Beanbag People
On a video call with two close associates about a year ago, before getting down to business, we did what we usually do: banter a bit about whatever was on our minds. I was sipping water from a glass container with a handle, and I referred to it as a “glass.” Often on the receiving end of my playful-but-annoying “corrections,” one associate said: “I don’t know, Jason. That looks more like a mug to me than a glass.” I’m trained as a linguist, so I couldn’t resist expounding on some facts about how human categories work. I noted that one of the earliest scholarly essays about what would later become “prototype theory” was about what people did and didn’t consider to be mugs, cups, vases, and bowls.
Our banter soon took a serious turn, though. I mentioned that I’d been thinking about writing an essay connecting prototype theory with questions of human belonging. One of my associates then shared a story about a recent experience she’d had.
In a conversation with friends, the topic had shifted to the “Asians” in the group. My associate identifies as mixed-race (with a white father from the U.S. and a mother from a Southeast Asian country). When the question came up as to whether she should be “counted” as a member of the Asian group, one friend said no. When my associate said she considered herself Asian, the friend said, “Yeah, but you’re half-white.” By itself, this comment wouldn’t necessarily have been hurtful, but in this case, the friend was using this designation in order to exclude my associate from the ingroup of Asians. My associate had also been raised in a predominantly white community, in which she regularly experienced being othered as “exotic,” and being asked, “What are you?” As she shared her story during our video call, she wondered aloud: “Will I ever be fully anything?”
In that moment, the gravity of not belonging hit me at a new level. As a white, cisgender, heterosexual man born and raised in the U.S., belonging is a birthright that I’m not forced to question. The pain in my associate’s eyes and voice gave me a glimpse into a different universe: one that’s in plain view if we choose to look.
The stakes couldn’t be higher. The need for belonging is as basic as the need for food or water, and the social pain of exclusion is processed in the same parts of the brain as physical pain. That’s what the current science says. This means that when we exclude, we are literally causing pain, and we are literally causing people to feel like their very survival is at risk. We also know that when we feel threatened — and our brain’s threat-detector, the amygdala, is active in our brains — we lose access to our neocortex, where our “higher” cognitive functions are carried out. We lose the ability to plan and strategize, along with a host of other functions crucial to succeeding in the workplace.
All of the scientific evidence points toward the need to create workplaces where people feel they belong — most importantly for reasons of basic humanity, and secondarily because organizations that truly foster belonging tend to perform better (which you can read about in many other places). This requires us to see every human as 100% human. 99.99% won’t do; only 100% works.
Here’s the problem: it’s really hard for our brains to do that. And it’s because of how the human brain pieces the world together.
What is a “prototype”?
The seminal study I mentioned above about cups and mugs heralded an intensive phase of scientific research into how humans categorize. Before this, categorization had mostly been a concern for philosophers. Nobody had bothered to take an empirical look at how humans actually form categories. That started to change in the 1970s, with the advent of “prototype theory.”
In essence, a prototype is a reference point. When we see something new, we automatically and quickly compare the new thing to these reference points. If we see an object with four legs, a flat surface and a straight back, and it looks like a human could sit in it, we’ll call it a chair. We recognize the shape immediately, and we would all agree that it’s a chair. But what if a chair happened to have only three legs? Or no legs at all, as is the case with some modern modular chairs? What about beanbag chairs? Are these all “real” chairs? Are they all equally deserving of the name chair?
Eleanor Rosch and her colleagues decided to investigate these questions empirically. In a series of studies, they discovered various forms of evidence that humans do indeed treat some members of categories as “better” members than others. For instance, are all birds equally “good” birds? The answer is no. Instead, what they found were the following “prototype effects”:
Direct rating: Subjects were asked how good an example of a category a given instance is — e.g., how good a bird a robin is, or a duck, or a penguin. Ratings were from 1 to 7. Robins and sparrows got the highest ratings.
Reaction time: Press a button to indicate true or false in response to statements like: “A penguin is a bird.” Robins and sparrows got the fastest response times.
Production of examples: either list or draw different kinds of birds, in whatever order the subject chose. Robins and sparrows were listed or drawn earliest.
