The Word That Turns People Into Insects

In the spring of 1994, in the hills of Rwanda, neighbors killed neighbors. People who had shared markets and churches and roads picked up machetes and used them on the families next door. Over roughly a hundred days, some 800,000 people were murdered, most of them Tutsi, most of them killed by hand, at close range, by people who knew them. ¹

The killers did not, for the most part, speak of murdering human beings. The radio station that helped orchestrate the slaughter, Radio Television Libre des Mille Collines, had spent months preparing the vocabulary. It called the Tutsi inyenzi: cockroaches. Insects to be crushed, vermin to be exterminated, a household infestation rather than a population of men, women, and children. ²

It is tempting to file this under the heading of pure evil, or ancient tribal hatred, or the peculiar madness of a particular place and time. But the mental operation underneath it is not exotic. It is not rare. It is a shortcut the human brain performs quietly, constantly, in ordinary life, long before anyone reaches for a weapon. The word dehumanize sounds like a term for history books. It is closer to a description of neuroscience.

A region of the brain that goes quiet

In 2006, the social neuroscientist Lasana Harris, then at Princeton, slid volunteers into a brain scanner and showed them photographs of other people. Ordinary faces, mostly: a businessman, a student, an older woman, a firefighter. As expected, a particular region of the brain lit up reliably each time. The medial prefrontal cortex sits behind the forehead and does a great deal of our social thinking. When we consider what another person might be feeling or intending, when we treat them as a mind rather than a body, this region reliably engages. Seeing a human face, for most of us, switches it on almost automatically. ³

Then Harris and his collaborator Susan Fiske showed the volunteers a different set of photographs: homeless people, and people described as drug addicts. For these images, the social region stayed dark. The medial prefrontal cortex did not engage the way it had for the businessman or the firefighter. The brains of the volunteers were, in effect, processing these faces the way they would process a chair or a pile of garbage. Not a person to be understood, but an object to be categorized. ⁴

Worse, another circuit did activate. The images of the most stigmatized groups triggered the insula, a region strongly associated with disgust, the same reaction the brain produces at the sight of overflowing sewage or rotting food. Fiske and Harris called this the failure to engage, a phrase deliberately clinical for a finding that is anything but. ⁴

What makes the result so unsettling is how mundane it was. These were not hardened ideologues or trained soldiers. They were undergraduates and ordinary adults, people who would describe themselves as decent, doing nothing more strenuous than looking at pictures. The dehumanizing response was not a decision they made. It was a reflex their brains performed before any decision was possible. The machinery that lets us see a stranger as a fellow mind can, under the right conditions, simply decline to switch on.

Where does such a reflex come from, and why would a social species evolve the capacity to un-see the humanity of its own members?

The distance that makes cruelty possible

Part of the answer had been sketched decades earlier, in a basement laboratory at Yale. In 1961, the psychologist Stanley Milgram ran an experiment that has haunted the field ever since. He recruited ordinary New Haven residents and told them they were assisting in a study of memory and learning. Their task was to administer electric shocks to a stranger in another room each time he answered a question wrong, increasing the voltage with every mistake, up to a switch labeled with an ominous 450 volts. ⁵

The stranger was an actor. The shocks were fake. But the volunteers did not know that, and the actor played his part convincingly, crying out, complaining of a heart condition, eventually falling silent. A man in a lab coat stood behind each volunteer and, whenever they hesitated, issued a calm instruction to continue. Milgram had asked colleagues to predict how many people would go all the way to the maximum shock. They guessed a tiny fraction, perhaps one in a hundred, and only the deeply disturbed. The real number was 65 percent. Nearly two-thirds of ordinary people delivered what they believed was a potentially lethal jolt to a stranger, simply because a mild authority figure told them to. ⁵

But the detail that matters most here is one that is often forgotten. Milgram ran many variations of the experiment, and one of the most revealing concerned distance. When the victim was in another room, unseen, a voice through a wall, obedience was high. When the volunteer was moved into the same room as the victim, obedience dropped. And when the volunteer was required to physically press the victim’s hand down onto a shock plate, to make cruelty a matter of touch, obedience collapsed to around 30 percent. ⁵

The closer the other person came, the harder it became to hurt them. Physical proximity forced the brain to register a full human presence, and that registration made the cruelty almost unbearable to carry out. Milgram had stumbled onto something that would echo through the neuroscience decades later: humanity is easiest to deny at a distance, easiest to preserve up close.

