The Body Knew Before the Internet Did

She came in on a Tuesday with the kind of breath people have when they have not been sleeping. A woman in her early thirties, mid-career, the sort of patient who reads the research before the first session and arrives knowing the clinical names for what is wrong with her. She had spent the previous night sitting upright on the edge of her bed, running through a single sentence she had typed three years earlier in a private group chat, screenshotted the week before by someone she had once thought of as a friend, now circulating. Her hands were cold. Her pulse, she said, would not come down. She kept opening her phone and then closing it and then opening it again, the way patients used to describe checking the stove. She told me she had not been able to eat solid food in four days. She was not, she said quickly, afraid of dying. She was afraid of being seen, and of what the people who had not yet seen the screenshot were about to think when they did, and she could feel that fear in her chest the way other patients feel a cardiac event: a tightness with weight behind it, a sense that something had already happened to her body before anything had actually happened in the world.

Clinicians have a name for what she was describing. Robert L. Leahy, who has spent decades writing about anxiety from a cognitive perspective, calls it cancel anxiety, and over the past several years it has become a recognizable presentation in offices across the country. The shape of it is consistent. A person becomes preoccupied with a past statement, a present opinion, an ambiguous photograph, a relationship they used to have, and the preoccupation does not behave like ordinary social anxiety. It is sticky, it is somatic, it interrupts sleep and appetite, and it produces the specific physiology of a survival threat in a person who knows intellectually that she is sitting in a heated apartment with food in the refrigerator. Therapists who treat it tend to notice the same thing I noticed with my patient that Tuesday. She was not being dramatic. Her body was responding accurately to what her body believed was happening. The question is what her body believed, and why.

Here is one way to think about it. For something like ninety-five percent of the time anatomically modern humans have existed, the question of whether other people approved of you was not a question about your feelings; it was a question about your continued physical existence. Hunter-gatherer bands ran on reputation the way modern economies run on credit. If the group decided you were unreliable, dishonest, dangerous, or simply not worth the trouble, the consequences were not social in any sense we would now recognize as social. Ostracism, even partial ostracism, meant losing access to shared food, shared shelter, shared defense against predators and rival bands, and shared reproduction. Kipling Williams, the social psychologist whose ostracism research has become foundational, has shown that even brief, trivial exclusion in laboratory settings produces measurable distress, drops in mood, and a felt sense of threat. He found this with a simple ball-tossing game played on a computer against strangers the participant would never meet. The threat circuit does not need much to activate. Naomi Eisenberger and her colleagues showed why, in a now-classic series of fMRI studies, that the same neural regions which light up under physical pain also light up under social rejection. Social pain, in the literal neural sense, is pain. The brain does not file ostracism in some abstract emotional folder. It files it next to the broken arm.

This is the frame anyone trying to make sense of cancel anxiety has to start with. The patient sitting upright at three in the morning re-reading a private message from 2021 is not having an irrational reaction to a minor social event. She is having a calibrated reaction to a survival signal her nervous system is reading correctly. The signal is real. What has changed is the environment in which the signal is being delivered.

Recent evolutionary work on shame helps explain the precision of the response. Daniel Sznycer and his collaborators, in a series of cross-cultural studies published in 2016 and 2018, have argued that shame is not a malfunction or an evolutionary leftover. It is, in their model, a sophisticated regulatory system whose specific function is to minimize the spread and impact of reputational damage. Shame, in this account, behaves like an internal accountant. It tracks information about the self that, if known, would lower the value other people place on you, and it generates the urgent, motivational state required to suppress that information, repair the breach, or, if neither is possible, withdraw before the damage propagates. One way Sznycer's model gets described, and the way that landed with me when I first read it, is that the shame system runs simulations. If people found out about this, how would they downgrade me, how many of them, how badly, how long. The wave of heat in the face, the urge to disappear, the loop of self-narration that will not stop, these are the felt edge of an estimation engine that is doing math about reputational survival in the background, whether or not the conscious mind is invited to the meeting.

These systems, ostracism detection and shame, evolved together for a particular kind of social world. The relevant world was a band of perhaps a hundred and fifty people, knit by face-to-face contact, where information spread slowly and stayed local, where a wrong move could be apologized for to the person actually wronged, where reputations were built and rebuilt across years and were known in their full context. Almost nothing about that world resembles the world in which our reputation-tracking circuits are now being asked to operate. Information now spreads to millions of people in hours. Audiences are anonymous, dispersed, and largely strangers. Context collapses, so a sentence written for one group, in one mood, on one day lands in front of an audience that did not see the conversation it answered. The past is retrievable. A line typed at twenty-two is still available at thirty-six. The shame system, which evolved to track what the people in your hut think of you over the season, is being asked to track what an unbounded global audience might think of an unbounded set of past artifacts forever. It cannot calibrate. It activates as if survival were at stake, because, in the only terms it understands, survival is at stake.

