. New Words for Old Wrongs.
https://philpapers.org/rec/MAINWF
Sterken, Rachel Katharine. Linguistic Intervention and Transformative Communicative Disruptions.
http://rachelsterken.org/onewebmedia/LinguisticInterventionsAndTransformativeCommunicativeDisruptions.pdfReport

Oliver Traldi
Oliver Traldi
Reply to  philosopher
2 years ago

Interesting papers, but it seems like here everyone is agreed that we’re talking about using *old* words for *new* wrongs.Report

Jon Light
Jon Light
2 years ago

What’s the difference between concept inflation and exaggeration? Is former a species of latter?

For example, suppose CNN keeps saying it’s Trump’s “worst week ever”, so we stop listening. Or what if Chris from Parks & Rec keeps saying he’s “*literally* the happiest he’s ever been”? Or that burrito place down the street with “world famous salsa”? Is that concept inflation?Report

Josh
Josh
2 years ago

Perhaps it is because we hold academics to higher standards, or perhaps it is the platform (we all must kowtow to the standards of editorial gatekeepers), but the choice of examples of concept inflation is rather telling. Egregious exemplars of the phenomenon go unremarked! Migration and demographic change is ‘white genocide,’ not enthusiastically celebrating a holiday is a ‘war on Christmas,’ ‘Taxation is Slavery.’ Economic migrants are ‘invaders.’Report

Smith&Jones
Smith&Jones
Reply to  Josh
2 years ago

I was saying something along these lines in my first comment on this post. Spencer replied to that. I’m not sure if he conceded the point or if he wanted to explain these examples as cases of hyperbole which he understands to be a different phenomenon.
Happy holidays!
(Also: Patton Oswald has a great bit on crying ‘white genocide’ in his newest stand up special.)Report