Similarity judgments: Subjects were asked to say whether statements like this were true: “A chicken is similar to a robin,” or “A robin is similar to a chicken.” “A chicken is similar to a robin” got higher truth ratings.
Judgments about generalization: People are more likely to think that a goose (a less prototypical bird) can get a disease from a robin (a more prototypical bird), than the other way around — i.e., that a robin can get a disease from a goose.
The findings of all these studies were overwhelming: humans simplify the world around them by identifying a set of attributes that “central” members have, and then seeing whether other possible category members share these attributes. With birds, for instance, robins and sparrows are in a particular size range; have feathers and beaks and wings and a specific shape of leg; and fly. In contrast, consider penguins: they’re bigger than robins and sparrows; have beaks but no feathers; have different-shaped legs; and have wing-like appendages, but they aren’t used for flying. Ostriches have beaks and feathers and wings, but they also don’t fly, and they’re extremely large. We could do a similar analysis for any bird, and to the extent that it shares core attributes with prototypical members like robins and sparrows, it will be considered a “good” bird — with other birds relegated to “peripheral” status.
A quick side note: some readers might be thinking, well, aren’t there ornithologists out there who can tell me for certain whether something is or isn’t a bird, period, end of story? Yes and no. Yes, in the sense that there are “expert models” of what is and isn’t a bird; no, both because ornithologists disagree among themselves, and more generally because expert models are still human models. Ultimately what we’re interested in is how regular humans put things into boxes called categories.
Humans need to simplify
Our brains use prototypes because we need some way to sort quickly through the essentially infinite quantity of information arriving through our senses every moment. And here we can begin to see how this relates to social categories. If you've ever taken a workshop on unconscious bias, you’ve probably heard the idea that bias is a kind of “mental shortcut” that the brain uses in order to conserve precious energy. One study found that, while the brain takes up only 2% of the body’s mass, it consumes 20% of the body’s energy. With energy at a premium, the brain needs to find ways to economize. One way of economizing is using mental shortcuts, and prototypes are an extremely common shortcut.
It might make intuitive sense that robins and sparrows are “better” birds than penguins and ostriches, or that dining chairs are “better” chairs than beanbag chairs. But the human need to create prototypes doesn’t stop with such obvious cases. It turns out that humans are so good at creating prototypes that we do it even for categories that, logically, should not have any prototype structure at all.
Take the concept of even number. The category has a precise mathematical definition: an even number is a multiple of 2. Any number that is a multiple of 2 is an even number; any number that isn't, isn't. No even number could possibly be “better” than any other, so prototype effects could never show up in an experiment.
Except they do. Using similar techniques as were used with birds and furniture, Armstrong, Gleitman and Gleitman found that some even numbers were judged as “better” than others. Of the six examples of even numbers offered up in the experiment — 4, 8, 10, 18, 34, 106 — 4 was judged to be the "best example" of an even number, followed by the others in rank order. How could this be?
Participants in the study were given the following instructions: “Don’t worry about why you feel that something is or isn’t a good example of the category. And don’t worry about whether it’s just you or people in general who feel that way. Just mark it the way you see it.” In other words, they were given permission not to think too hard about their choices. And so, some even numbers were judged as “better” than other even numbers — even though 100% of subjects responded “no” when asked, “Does it make sense to rate items in this category for degree of membership in the category?”
There must be something powerful about human categorization that allows us to take a crisp, simple category like even number and turn it into something fuzzy and complex. It seems that human beings are just really, truly, deeply hard-wired to see some category members as “better” than others.
Linguists like to use what we call “tests” to probe various aspects of language and conceptual structure. In the case of prototypes, we can use what’s called the but test. Consider the following sentences:
(1a) It’s a chair, but it’s shaped like a beanbag.
(1b) It’s a chair, but it has four legs.
I’m guessing that most readers will feel that (1a) is more natural than (1b). In both sentences, the use of chair, without any modifiers (what linguists call “unmarked”) calls to mind a prototype; but then tells us something unexpected is coming. In (1a), “shaped like a beanbag” is indeed unexpected; (1b) is odd because there’s nothing unexpected about a chair having four legs.
We can apply the but test to any category:
(2a) It’s a bird, but it doesn’t have feathers.
