This article has been written with consultation from Alaska Elisabeth McMillan, a MA candidate in philosophy at the University of Toronto, whose focus is in the intersection of ethics and the philosophy of science. Anything not in direct quotations is my own extrapolation from her words, and opinions and views expressed in the article are only mine.
I’ve been working on this article for so long that if I look at it long enough to put references in it, I will commit arson. I may edit it later with citations but also? I may not. You’re not my dad and you can’t make me.
Covert hostility
There’s a way of saying I’m really interested in this thing which is passive-aggressive and mocking — a kind of office-culture speak which actually means “I believe this thing is worthy of derision,” or just “I think it’s stupid.” I thought about starting this article with the assertion that when I say I’m really interested in the way the cultural response to the concept of Body Mass Index has shifted in recent years, I am not doing that, and am in fact being very sincere. But I’m not sure that’s true. I wonder how much covert hostility is in me, and the answer is not immediately clear to me.
In the last few years, I have noticed that bringing up BMI anywhere seems to have approximately the same effect on the vibe of a conversation as suddenly taking my pants off, or making a really offensive joke. Even when I speak about it with people who are not over- or under-weight, the conversation tends to go in exactly one way — which is to say, poorly. I’ve noticed that the level to which public dialogue around BMI has become hostile has risen, in recent years, in a way that seems correlated to a distinct decline in how horrible I have found it to be fat in public. This is interesting, and I can’t help but speculate that these two things may be related.
This blog is named for a dark joke about my BMI at my highest weight. At some point in the last few years, I spontaneously developed the capacity to engage with the number with a kind of emotional distance that had never been possible before. Absolutely loaded on my own character development, I started writing this article thinking I was going to write what was, essentially, a defence of the concept of BMI, and a lambasting of BMI-rejectors as gullible in the face of 2010s woke-scold clickbait, or else hysterically emotional enemies of science. I was so ready. I was going to be like the Red Scare Podcast girls of being fat on the internet. It was going to be so awesome.
As tends to happen, though, when one takes a serious look at a subject, I eventually found the place where my position was informed by a failure of empathy — with others, but also with a version of myself which is not so far in my past. This rang very false, and made me sound stupid. And I can’t have that! So, here’s something else:
Body Mass Index: what?
Body Mass Index is a numerical value. This numerical value is achieved by doing a math equation. This math equation takes the weight of a person in kilograms and divides it by a square of their height in meters. The resultant number gives a rough estimation as to the thickness or narrowness of a person. It doesn’t give any information about fat-to-muscle distribution, the weight of their skeleton or organs, or anything else — it’s just a very general grouping of people based on how much air they displace.
Something important to understand is that the number is made out of qualities that a person has, but the number itself is not a quality the person has. The number is a constructed category made from first taking data about a person’s qualities and then running a math equation. To make sure I understood this correctly, I texted a bunch of questions to my best friend, Alaska — an MA candidate in philosophy at the university of Toronto.
“I'm sitting in my apartment right now,” Alaska explains, over voice notes, “And I'm looking at my purse. My purse is brown — and so we might want to say that brown is a quality that my purse has.” On the distinction between the brownness of a purse and the BMI classification of a person, Alaska explains that a person’s Body Mass Index is “not a quality of a person in that it's not something that would exist outside of the classification framework that we've created.”
My weight might be variable and subject to change based on my caloric intake, activity level, or weird changes in the effects of gravity, but it’s still a measure of some quality my body has. This can also be said of my height. My BMI, however, is the result of plugging these qualities into human-constructed process of abstraction. This abstraction process manufactures a value and sorts me into a category which is, itself, constructed — it doesn’t, on its own, say anything about me.
“We've created these buckets, and we've put populations into them,” Alaska explains. “And so now you could say, well, Jerry's in the blue bucket — but that doesn't mean that there's something about Jerry that is blue-bucket-y.”
The buckets created by the classification of BMI are underweight, normal weight, overweight, and obese. Statisticians have run a great number of large-scale tests and found that there is a statistically relevant increase in the likelihood of disease in underweight, overweight, and obese poplulations. This does not mean those people will definitely get the diseases we’re talking about — it just means that the likelihood goes from 1 in 300 000 to 1 in 100 000, for example — a threefold increase in statistical likelihood which still leaves the odds very much stacked against getting a specific disease.