This is where the individual problem and the cultural problem turn out to be the same problem at different magnitudes. Cancel anxiety in a single patient and cancel practice as a public phenomenon are not separate things linked by a metaphor. They are the same evolutionary circuit, running in the same nervous system, firing at the scale of one body and at the scale of a crowd. The crowd, when it gathers around a person to expel them, is using ancient machinery for sorting members from non-members. The individual, when she lies awake at three in the morning, is using ancient machinery for predicting the crowd. That is why the fear is rising in the same period in which the practice has expanded. It would be strange if it were not. The phenomenon also crosses every political line you could draw. A devout religious community can cast someone out for the wrong question. A progressive workplace can cast someone out for the wrong sentence. A small subculture can cast someone out for breaking its private codes. A friend group can do it over dinner. The circuit does not care about ideology. It cares about belonging, and it will use whatever symbolic vocabulary the surrounding group provides.

Treatment is where this gets clinically harder than it tends to look in a manual. There is real and good work to do at the level of the individual nervous system. Cognitive behavioral approaches can help a patient see the catastrophic predictions her shame system is generating and weigh them against actual probabilities. Exposure work can help her tolerate the bodily sensations of being seen and judged without collapsing into avoidance. Acceptance and commitment therapy can help her stay in contact with what she values when the threat signal is loud. Attachment-informed work, which is most of what we do at Unscarred, can help a patient understand why her particular nervous system reads social uncertainty as annihilation, and can begin the slow process of building a felt sense of safety inside her own body that does not depend entirely on the crowd's vote. All of this is useful. None of it is the whole truth.

The whole truth is that the standard move of telling a frightened patient that her fear is irrational, that the threat she perceives is not real, that she is catastrophizing, does not work here, and trying to make it work risks something close to gaslighting. In the current environment, the fear is partly accurate. One post, one screenshot, one out-of-context clip, one sentence spoken in a private setting and surfaced in a public one, can in fact end a career, a friendship, a marriage, a community membership built over a decade. People know this. The patients we are seeing know this. Pretending otherwise does not soothe them; it tells them their clinician has not been paying attention. A treatment frame honest enough to be useful has to start by conceding that the social threat is, at this moment in history, partially real, and then ask the harder question of how a person builds a life that is not run, second by second, by the prediction of that threat. That is a different question than the one our standard protocols are designed for, and it deserves a more honest answer than the one our standard protocols tend to give.

That harder question opens out onto a cultural one we cannot ethically pretend is somebody else's department. Individual treatment, however good, cannot fix a culture that systematically activates the threat. We can teach a patient to regulate her response to a social environment that is producing more cancel anxiety every year, but the regulation work compounds against an incoming tide. We treat the symptom in the office while the environment keeps producing more of it. There is something demoralizing, both clinically and morally, about a situation in which the most we can offer is better coping inside a system that does not need to be this cruel.

A culture that took this seriously would do specific things, and it would do them across political lines, because no faction has been innocent here. It would build proportionality between mistake and consequence. The same response is not appropriate for a teenager's clumsy joke, an adult's careless tweet, a pattern of harassment, and a serious crime, and a healthy social order makes those distinctions on purpose rather than letting the algorithm flatten them into one undifferentiated category of “bad person.” It would leave room for repair and for growth. Most ethical and religious traditions humans have built across the last several thousand years contain the assumption that a person can do wrong, be held accountable, change, and be re-admitted. Permanent expulsion as the default outcome for a wide range of offenses is a historically strange position to take, and it is not the position our nervous systems are built for. A culture that took this seriously would also stop using public shaming as a substitute for the harder work of actual accountability. Shaming is fast. It is satisfying. It allows the people doing it to feel righteous without doing anything else. Accountability, in any tradition that has thought carefully about it, involves the person who caused harm understanding what they did, the person harmed having a real say, a process of repair when repair is possible, and a reintegration that does not pretend the original offense did not happen but also does not require permanent banishment. We do not currently have a public version of that process. We have an outrage cycle, and we are using it for jobs it cannot do.

None of this argues that accountability should be soft, or that there are no behaviors which warrant serious and lasting consequences, or that people harmed in private by powerful figures should go on being harmed in private. It argues, more narrowly, that a healthy culture distinguishes between the kinds of wrong it is responding to, and that one of the costs of failing to make those distinctions is the rising tide of patients I and my colleagues are now seeing, sitting upright on the edge of the bed at three in the morning, in bodies that have correctly detected a threat that did not need to exist at this magnitude.

What we can do in a therapy room is real. The nervous system can be taught to recover. The catastrophizing simulations can be slowed. The sense that one's worth depends entirely on the crowd's verdict can be loosened, gently, in the company of someone who is not running that simulation about you. For patients whose cancel anxiety is layered onto an anxious or fearful-avoidant attachment pattern, which is most of them, the work of Anxious Healing or Fearful-Avoidant is exactly the work this fear requires: a slow, embodied repair of the part of the system that learned, long before the internet existed, that being seen badly by the people who matter is the same as not surviving. That work is worth doing for its own sake. It is also, in this moment, the closest thing many people will have to a private antidote for a public problem we have not yet collectively decided to solve.

The essay you are reading is an argument that we should decide to solve it, and that the clinicians who are watching the cost of not solving it have a particular responsibility to say so out loud, in their own voice, in the room and beyond it.