All writing

The Adult in the Room · Part 1 of 8

What Did He Do?

The first question everyone asks about a bullied colleague. Four decades of research says it's the wrong one.

· 7 min read

Sources 3

This series follows one account, told across eight pieces. Names, places, and identifying details have been removed or changed throughout.

Here is a man as people described him in the summer of 2019. Twenty-two years old, three months into a new town and a few weeks into his first real job. Optimistic. A little mysterious, in the way a quiet newcomer often seems. Easy enough to get along with. He made friends fast, the way you do when you don't yet know which conversations cost something.

Here is the same man as people described him toward the end. Guarded. Hard to read. Keeps to himself. Somebody who had, one supposes, gotten a reputation.

Seven years apart. Same man. The only thing that happened in between was what other people did to him.

That sentence is easy to nod at and very hard to actually believe, because the moment you meet someone in year six, your mind does something automatic and almost impossible to interrupt. You look at the guarded man and you think, without ever choosing to think it: well, I can see it.

You are looking at a result and reading it as a cause. Nearly everyone does. And the reason this matters more than any other single thing in this series is that it is the engine that keeps the whole machine running.

Tell someone at a party that a coworker of yours was hounded out of a job. Watch what they ask.

They will not ask who did it. They will not ask what the company did about it. Almost without exception, the first question out of a decent person's mouth is some version of:

"What did he do?"

Not because they're cruel. Because they're frightened. And because we were all taught the same story about how this works, at seven years old, from someone who meant well: bullies pick on the weak ones. The kid with the inhaler. The kid who cried. There was always, in the telling, some quality in the target that explained the targeting, and the quiet promise underneath was that if you weren't that kid, it wouldn't be you.

We carry that promise into adulthood without ever taking it out and looking at it.

So let's look at it.

In 2009, researchers studying a group of 435 healthcare employees set out to find the bullying target's personality. They had 42 identified targets in the sample, and they measured all five of the standard personality dimensions using the NEO-FFI, then ran logistic regression and cluster analysis to see what separated the bullied from everyone else.

They found a difference on two dimensions out of five.

Targets scored higher on conscientiousness and lower on agreeableness.

Those two words are worth unpacking, and worth being careful with. High conscientiousness describes someone who does the work carefully and notices when a corner has been cut. Low agreeableness is the more ambiguous of the pair. On this inventory it picks up candor and an unwillingness to smooth things over, and it also picks up suspicion and being hard to get along with. Taking only the flattering half would be easy, and it is precisely what the popular version of this finding does, arriving on LinkedIn as bullies target the best among us. That is further than the evidence goes.

The authors themselves went the other way. Their cluster analysis, looking for a distinctive type of victim, found none, and their conclusion was that personality patterns do not easily distinguish targets from non-targets, and that single-factor explanations of bullying are unlikely to be sufficient.

So the honest summary of the strongest evidence on who gets bullied is this: there is no victim type. Not a weak one, and not a heroic one either.

And there is a further problem with reading anything into those two traits, which the study cannot solve and which is about to become the entire point. This research was cross-sectional. It measured 42 targets at a single moment, after the bullying. Which means the traits it found may not describe who those people were before any of it began.

Now the part that reorganizes everything.

Last year a team in Sweden ran the question the right way round. Instead of finding bullied people and inspecting their personalities after the fact, which is what nearly every earlier study had done, they followed 2,024 employees across two waves seven months apart, measuring bullying both behaviorally, through the 22-item Negative Acts Questionnaire, and by self-labeling, then modeling the whole thing with cross-lagged structural equation modeling so they could see which way the arrows actually pointed.

Both directions were real. That's the honest headline and I'm not going to bury it. Neuroticism did predict some later exposure to bullying, working through interpersonal conflict and emotional distress. Anyone telling you that target characteristics are pure myth is overselling.

But here is the finding that should change how you read every bullied person you will ever meet:

Exposure to bullying predicted changes in personality. Being bullied moved people on neuroticism and conscientiousness over the study period, and it did so specifically through the accumulated weight of stress, anxiety, and depressive symptoms. The authors describe personality not as a fixed risk factor a person carries into the room, but as a dynamic system that the social environment acts upon.

Which means the anxious, brittle, hard-to-read version of a person is, in part, manufactured by the treatment, and then read backward as its cause.

This is the trap, and it is nearly perfect. By the time you meet the target, the evidence for "well, I can see why" is genuinely there. It's on their face. It's in the way they answer a simple question a half-second too carefully. You are not imagining it and you are not a bad person for noticing it. You are simply arriving at the end of a process and mistaking its wreckage for its origin.

The bullying builds its own alibi.

There is a second reason the question comes so easily, and it has nothing to do with the target at all.

Psychologists have spent decades on what happens when people encounter undeserved suffering. The mind's cheapest available move is to locate a reason inside the person it happened to, because the alternative is unbearable in a specific way: if it happened for no reason, it can happen to you. What did he do is not really a question about him. It's a request for reassurance, addressed to the universe, phrased as curiosity.

And it works. That's the miserable part. Ask it and you feel better. Ask it and the guarded man in year six becomes a man with a reputation instead of a man with a history, and the machine that made him gets to keep running unexamined, because everyone in the building has already privately explained him to themselves.

Which is where I keep landing as a matter of what we actually teach children.

We tell them bullies pick on the weak. We mean it kindly. We are trying to say this isn't about you, there's something wrong with the person doing it, and we do not notice that we've handed them a sentence with a trapdoor in it. Because the child hears the other half too. If bullies pick on the weak, and you are being picked on, then the arithmetic runs in exactly one direction, and no adult ever has to say it out loud for a nine-year-old to work it out.

That's where the shame is installed. Not by the bully. By the explanation.

And it lasts. It follows a person through a door on their first day at twenty-two and it's still there at twenty-nine, in the form of a question they have quietly asked themselves ten thousand times, which is the same question the party guest asked, turned inward: what did I do?

The available evidence says: probably, mostly, nothing. Possibly you were careful about the work and unwilling to be smooth about it.

So before the next series of these, one question worth carrying around.

Think of the last difficult person you worked with. The prickly one, the guarded one, the one everybody had a shorthand for.

How certain are you that you met the cause, and not the result?

Sources

Identifying details in the opening account have been removed or changed.

More writing ›