“Underweight” and “obese” are further subdivided into smaller categories to indicate more abrupt jumps in the statistical likelihood of disease. It is a tool that statisticians offer to healthcare professionals as a way of generating hypotheses — working out which questions to ask next — about which set of tests to order, or care strategy to prescribe.
The gap between storytelling and random happenstance
Scientific explanations of things can be very roughly categorized into mechanistic and correlative ones. A correlative explanation is the very hands-off presentation of a statistical relationship between two variables. These variables covary, which means that they change together. A common example is that as ice cream sales go up, so do levels of violent crime. As the ice cream sales come back down again, the levels of violent crime do as well. These variables predictably move at the same time — but we cannot say with any certainty that increased consumption of ice cream causes or otherwise brings about the increase in violent criminality. To say that increased ice cream consumption is has a direct causal link to violent crime would be offering a mechanistic explanation of the phenomenon.
“Both of these explanations are useful and meaningful and help us understand the world around us,” Alaska explains. “We use both of these explanations to inform our classifications. Which explanation we use implies different things about what we can infer from our classifications.” Both explanatory models help to issue hypothesis about things, but “the kinds of hypotheses that they're going to help us issue are going to be different.”
People say correlation doesn’t equal causation a lot, and what they essentially mean is that while these two ways of explaining stuff are interrelated and inform each other very often, the power that they have to inform each other is affected by other, more complex assessment strategies — in my ice cream and violent crime example, it’s been determined that hot weather has a mechanistic relationship to both increased ice cream sales and violent crime rates — the ice cream and the violence are not related to each other, they’re related to some third variable we weren’t looking for. Somebody who isn’t me would also have to assess in which direction the mechanistic causal process is going, and do a bunch of other science things I’m not smart enough to understand. Basically, we don’t have to tell wild stories about why things happen. We also don’t have to dismiss everything as coincidence. There are more rigorous, mysterious, scientific, third things we can do — and gaps between all of that which are wicked easy to cross for understandable but ultimately unscientific reasons. This will come up later.
This math equation’s pattern of harm-doing, refusing accountability, and just being unhelpful generally
One of the things that comes up a lot when I try to talk about Body Mass Index is “its racist origins and applications.” I will admit that I have become skeptical, in quite a reactionary and not-intellectual way, of basically all accusations of racism against things or concepts in recent years. I never used to be. In fact, I’ve always found the intellectual steps required to label a concept or tool as racist to be very intuitive. I have, however, been hardened by two things: the first being the more frequent experience, as I got older, of tangible acts of racism in my material life, which came with the sinking feeling that they had some quality which was different than the more abstract theoretical racism I wrote about in the early parts of my undergraduate degree. The second, that I spent a great deal of time navigating a social justice youth culture which is still recovering from a somewhat overzealous application of the term to everything from white men holding brown babies to Taylor Swift’s general vibe.
I find this all rather silly. I have emerged from my twenties pretty much immediately dismissive of the use of the world racism for anything other than hate speech, physical violence, or organized white nationalism. It’s not that I don’t think any tools or concepts can be racist by design, it’s just that I’m going to want to be very rigorous of my assessment of this assertion. And I’m going to be kind of a dick about it. Also, I’m an Indigenous woman. So you can’t be mad at me.
The argument that BMI is racist goes like this:
The man who invented the BMI was not a doctor
The man who invented the BMI was interested in the study of the ideal man, held this as a standard, and considered all deviation from it to be diseased or disordered
This man’s research was fundamental in the establishment of numerous schools of thought and assessment which have caused damage to populations subjugated by global white supremacy.