Two ways to strip a person down

If dehumanization is a switch, it turns out there are at least two different things it can turn off. In 2006, the same year as Harris’s scanner study, the Australian psychologist Nick Haslam proposed that we deny other people their humanity in two distinct ways, by withholding two different sets of traits. ⁶

One set Haslam called human nature: the qualities of warmth, emotion, curiosity, and vitality that we share with other living things. Deny these, and a person becomes cold, rigid, mechanical, empty inside. This is mechanistic dehumanization, the reduction of a human being to something like a machine or an automaton. It is the way a bureaucracy can come to treat the people in its files, or the way a torturer can regard a prisoner as a problem to be processed rather than a person in pain. ⁶

The other set Haslam called uniquely human traits: refinement, rationality, moral sensibility, the civility that we imagine separates us from animals. Deny these, and a person becomes a beast, driven by instinct, dirty, savage, incapable of self-control. This is animalistic dehumanization, and it is the older and cruder of the two. ⁶

History, unfortunately, offers an endless catalogue of the animalistic form, because propaganda reaches for it almost by instinct. The vocabulary is startlingly consistent across atrocities separated by centuries and continents. Rats. Vermin. Cockroaches. Apes. In the propaganda of Nazi Germany, Jewish people were drawn as rats swarming through Europe, carriers of disease to be exterminated for public hygiene. In the legal architecture of transatlantic slavery, whole peoples were classified not as persons but as property, chattel to be bought and inventoried. The Rwandan radio’s inyenzi was not an original insult. It was the reappearance of an ancient one. ²

The pattern is so reliable that some scholars have argued language is the leading edge of violence, the first weapon deployed, with physical harm following once the words have done their preparatory work. Once a group has been successfully re-described as insects or vermin, the ordinary moral machinery that would object to their killing has already been switched off.

Blatant dehumanization, in broad daylight

For a long time, researchers assumed that in the modern world, dehumanization had gone underground. Overt racism had become socially unacceptable in many places, so surely the crude business of ranking other groups as subhuman had retreated into implicit bias, something people harbored without quite admitting it even to themselves.

Then the psychologist Nour Kteily, working at Northwestern University, built a test so blunt it seemed almost too simple to work. In a 2015 paper, he presented participants with the famous “March of Progress” illustration, the familiar image of a hunched ape gradually straightening into an upright modern human across a series of silhouettes. He labeled the ape end 0 and the fully upright human end 100. Then he asked people, plainly, to rate how evolved various human groups were, sliding them along the scale. ⁷

One might expect participants to refuse the premise, or to place every group at 100 out of embarrassment. Many did not. Confronted with a simple slider, sizable numbers of people openly rated other national, religious, and ethnic groups as less than fully human, placing them meaningfully below the top of the scale. Kteily called this blatant dehumanization, and the word blatant was the point. This was not a hidden bias teased out by clever reaction-time measures. It was an explicit judgment that people were willing to make in the open, on a form, with no apparent shame. ⁷

And the scores did work. Where people placed an out-group on the ape-to-human scale predicted real attitudes with disturbing reliability. Higher dehumanization predicted greater support for aggressive drone strikes, for harsh interrogation and torture, for punitive immigration policy. The slider was not measuring an abstraction. It was measuring a willingness to see other people harmed. ⁷

The excuse, not the cause

Here the story turns, and turns in a direction that inverts the intuition most of us carry. It is natural to assume that dehumanization causes violence: first you come to see a group as less than human, and then, freed from moral restraint, you harm them. The dehumanizing belief is the accelerant, and cruelty is the fire that follows.

But a growing body of work suggests the arrow often runs the other way. Frequently, the desire to harm, or the fact of having already harmed, comes first. The dehumanizing story arrives afterward, as justification. We do not always dehumanize a group and then mistreat them. Sometimes we mistreat them, or benefit from their mistreatment, and then reach for a story that makes it bearable. The narrative that the victim was less than fully human is a way of quieting the conscience after the fact, of being able to sleep. Cruelty writes its own alibi. ⁸

This reframing matters enormously, because it changes what dehumanization actually is. It is not only a cause of atrocity, sitting upstream where we might hope to intercept it. It is also a psychological convenience, a tool the mind employs to protect its own image of itself as good. If we have done something monstrous, or looked away while others did, insisting that the victims were vermin lets us remain, in our own eyes, decent people who acted reasonably. The dehumanizing belief is less a fixed conviction than a story we tell to preserve our self-regard.