(2b) It’s a bird, but it has feathers.
(3a) It’s a tree, but it has several trunks.
(3b) It’s a tree, but it has one trunk.
(2a) and (3a) probably sound more natural than (2b) and (3b), because our prototypical bird has feathers, and our prototypical tree has one trunk.
What happens when we apply the but test to human categories? Let’s have a look:
(4) I have a primary care physician, but it’s a man.
How does that sentence read for you? If you were raised in a society in which your brain has been trained to think of doctors in mostly male terms, then the sentence probably strikes you as a bit odd.
One of the tricky aspects of digging into these questions is that it forces us to confront what is actually going on in our minds, as opposed to what we hope or wish is going on. If we’re honest about how we react to sentences like (4), we’ll quickly see that our brains are stuffed with stereotypes and unconscious biases. It can be upsetting to see, but we need to see it if we’re going to make meaningful progress.
As one last example of the but test, I invite you to fill in the blank with various types of human, and see how your brain reacts to each of them:
(5) That person is a _____ person, but they’re intelligent.
The privilege of being “normal”
Back in the early 2000s, I was in my dissertation adviser’s office, along with a fellow graduate student. The student — a 20-something Sri Lankan–American woman — made an offhand remark about white men, and followed it with, “You know. Prototypical people!” All three of us chuckled. I had the luxury of finding the comment amusing; I imagined that for my white female adviser and my woman of color colleague, there was some sting to go along with the amusement.
As we’ve seen, the human brain treats prototypical members of categories as “special” in a number of ways. When we think about this in human terms, the consequences are sobering.
Take the following result of Rosch’s experiments: that peripheral birds (like penguins) were judged as “more similar” to central birds (like robins) than vice versa. On the face of it, the finding makes no sense. Similarity is a symmetrical relationship: if A is similar to B, then B is equally similar to A.
But it doesn’t work that way in the brain, which needs reference points in order to understand categories. In the case of birds, we use robins and sparrows as a sort of anchor for the category: if I think something may or may not be a bird, I’ll compare it to something I know for sure is a bird, and I’ll use that to make my judgment.
This means that robins and sparrows establish a standard for the category: with respect to the category bird, robins and sparrows have a privileged status. Just as a dining chair has privileged status compared with a beanbag chair.
Applying this finding to humans reveals some difficult truths about how our minds work. Recalling my conversation with the professor and graduate student: the joke about white men as “prototypical people” makes light of the privileged status of white men vis-à-vis the category human. Sick and harmful and destructive as it is, the characteristics of (cisgender, heterosexual, physically able, etc.) white males have come to be used as a sort of yardstick against which other humans stack up more or less well.
At the end of the day, there are two truths that are impossible to reconcile.
First, humans judge members of categories as “better” or “worse” than others. It’s inescapable: we do this even with “even numbers.” We hold a vast quantity of unconscious ideas about how the world is divided up that are completely counter-logical and unscientific.
Second, to be a member of the category human requires nothing less than full membership for everyone. Each of us is fully human, full stop.
In the battle between these two truths, the one that wins most often, hands down, is the first one: humans can’t help seeing some members of all categories, including humans, as “better” than other members.
Being considered “normal” is the ultimate form of privilege, because it’s what confers 100% category membership — in this case, full humanity. By extension, anything other than “normal” is automatically lesser: something below fully human. Not actual people; something more like beanbag people.
As if that weren’t enough, when you’re “normal,” you don’t notice it. Adrienne Rich once wrote that “all privilege is ignorant at the core.” When things go our way, we tend to chalk up our results to our own talents, skills and hard work, and not to our environment. I experienced this in a deeply physical way once, during a running race. For a good part of the race, I was cruising along, faster than I thought I would, and feeling great about myself. Then the race hit a turnaround point, and I was blasted in the face by a forceful wind. Suddenly I didn’t feel so great about myself — and I slowed down.
This is why it’s hard, and therefore especially important, for “prototypical people” to continually force ourselves to look at how lucky circumstance has given us untold, and unearned, advantages, and to probe ever more deeply into the profoundly different experiences of those whose lives have been shaped by constant experiences of othering. Privilege and ignorance go hand in hand. That’s not an insult; it just is.
