The Obese Police

Language is constantly evolving, but you know a change has hit the big time when the AP Stylebook makes it official. In light of all the recent news attention to Ozempic and related drugs, the usage guide’s lead editor announced in April that the entry for “Obesity, obese, overweight” had been adjusted. That entry now advises “care and precision” in choosing how to describe “people with obesity, people of higher weights and people who prefer the term fat.” The use of obese as a modifier should be avoided “when possible.”

In other words, the new guidelines endorse what has been called “people-first language”—the practice of trading adjectives, which come before the person being described, for prepositional phrases, which come after. If you put the word that indicates the condition or disability in front, then—the thinking goes—you are literally and metaphorically leading with it. Reverse the order, and you’ve focused on the person, in all their proper personhood. This change in syntax isn’t just symbolic, its proponents argue: A fact sheet from the Obesity Action Coalition promises that people-first language can “help prevent bias and discrimination.” Changing words is changing minds.

People’s minds sure could use some changing. The world is an awfully inhospitable place for fat people—I know firsthand, because I used to be one. But I also know secondhand, because the discrimination, bias, and downright cruelty are on display for anyone who’s paying attention. Nobody with a shred of decency wants a society where fatness, obesity, high BMI—whatever you call it—is an invitation to humiliation and scorn. So if using people-first language really can reshape people’s attitudes, or if it really makes the world even just a sliver more accepting, I’m in.

I am not at all convinced, though, that a diktat about language will ever make a dent in deeply entrenched enmity; and although the push for people-first language is undoubtedly well-meaning, there’s a whiff of condescension in the idea that people can’t recognize kindness and compassion without signposts put up by social scientists. Around every use of obese or fat or people living with obesity, there are lots of other phrases, and it’s those other phrases—not the people-first or people-last ones—that convey how the writer or speaker feels about fatness.

This puts me at odds with just about the entire medical establishment. “Because of the importance of reducing bias associated with obesity, The Obesity Society and all members of the Obesity Care Continuum have affirmed people-first language as the standard for their publications and programs,” Ted Kyle and Rebecca Puhl wrote in a 2014 commentary for the journal Obesity. The American Medical Association did the same in 2017. People-first language for obesity is now preferred at the National Institutes of Health and the Obesity Action Coalition. Ditto the American Academy of Orthopaedic Surgeons, the College of Contemporary Health, Obesity Canada, and the World Obesity Federation. You need to follow suit if you want to publish academic work in certain journals, present at certain conferences, or—as of this spring—write for any outlet that uses the AP Stylebook.

The problem is, there’s not much evidence that people-first language really can reduce bias, let alone eliminate it. The first position statement on the topic, put out by the Obesity Society in 2013 and co-signed by four other groups, offered just two references to prior research. The first pointed to a study done more than a decade earlier at Ball State University, where psychology researchers asked a few hundred students to describe a hypothetical person with a disability, and then surveyed the same students on their disability-related attitudes. The authors found that people who didn’t use people-first language in their descriptions had more or less the same attitude as people who did—although on a few specific items in the survey, they did show some signs of greater bias. (As the paper notes, “results were mixed.”) In any case, the study gave no reason to believe that students’ word choice was affecting their beliefs, rather than vice versa (which makes more sense). Still, advocates in the obesity field have been pointing to this research, again and again, as evidence that “people-first language affects attitudes and behavioral intentions,” as those advocates put it.

The Obesity Society’s second cited reference in support of people-first language points to a study that came out in 2012, led by Puhl, who is now the deputy director of the Rudd Center for Food Policy and Health at the University of Connecticut. Puhl and her co-authors surveyed more than 1,000 adults on how they’d feel if a doctor at a checkup used each of 10 terms to describe them, including obese, unhealthy weight, high BMI, chubby, and fat. On average, people said that unhealthy weight and high BMI were more desirable, and felt less stigmatizing, than most of the other options; obese and fat were just the opposite. But no one was asked about obese versus person with obesity.

For a paper published in 2018, a group of researchers at the University of Pennsylvania’s Center for Weight and Eating Disorders finally posed that question, in a survey of 97 patients seeking bariatric surgery. Respondents were asked how much they liked each of seven “obesity-related terms,” including some that were people-first (for example, person with obesity and person with excess fat) and some that were not (obese person, fat person). The former got higher ratings, overall.