Buddy’s construction of the ideal man was constructed from the study of white guys
As a result, BMI does not map properly onto female or nonwhite populations, who have relevant anatomical differences
Also, BMI was never meant to assess the health of a person, but populations of people
At least in the USA, nonwhite populations tend to consistently score as having higher levels of “abnormal” body compositions, on the BMI scale, compared to white populations
BMI is an attempt to impose a white body standard on the world
I find this argument incoherent for a lot of reasons, which is too bad, because there are plenty of reasons to dismiss or downplay the importance of BMI that have nothing to do with anything mentioned up here. In an attempt towards charity, let me break it down:
The man who invented BMI, or at least the ideological precursor to BMI, was absolutely not a doctor. His name was Adolphe Quetelet, he was from Belgium, and he was a mathematician. Specifically, he was into statistics and probability theory — new disciplines, at the time, and applied mostly to astronomy. So, he was an astronomer, and a statistician. Eventually, he had the idea that we could use statistics in the study of human populations. This was very huge, at the time. The time was the mid-19th century — Quetelet was born in 1796, died in 1874, and did most of the work which lead to the Body Mass Index between 1820 and 1850. Quetelet wanted ask questions about, for example, how likely it was for a crime to happen in a specific area, and then ask about the income level of that area, and then ask about how those things might be connected. If this sounds familiar, it’s because this dude was instrumental in the invention of sociology, as a discipline — all social sciences, really. Statistics are still a huge part of those fields — largely because of his work. So, no, he was not a doctor. BMI is not a measure of health, it’s a statistical tool. Quetelet was a statistician.
The second point is that Quetelet was interested in the study of the ideal man. I see people use this terminology on blogs which seek to undermine BMI as a tool. Interestingly, other sources quote the original French as the study of l’homme moyen — which translates to the average human or the most normal human. “Homme” directly translates to “man,” of course, but it’s meant here in that way that “the dawn of man,” in English, means “the beginning of the human race.” I don’t mean to imply that writers interested in the critique of Quetelet’s methods are intentionally misquoting him — there could very well be another version of the original text that I can’t find — but I do think that the huge gap in the kinds of implications carried by “the study of the ideal man” versus “the study of the average human” is significant, and worth considering. Quetelet was far from the first person to muse about the proportions, mathematically speaking, of the most average of all average people — it’s also what da Vinci was doing in his famous study the vitruvian man. I think it’s probably much easier to be forgiving of da Vinci, though, because the results of his exploration were very pretty pictures, while the results of Quetelet’s are often people feeling bad about themselves at the doctor’s office.
In the interest of charity, and towards point three, I will concede that the designation of any standard of normalcy is both a completely artificial designation and also has implications that have caused a great deal of human suffering. The concept of normal is always an averaging-out — the flattening of human difference in the interest of some abstracted thing which doesn’t actually exist in real life, and also paradoxically contains within it at least the possibility that deviations from those standard could be considered diseased or disordered. This is a possibility Quetelet explored with a great degree of seriousness.
I’ve obviously found myself struggling against my own understanding of what’s normal, and what it means when I, or the people I meet, deviate from this standard, because everyone does. And we deviate from it all the time — because the very concept of normalcy is itself a human construct. I’ve spent thousands of dollars on therapy trying to cope with my own anxiety about how I deviate from standards of normalcy, spent lots of time fumbling towards compassion for other people whose behaviours don’t line up with the standard of normalcy to which I am accustomed. Quetelet is often credited with inventing the concept of the normal person — a persistent ideological concept, which underwrites everything from Marxism to public health to experimental psychology, as Todd Rose explains in this article for the Atlantic. The Normal Person flits in and out of my conception of the real — sometimes I can almost see him, hear him, touch him. Sometimes I think he’s never been anything but a very compelling human myth.
More towards point three, Quetelet’s work informed the practice of statistical models being applied to criminology and sociology in ways which were controversial at the time in which he wrote them, and continue to be. Statistical models are used to guess the qualities of perpetrators of specific crimes. They are used to assume which social problems are likely to be experienced by groups of people experiencing x or y manner of social disease. Often, this practice is helpful — until it is not, and it leads to extraordinarily violent outcomes. Part of the problem with the application of statistics in this way is that it’s very hard, even for trained and expert professionals, to avoid making correlative-mechanistic leaps about the character profiles of the individual people contained in these data sets. Sometimes you have to make those leaps to make sense of the data. Sometimes people make those leaps to service their own bias — racists do that all the time.
