"Measuring Conceptual Inflation: the Case of ‘Racist'" - Nat Hansen (University of Reading) & Shen-yi Liao (University of Puget Sound)

By Nat Hansen & Shen-yi Liao

Has the word “racist” lost its meaning?

In a campaign rally in 2024, J.D. Vance interrupted his own argument in favor of voter I.D. laws to say that “Democrats say that…it’s racist to do anything. I had a Diet Mountain Dew yesterday and one today. I’m sure they’re going to call that racist too”. If it’s racist to do anything—including doing the Dew—then “racist” has lost its meaning. And that’s bad. We want words like “racist” that express intense moral criticism to retain their power, otherwise we won’t be able to call out genuinely bad things in the way they deserve.

Vance isn’t alone in his worry. Pundits and theorists from across the political spectrum have claimed that the meaning of “racist” is being worn out through overuse. “Conceptual inflation” is the name the liberal philosopher Lawrence Blum has given to the idea that the meaning of “racist” is being diluted. As with monetary inflation, people use the word more, but each use is worth less than before.

“Racist” isn’t the only word that people have worried is suffering from conceptual inflation. “Concept creep” is what the psychologist Nick Haslam calls the widespread use of words with technical meanings for harm in clinical psychology, like “abuse”, “trauma”, and “mental disorder”, to cover less severe examples than they once did. It has been claimed that concept creep makes words like “trauma” mean less and less generation by generation; Gen Z students are alleged to be “so reactive to potential difficulty as to consider everyday challenges triggering” that they call everything “trauma”, including “breaking an acrylic nail”.

We know monetary inflation is taking place because economists can measure it with the consumer price index. It can tell us when inflation is really happening, and also when it isn’t, even when we feel that it is, as happened in the recent “vibecession”. Is there a way that we can measure the inflation of the meaning of words like “racist” in a way that doesn’t rely on mere vibes?

As philosophers who prefer running experiments to sitting in armchairs, we wanted some evidence. There is data showing a substantial increase since the 1970s in the frequency that words like “racist”, “sexist”, and “homophobic” occur in newspapers and academic journals. But there are lots of reasons that the frequency of a word may increase over time that don’t mean it is losing its meaning. Take a taboo word that appears much more frequently in print today than in the 1970s, like “fart”. It’s unlikely that the greater frequency of the word is evidence that its meaning has been inflated—though, as the kids of one of the authors pointed out, there probably are in fact more farts now because there are more people—it’s just that it’s now much more acceptable to talk about farts and print the word.

The gold standard for measuring meaning change over time would be to study the way a particular group of people use the word over time, like in the British Up documentary series, which interviewed the same group of people every seven years, starting when they were children in 1964 through the final installment in 2019, when the interviewees were 63. Unfortunately we can’t go back in time to run the historical part of such a “real time” study, against which we could compare present use to determine whether the meaning of the word has changed. But there is a method used in sociolinguistics, known as “apparent time” that uses the way people of different ages currently use language as a window onto the past.

The idea behind the apparent time method is captured in a remark of the French painter Paul Cézanne, who said: “it’s in the look of people my own age that I see the past again”. In an apparent time study from 2020, linguists Sali Tagliamonte and Katharina Pabst showed how people of different ages used different “adjectives of highly positive evaluation”, like “amazing”, “cool”, “awesome”, “great”, and “wonderful”. Your grandparents probably would not have praised something by calling it “awesome”, they would have said that it was “wonderful” or “great”. Kids these days would probably use another word—“sigma”, or something else middle-aged folk haven’t heard of yet (a young linguistics student at UCL told me “leng” is the current adjective of highly positive evaluation in London).

In our paper “Measuring Conceptual Inflation: the Case of ‘Racist’”, recently published in Ergo, we report on several apparent time studies we ran to determine whether “racist” was suffering from conceptual inflation. We were surprised: we didn’t find clear evidence that it was used differently by people of different ages. In contrast, we did find clear evidence that words like “slutty” and “queer”—examples of (partially?) “reclaimed” slurs—were undergoing meaning change over time, being applied more widely by younger people and losing some of their moral intensity.

Of course, the apparent time method can’t detect if every user of language, young and old, has inflated the meaning of “racist” in comparison with the way it was used in the past. If that was the case, the apparent time method wouldn’t be able to detect change, because it’s just looking for whether older speakers use terms differently than younger speakers. But the fact that our apparent time studies did find changes in the way younger speakers use terms like “queer” and “slutty” make it less plausible that only “racist” has changed in this hard-to-detect way. And in another recently published paper, “Conceptual Inflation”, published in the Taiwanese journal EurAmerica, we use corpus data to challenge a widely-held view that “racist” originally was used to apply to individuals and mental states (racist people, racist beliefs), but has recently been expanded to include social institutions (racist policies, racist laws). Contrary to that standard view, we found that the institutionalist sense of “racist” appears to be as longstanding and as frequent as the individualistic sense, at least in uses of “racist” that appear in the Corpus of Historical American English.

So why do people like Vance contend that the meaning of “racist” is being inflated when we couldn’t find evidence that it is? We think three factors explain this difference between perception and reality. First, in our studies, we found there is a lot of variation in how different people use “racist”—some people use it more narrowly, some use it more widely, and people differ in terms of how bad they think it is to be called “racist”. But this isn’t a widespread societal change, it’s just diversity in language use, which is fairly enduring over time. Second, as mentioned above, “racist” now occurs more frequently than in the past, so people are more likely to notice the diversity in how the term is used, especially when their own actions get called “racist”. Finally, it’s easy to mistake something that you just noticed (like a young person calling something “racist” you think is no big deal) for something that is genuinely new. Linguists have called this a “recency illusion”.

Put those three factors together and it’s easy to see why people are noticing more and more uses of “racist” that differ from their own, especially those that target their own actions or beliefs, and interpreting those different uses as evidence of inflation.

Debates about what racism is and what we should do about it are not going to go away. It’s tempting to switch from talking about difficult solutions to political problems to arguing about the language we use to describe those problems, because our language seems easier for us to change. As philosophers of language concerned with the meanings of words, we recognize we are susceptible to this temptation too. But our goal in developing reliable methods of measuring whether conceptual inflation is actually happening, or just a persistent illusion, is to shift focus back to the difficult solutions.