But even the Penn study had complications. For one thing, not every people-first phrasing was preferred: Patients said they liked the term heavy more than person with excess fat, for example. Also, when asked to choose between obese person and person with obesity, the men in the group didn’t go for people-first—they preferred the more old-fashioned terminology. In a 2020 review, Puhl found that preference for weight-related terms differed not only by gender, but also by race or ethnicity, age, and body size. “People generally prefer more neutral terminology, like higher weight,” she told me recently, but some African Americans might like the word thick, while adolescents at a weight-loss camp favored chubby and plus size (but not curvy). Aspiring health-care providers were fond of unhealthy weight, understandably. Taken all together, she explained, overweight did pretty well, while fat and obese did not.

But again, very little could be said about anybody’s preference for (or against) people with obesity: Out of the 33 studies that Puhl used for her analysis, exactly one—the Penn survey—included people-first phrasing. As for whether using obese as an adjective might actually cause harm, and whether people-first constructions could ever ameliorate that harm, Puhl acknowledged that the evidence is thin. We have surveys on preferences, along with the occasional study (such as this one, on substance abuse) that shows people having slightly different reactions to written passages using different language. And that’s about it.

Read: The medical establishment embraces leftist language

It’s hard to imagine what persuasive evidence of harm from using obese as an adjective would even look like. How can we tease out a causal effect of language on social conditions? And, to muddy the waters even more, many fat activists make the case that all forms of the word obesity are stigmatizing. If you’re defining people with a certain BMI or above as having a disease, then how you choose to write your sentences doesn’t really matter, Tigress Osborn, the executive director of the National Association to Advance Fat Acceptance (NAAFA), told me. “Obesity as a disease state is dehumanizing in and of itself,” she said. Whether it’s used as an adjective or noun, the O-word pathologizes fatness.

Some doctors have subscribed to this belief. In 2017, the American Association of Clinical Endocrinologists and the American College of Endocrinology put out a statement citing what they called “the stigmata and confusion related to the differential use and multiple meanings of the term ‘obesity,’” which proposed a new alternative: “adiposity-based chronic disease.” But activists like Osborn opt for plain old fat. She described going to a diversity symposium when she was in college and meeting a NAAFA member who was unapologetic in her use of the word. “She was the first person in my real life who used fat as an adjective and not as an insult,” Osborn said. That’s how to destigmatize the word, she added: Just use it in an ordinary way, to describe an ordinary human condition. “You can’t destigmatize a word you can’t even say.”

When I asked Puhl and Osborn for some actual guidance on all of this, both responded with advice that is consistent with common sense and common courtesy. They talked about context: The language a doctor uses with a patient is going to be different from the language a journalist uses in an article about obesity statistics, which is going to be different from how we talk with friends and family. If the person right in front of you has a clear language preference, honor it. If you’re addressing a group, mix it up. If you feel respect and compassion, that will come through.

As a journalist on the obesity beat, I write about obese people pretty often, so I bristled when a well-known obesity researcher chastised me not long ago for using obese as an ordinary adjective. “Join the people who care,” he wrote. But the idea that word order telegraphs moral priority simply doesn’t jibe with how people actually speak and write, and insisting that it does burdens us with, at best, linguistic awkwardness and, at worst, abominations like people with overweight. True, you wouldn’t describe someone with cancer as being cancerous or someone with dementia as being demented, because those words have their own colloquial meanings. There are, however, other perfectly respectable health-related adjectives that get used routinely: diabetic, asthmatic, anemic, immunocompromised, myopic. And, I think, obese.

Language is, by its nature, majority-rule. A word’s meaning changes when enough people use it in its new, changed way. And I understand the hope and the compassion behind a top-down effort to change the way we talk about fatness. But I do not, cannot, see the value in replacing garden-variety adjectives with phrases that only call attention to themselves.

If ideas like this get traction, it’s because we don’t have many effective strategies to combat bias, so well-intentioned people latch on to anything that looks even remotely promising. But our public discourse shouldn’t be victim to attempts to rally consensus for a position that is largely unsupported by the evidence. Using people with obesity will not make much difference in the end. But the policing of language and, by extension, the ideas that it expresses, certainly might.