Articles about the racist character of Body Mass Index bring up, very often, that Quetelet’s work heavily informed the field of eugenics in its early-twentieth-century iteration. Darwin is, of course, also a profound influence on the whole mess — despite the fact that he profoundly disagreed with this development of his work. The first major book by the first real eugenicist, Francis Galton (a half-cousin of Darwin) was published three years before Quetelet died, when he was a very old man, and after his capacity towards intellectual labour had been heavily impeded by a stroke. I simply don’t think it’s reasonable to hold person A individually responsible for the insane speculations person B makes about their work decades after it’s published, and Quetelet’s personal responsibility for a racist quack’s invention of a pseudoscience seems like a reach, at best, to me. This kind of argument always feels akin to saying Salinger is somehow personally responsible for the murder of John Lennon.
Quetelet’s work is also the basis for ergonomics. Maybe I’m being overly charitable, but this seems, to me, like a more faithful application of his work.
Point four, that Quetelet’s study of the average man was based on white men, is true. It reflects the medical standard, as well as the standard for personhood, that was both boneheaded and stupid and also generally accepted in Europe at the time. Basing the medical model of personhood around the body of a white man is an abstract concept I have no trouble calling racist, and also misogynistic, and which is also much bigger than Quetelet, or his choices. I continue to identify what I would call a weird rhetorical mistake, made by the popular turn away from Body Mass Index, in the implication that seems to be made by writers — an implication that, should proper studies be conducted on more racially or gender-diverse populations, the correlative connections we see between body mass and things like heart disease or diabetes would be blown wide open. This implication is either ignorant to or dismissive of the fact that those studies have happened. The results were that Body Mass Index was determined to underestimate the likelihood of obesity-correlated diseases in Asian populations due to some specific quirk about having a body adapted to being in Asia — which means that there were a bunch of people whose body compositions did, in fact, put them at a higher disease risk who didn’t know, because they were being measured against a white standard. Asian people, apparently, tend to actually have a higher body fat distribution, even at lower weights, and BMI has been adjusted down — with “overweight” beginning at 23 instead of 25 — for people with a high percentage of Asian ancestors. While studies on Black populations did seem to find a slightly higher lean body mass percentage compared to nonblack populations of the same weight, the difference was nowhere near the level required to adjust the BMI standards for Black or other nonwhite populations. This is left out of all of the readily-available reading on the subject — I guess because it’s not very exciting or interesting.
A much more interesting angle on question of whether or not BMI is racially biased, which I have read a lot, brings up that the calculation further stigmatizes health problems caused by racism, and does this by failing to control for social factors which make it harder for racialized populations to access healthcare. This argument often uses the fact that BMI is a standard developed by and for white people as a premise for its argument, and this premise isn’t satisfied by the fact that BMI does accurately predict correlative health outcomes for nonwhite populations. I’m tempted to call this incoherent, except that I think it gestures towards a very real problem in healthcare more broadly — that historically marginalized populations might not be interested in using tools which have a kind of historically marginalizing flavour, if you will — the way that BMI does. This resistance may be dissociative, but as long as it exists it’s good to know it’s there, if the goal of healthcare is not convincing people of the truth but of meeting them where they are with the highest standard of care they can accept or participate in.
I’ve mentioned before that BMI does, in fact, accurately map disease prediction in most nonwhite populations (excluding Asians). The BMI-is-racist argument counters that this does not take into consideration the racism-specific living conditions of nonwhite populations. I fully accept that nonwhite people in North America are more statistically likely to be deprived of the basic requirements for good health and nutrition. It seems intuitive that this is caused by the conditions imposed by systemic racism, which I very sincerely believe is a thing. But that doesn’t seem to me to have much bearing on whether or not BMI is useful or not — because it seems like this argument is framing BMI as a test someone can either get a good or a bad score on. In the absence of a good score, an explanation has to be offered as to why the person has scored poorly, lest it reflect something bad about the person’s character in some way. But that’s not what BMI is. BMI is a statistical classification tool that makes it easier to make more accurate guesses about which diseases are more or less likely to present as a problem.