Turning the switch back on

If that sounds bleak, it also points, unexpectedly, toward a remedy. A switch that can be turned off can be turned back on.

Recall Milgram’s finding about distance: cruelty grew harder as the victim came closer, until physical proximity made it nearly impossible for most people. That principle scales up. In 1954, the Harvard psychologist Gordon Allport proposed what became known as the contact hypothesis: prejudice between groups tends to diminish when their members come into genuine contact with one another, provided certain conditions hold. The contact must occur between equals, in pursuit of a shared goal, with some measure of institutional support, rather than as a fleeting or hostile encounter. ⁹

Allport’s idea could have remained a hopeful hunch. Instead, it became one of the most heavily tested propositions in social psychology. In 2006, a comprehensive review by Thomas Pettigrew and Linda Tropp pooled hundreds of studies across many countries and forms of prejudice. The pattern held. Genuine contact, under reasonable conditions, reliably reduced prejudice, and the effect appeared across cultures, age groups, and target groups. The reduction was not merely a matter of learning new facts about the other side. Contact worked in part by reducing anxiety and increasing empathy, by making the abstract, dehumanized “them” resolve back into particular people with faces and histories. ¹⁰

This is the quiet counterweight to the machete and the radio. The same brain that can decline to see a homeless man as a person can, given proximity and equality and a shared task, be brought to see him as one again. Rehumanization is not a soft platitude. It is a measurable psychological process, and it begins with the reappearance of a single face where an abstraction used to be.

Coda

The machinery described here is not the property of monsters. It ran in undergraduates looking at photographs. It ran in ordinary New Haven residents obeying a man in a lab coat. It runs, in small everyday doses, whenever a group of people is compressed into a label and the mind stops bothering to imagine what is behind their eyes. The genocide is the far end of a road that begins with a slur, a joke, a glance that says you do not count. The most dangerous word in any language may be the smallest and most innocent-looking one: they. It is worth pausing, the next time a group of human beings has quietly become a they, to ask who taught us to stop seeing a person, and whether the switch might be turned the other way.

Sources

1. Human Rights Watch, Leave None to Tell the Story: Genocide in Rwanda, 1999. — https://www.hrw.org/reports/1999/rwanda/
2. Yanagizawa-Drott, D., “Propaganda and Conflict: Evidence from the Rwandan Genocide,” Quarterly Journal of Economics, 2014. — https://doi.org/10.1093/qje/qju020
3. Amodio, D. M. & Frith, C. D., “Meeting of minds: the medial frontal cortex and social cognition,” Nature Reviews Neuroscience, 2006. — https://doi.org/10.1038/nrn1884
4. Harris, L. T. & Fiske, S. T., “Dehumanizing the Lowest of the Low: Neuroimaging Responses to Extreme Out-Groups,” Psychological Science, 2006. — https://doi.org/10.1111/j.1467-9280.2006.01793.x
5. Milgram, S., “Behavioral Study of Obedience,” Journal of Abnormal and Social Psychology, 1963. — https://doi.org/10.1037/h0040525
6. Haslam, N., “Dehumanization: An Integrative Review,” Personality and Social Psychology Review, 2006. — https://doi.org/10.1207/s15327957pspr1003_4
7. Kteily, N. et al., “The Ascent of Man: Theoretical and Empirical Evidence for Blatant Dehumanization,” Journal of Personality and Social Psychology, 2015. — https://doi.org/10.1037/pspp0000048
8. Rai, T. S., Valdesolo, P. & Graham, J., “Dehumanization increases instrumental violence, but not moral violence,” PNAS, 2017. — https://doi.org/10.1073/pnas.1705238114
9. Allport, G. W., The Nature of Prejudice, Addison-Wesley, 1954. — https://archive.org/details/natureofprejudic0000allp
10. Pettigrew, T. F. & Tropp, L. R., “A Meta-Analytic Test of Intergroup Contact Theory,” Journal of Personality and Social Psychology, 2006. — https://doi.org/10.1037/0022-3514.90.5.751

Related reading

• The Line Test That Broke Our Faith in the Individual
• What the Marshmallow Really Measured

More from the Mind edition →