The work ahead
Nothing below 100% membership in the category human could ever be the basis of a just society. Yet, as we’ve seen, we are wired to see all categories in prototype terms. This means we have work to do. And by “we,” I mean the so-called “prototypical people.” Or, put in less binary terms: the more “prototypical” (privileged), the more work is required. That’s because it’s the more privileged folks who:
have less awareness of the issues at hand
benefit more from the status quo
control more of society’s levers of power, both to shape narratives and to change policies.
What is the specific nature of the work to be done? There are fields of scholarly and popular writing on the topic, plus untold workshops, surveys, audits, reform efforts — and on and on — devoted to answering this question. What I’ll highlight here are specific recommendations for mindset shifts and ongoing actions that more privileged people can take. These actions won’t directly counteract the mind’s need to use prototypes, but they can, over time, help mitigate some of the most harmful effects of prototyping:
Be honest about your mind’s workings. Internalized superiority is real. White supremacy is just one type of internalized superiority. Don’t try to wish them away; they’re not going anywhere. Recognize them for what they are: deeply ingrained stains and habits resulting from thousands upon thousands of years of human history.
Beware of self-positivity bias. Scores of research studies have shown that, when it comes to judging people’s actions, humans inevitably go easier on themselves than on others. It’s insidious. Be ready to take an honest look at your own behaviors.
At the same time, cut yourself some slack when it comes to the harsh workings of your unconscious mind. You can’t control where your unconscious mind will go. As we’ve seen above, stereotypes are everywhere in our minds. They just are. What matters is reducing the frequency of acting on our stereotypes. That’s why it’s so important to learn to recognize them; otherwise we can’t even begin learning how to interrupt them.
Remind yourself that none of this has anything to do with “being a good person.” Especially but not exclusively in U.S. culture, we’re taught that we should “be good people.” It’s hard to square our positive self-concept with the mind’s nasty tendency to generate disparaging thoughts towards others. Since those disparaging thoughts will always be there, and since we need some sort of positive self-concept, we need to learn to separate the truth that we can “be good people” from the fact that our minds are engaged in a lot of nasty judging.
We need to be careful not to keep centering ourselves when we talk about these topics, especially with more marginalized folks. Too often we’re so shocked or upset when we engage in this learning that this shock or upset becomes the focus of what we talk about, including with people with less privilege. We have to remind ourselves that this isn’t about the privileged folks and our struggles for self-awareness and self-betterment; it’s about the real human pain caused every day by systems and thought patterns burned deeply into our individual and collective psyches, and into our daily habits. More crucially, it’s about questioning, interrupting, and reforming these systems and thought patterns.
Stop thinking and talking about “racists,” or “sexists,” or any other type of person. Do focus on racism, sexism, homophobia, transphobia, ableism, and other forms of oppression. When we focus on a type of person, we’re stuck in a binary, either-or, in-or-out choice that doesn’t help us deal with the endless degrees of nuance that are required for having helpful, action-oriented conversations about these topics.
Think before you speak. Are you about to compliment a person of color, or a woman, for something they’ve done, or for some positive characteristic you believe they’ve demonstrated? Why are you complimenting them? Is it because part of your unconscious mind is surprised that someone “like them” would act in this positive way? If so, then it’s not a compliment; it’s a microaggression. It does real harm.
Do your own research. Don’t ask people of color to explain racism or tell you what you’re supposed to do and not do. The same goes for women and sexism, disabled people and ableism, and so on. To do so is to burden them with emotional labor on top of what they’re doing just to get through the day, every day.
Don’t think you’re some sort of hero. There is nothing heroic going on. If you want heroic, try living for one day without your privilege.
Don’t expect rewards. Don’t expect kudos. Don’t even expect to see concrete results of your actions.
Never stop learning and growing. There are way stations and epiphanies, but there is no final destination.
If this sounds preachy, I understand. I’m preaching to myself, first and foremost. In this work, I fail far more often than I succeed. You probably will too. If we truly long for a world where everyone belongs, where people like my associate no longer feel less than 100% human, and where all people are 100% people and not beanbag people, we have no choice but to keep going.