Local intellectual attempts to pull head from own ass
I think it would be easy for me to close the article there, and be like isn’t it cool how I solved this problem forever. I could hit publish and go on with my day and probably lots of people would agree with me. I could say all of these things as if I’m describing a way of thinking which is silly and I do not understand. Like I am above all that. As if I have no idea how it could enter somebody’s mind that BMI as a test you can either pas or fail. Where that idea could possibly come from.
But I do know. I have felt that thing. I have seen the wrong number and I have known it not as a flat piece of data, but, in fact, as a referendum on my health and my wellbeing and my future. And my worth as a person and a woman. I have seen the deep, esoteric meaning lurking under the clean and impersonal data point, and I have felt the bad thing.
This means that this is the place where I’m forced to reckon with the failure of empathy which underwrote my earlier assumptions on how I’d write this article. So, personal story time: I have, in the past few years, managed to iron out a lot of the fundamental self-image and self-esteem problems which completely controlled my life for my teens and my twenties. I’ve come so far that it’s sometimes very hard to get into an empathetic headspace about how I used to think and feel about these things — probably because I’m more than a little concerned that the monumental self-loathing and self-destructive behavioural patterns that underwrote fifteen some-odd years of my life might find its roots in me again. But the fact is that I came of age wracked with angst about my body — and more specifically, what my body meant. How was I read by others? What meaning were they making? Which roles was I playing, in which stories? In the medical stories, the relational ones, the ones about gender and sex and death?
All of this fundamentally underwrote what I thought about my worth. I could chalk this extreme distortion of my self-concept up to my own hysteria or my weakness to external pressure or influence - but that doesn’t seem very fair to my younger self, who was very smart and principled and tried very hard to be good to herself, despite everything. There is obviously a persistent public perception issue when it comes to the relationship something like BMI has to a specific person. I know because it got inside my head. I lived it for a long time.
The layperson’s lament
I asked Alaska what she thought this public perception problem was. “I only have access to public perception from the standpoint of my own perception,” she said. “On the one hand, we have the treatment of BMI as though it's measuring a real thing. We treat BMI as though BMI is causally responsible for negative health outcomes. In the same way having high blood glucose might, or having too much arsenic in your system might, or even drinking too much water might. On the other hand, we have social bias. We value people based on how they look. And BMI roughly corresponds to our value-based judgments of people's appearance.”
Growing up a girl, I was constantly bombarded with the messaging that the truly important things I had to look forward to in my life were all attached to love and desire — being attractive to boys, great love stories, marriage, and children. Since women are also socialized to be very passive in the roles they play in love and desire, my success or failure in this pursuit followed directly from how closely I could come to a cultural standard of attractiveness. Coming acceptably close to a cultural standard of attractiveness was the thing which would cause the waterfall effect of being the subject of male desire, which would allow me to continue to play my passive role while also getting the things my life was for: which was not meaningful work, or friendship, or community, or self-actualization — but great love stories, marriage, children, and other things which are fine to want but not at all a requirement, and in no way indicative of someone’s success or failure at the project of womanhood. My proximity to beauty would, at least, lead me away from the invisibility or outright disdain which were awarded to unattractive women — a fate from which my intellect, talents, or personality couldn’t possibly save me.
This is, of course, all horseshit. I don’t believe any of it, on an intellectual level, and it’s stuff my parents and some great literature all tried very hard to encourage me not to believe — but it’s cultural messaging that exists, out there, generally, in the form of social bias and the legacy of misogyny. Women have to be pretty much constantly vigilant against the internalization of a huge array of normative claims about our value based on our appearance — and sometimes, something sneaks in, despite our best efforts. And it can take like, years of therapy to wrench it back out again.
“It's not just that people think that being thin is beautiful,” Alaska elaborates. “If it was just that, we would have less of a problem. It's that people think that being fat is ugly. I don’t — lots of people don't think that, but I'm just sort of speaking in generalities here: our social norm prescribes thinness as a kind of beauty and fatness as a kind of ugliness. We correlate fatness with being unhealthy, and sometimes with other things, like laziness. And people also have this classification of people [BMI] that gives them a kind of like prescriptive advantage over it just being a matter of aesthetics.”
To further elucidate the point, she brings up how a specific person may have an aesthetic preference for dating men who are tall, as opposed to men who are short. “But then it’s like, from there, I said that the reason I liked tall people is because people who were short were more likely to have poor health outcomes, but they also were responsible in some way for their shortness. I would be equipping myself with a reason to move beyond my appreciation of tall men as an aesthetic thing, and move toward it being a kind of normative claim — not just that I prefer tall men and think tall men are beautiful, but that tall men are better, and that if you're a short man, you should think about the fact that you're not as good as a tall man. And that's obviously really fucked up. Right? If you if you don't think that's really fucked up, I'm confused.” She stops for a second, and then ads: “But I know you. I know that you would think that's really fucked up.”
I do think it’s fucked up, by the way. Short dudes come ruin my life.
“We end up moralizing preferences in a way that sort of draws them external to us, where people feel justified and they seem justified in saying things like, Well, I only call you ugly because I care about your health, or I only point out how fat you are and how I don't think that that looks good, because I care about your health.”
Bring back smoking in movies. We deserve it.
But it’s probably not about caring about people’s health, Alaska elaborates, because there are plenty of examples of social bias in favour of objectively unhealthy things. “For a long time we thought smoking was incredibly sexy,” Alaska offers. “We knew it was bad for you. But we kind of liked that it was bad for you! It kind of added to the sex appeal, right?” Though cigarettes never became a serious fixture in my life, I grew up in the era of smoking in movies, was a child during the grunge boom of the 90s, and started hanging out with Alaska when we were both art freak / crust punk hybrids in the Halifax underground — a place where the four food groups were black coffee, malt liquor, candy, and cigarettes. Changing the personal relationship people our age had to cigarettes only happened “once we had moved it out of the domain of being sexy and into the domain of being ugly. And now there are a whole bunch of people in their 30s who are addicted to smoking cigarettes, even though when we started smoking — I don't smoke anymore — but, when we started smoking, we knew it was bad for you.” She explains that the kind of legitimacy something like BMI can offer “equips us with this normative tool that allows us to make the transit from aesthetics into moral judgment. And that's really dangerous for people individually.” Dangerous, I presume, because we ultimately feel compelled to act based on our moral convictions in a way we don’t feel compelled to act based on our arbitrary aesthetic judgements — and should not, because that would be weird.
So I do understand that it’s hard out here — and I think that the cultural knocking-down-a-peg of Body Mass Index, as a concept, seems to be a part of a general movement away from moralizing about bodies in general. I’ve noticed my doctor using different language about my weight, and my body, in the last few years. He weighs me, but his designation of my health is based on the routine tests I get every few times I visit — blood glucose, cholesterol, blood pressure, and so on. I can buy clothes that are cool and look good on me as easily as I would buy clothes in a small or medium. Hearing a bunch of extremely mean things about my body, and fat people in general, used to be something I just expected to hear from the media and in casual conversation several times a day — now, if I really want to hurt my own feelings, I have to go looking for it. None of this used to be the case. I don’t think it’s a coincidence that I don’t feel completely frozen, broken-down, and defensive about my body anymore — I can know what my weight is, and what my BMI is, and the knowledge brings me no distress at all.
This would, of course, be the ideal response for most people. I don’t know if I think it’s realistic to put BMI back in the box, so to speak — which leaves the option of a long, slow reeducation on what BMI is, what it’s for, and just how distant a relationship it has with any qualities that exist in the body. It would be in the assertion that the relationship BMI has with disease and discomfort is a correlative one, and only examined to be able to ask more effective questions, and provide each person with the information they need to decide which course of action makes the most sense for them. How we get there, I don’t know. I’m not a doctor. I’m not even a statistician.
The paid edition of MORBID includes weekly insights into the weight loss efforts of a recovered bulimic art hoe just trying to have a good time and reduce her chronic pain while she’s at it. Subscribe below.
🙌