G7: Moral Panic and the Folk Devil

Two headlines

Suppose you read two headlines on the same morning. Both are invented for this page.

The first says that a new kind of event at the town library, where performers in drag read stories to young children, is putting those children at risk. The second says that a new kind of online forum, popular with teenage boys, is turning them toward extremist violence.

Notice your first reaction to each. Did one strike you as an obvious overreaction, the kind of story written every few years about whatever is new, and the other as a real problem that people are failing to take seriously?

Now ask both a narrower question. How many children would have to be harmed before the headline was proportionate? Out of how many? Compared with what other risks? If you didn't think to ask those questions of the headline you agreed with, you have just met this grammar in yourself.

There is a second test. If you reached for the phrase "moral panic" to dismiss one headline, did you reach for it for both? The phrase is a useful tool, and also a way of ending a conversation, and it is easy to aim it only at the other side's alarms.

What the grammar is

A social problem gets pinned on a specific group or practice, and the intensity of public alarm comes loose from the actual scale of harm. What makes it this grammar is not that the alarm is wrong, since sometimes it isn't, but that its loudness no longer tracks the size of the problem.

This site calls a recurring pattern of argument like this a grammar: a shape the same mistake takes across many different subjects, so that once you can see it in one argument you can see it in others.

The sociologist Stanley Cohen described the classic form of this one in Folk Devils and Moral Panics (1972). In outline, a deviant figure is identified, the "folk devil" of the title; the media amplify the alarm; claims inflate beyond the evidence; campaigners take up the cause, the people the sociologist Howard Becker called "moral entrepreneurs" in Outsiders (1963); and eventually the panic fades away, usually without anyone going back to check whether the original claims were true. This site's word for that fading is that the panic composts: it rots quietly into the background and feeds the soil for the next one.

Source still being found. the stages of the cycle as Cohen himself set them out in Folk Devils and Moral Panics (1972), quoted from the book, including how he drew on Becker's "moral entrepreneurs"

The newer wrinkle is that every scare tends to arrive with its debunking attached: a counter-panic saying "your panic is the real panic," which can be just as disconnected from the scale of harm, in the other direction.

Source still being found. evidence that counter-panics now launch at the same time as the panics they answer, rather than arriving later

The engine of both is slippery-slope reasoning. The panic says that today's oddity is tomorrow's epidemic. The counter-panic says that no warning of this kind has ever come true. Neither checks how often this kind of slope has actually been slid down.

The grammar has two poles, and each is a real failure.

Pole A, the panickers, treat a vivid story as a measure of size. One terrible case, told well, stands in for a trend; laws get written around anecdotes; and the people accused, whether a subculture, a profession or a kind of teenager, are harmed faster than evidence about them can accumulate. Their characteristic blindness is to effect size: they ask whether something can cause harm, and never how much.

Pole B, the panic-callers, treat the label "moral panic" as a verdict. They are right that many alarms in the past turned out to be overblown. They are wrong when they conclude that every alarm is, because some alarms were not panics at all, and the warnings about tobacco proved right. Their characteristic failure is premature dismissal, using the phrase to end an inquiry rather than to discipline it.

The clearest documented case is tobacco. In 1958 the statistician R. A. Fisher, who had spent much of his career on "how experimentation should be carried out," published "Cigarettes, