Most people who behave cruelly online do not think of themselves as cruel, and in fact may not be especially so. Subjectively, they may be speaking from a sense of conviction, righteousness, or simply what feels like playful competition.
Give Us This Day Our Daily Dread
As a psychiatrist and psychoanalyst active online and in my work, I have spent years watching and being directly impacted by how ordinary people account for harm they cause—usually, they feel they are in the right. Often, they believe they are doing what is best. Sometimes, there is a tinge of enjoyment, especially if justified and rationalized ethically and morally. Some days, going online feels like a risky proposition, particularly if one of the more persistent low-level proto-trolls has latched on to some idea. How to handle that, I’ve discussed in my work on BRODA, a common set of moves on socials: bait, refuse open dialogue, attack. There is often a theme of the alleged troll expressing being victimized, falsely accused, shifting to a personal register, flipping blame (DARVO), and the like.
Wrongdoing more generally was once imagined to be the work of a few deviants, evil people. After WWII, when perfectly lovely neighbors turned a blind eye, or worse, the explanation became less clear, more troubling, and easy to ignore. More people than we’d like to imagine, under conditions like anonymity, can slip into everyday evil: given a crowd, perceived distance from the target, and a cause that makes the attack feel like justice. Famous mid-20th-century experiments paint a clearer portrait—Milgram’s obedience studies, Zimbardo’s prison simulation, contested as they now are, pointed at the same uncomfortable thing. The numbers do as well. This type of behavior is context-dependent, and people who think they’d never act so are often surprised, or they don’t notice… even when confronted with the facts.
Pathological narcissism is relatively uncommon, hovering between 1% and 6% of people, depending on the study. The darker traits it travels with—manipulation, callousness, the appetite for others’ pain—are more rare, shading off into a thin tail rather than marking out a population. Even granting the occasional claim that a third of online trolls score high on those traits, the sheer volume of everyday cruelty online isn’t exclusively coming from those folks.
Becoming a Bully
A consistent pattern of behaving this way rarely arrives all at once. It sets in by degrees, the way some diseases do—a symptom here, easily explained away, then another, until one day the realization breaches. A history of being victimized, perhaps; personality traits like narcissism and neuroticism, not fully set but predisposing; the more anonymous, detached perch social media affords, and the vigilante mentality that comes with it; a resentful alter ego discharging retaliatory impulses repressed elsewhere in a life. There are many factors—though I’ve watched a handful of people online earnestly argue that bullying is good for you, that it builds toughness, and then bully instrumentally on that logic.
With insidious onset, insight fails early and under the radar. Someone who knew, in the moment, that they were being cruel might simply stop—though sometimes there is a callousness cloaked in the more stoic philosophies. This lighter bullying can’t stop because there’s nothing to stop, only a series of reasonable responses to people who deserved it.
There are reasons this happens, and a measure of compassion is useful. Harm travels down the generations; people who were hurt are more likely to hurt, and the pattern repeats through blindness more than decision. But the same research that shows the cycle shows it isn’t a given—most of those treated badly do not go on to perpetrate. Etiology is not destiny. What lets the pattern hide is moral disengagement: the running commentary that reframes each cruelty as fairness, accountability, a joke, a defense. Compassion doesn’t excuse the behavior, and neither does clinically significant trauma—but compassion, for oneself and others, is what makes it possible to challenge hurtful norms effectively.
The genuine sadist—the true troll—acts with eyes wide open. No slow onset, no failure of insight, no symptom ignored. They know exactly what they’re doing and enjoy it, and innocence is useful camouflage: a wolf in the proverbial sheep’s clothing.
Five Questions for Self-Appraisal
Drawing on the literature below, five self-reflective questions stand out for spotting whether bullying—or near-bullying—behaviors have crept in. They borrow from tested rating scales but are not themselves a validated instrument.¹ Worth noting before you start: when these patterns get named mid-argument, things can get spicy. Met with a sincere “I feel like you’re crossing a line, please stop,” some back off and some advance. And defensiveness, it should be said, is not by itself an admission of guilt.
- The Power Check: Do I pile on smaller accounts because it’s consequence-free—or go after larger ones to self-aggrandize, especially when I sense a vulnerability or have an “injustice” to justify the attack?
- The Harassment Test: Do I repeatedly insult, mock, or DM the same person, across one account or several? Do I repost to humiliate on a snap judgment, without checking whether it’s warranted?
- The Mob Test: Do I use my platform or group chats to coordinate dogpiles, mass-reporting, or waves of hate—rationalized by a sense that the target “deserves it”? Do I belong to a group that praises this?
- The Mask Test: Do I use burners or anon profiles to say what I’d never attach to my name or say to a face? Does it feel like sniping from cover—with some guilty pleasure, an aliquot of schadenfreude?
- The Payoff Test: When my comments upset, silence, or drive someone off a platform, do I feel satisfaction, entertainment, or “they had it coming”? Do I double down, go in for the kill, or dial it back?
What to do?
Hopefully, these types of questions are useful, if not fully welcomed. I shared a few online, and got mixed responses. Who would answer these questions honestly? Interestingly, research has found that at least with narcissism, people are likely to disclose being narcissistic based on a single question: “To what extent do you agree with this statement: ‘I am a narcissist’?” (Konrath et al., 2014).
Next steps, if anything rings a meaningful bell, are to collect more information, observing one’s own behavior and asking safe and trusted loved ones, and do one’s own research. These five questions are not definitive or diagnostic, but can show up yellow (or even red) flags. People can and do change these patterns, and often feel better about themselves and enjoy healthier relationships. Sometimes the assistance of a professional can be useful, or even necessary, but many things we can shift with self-awareness and consistency.
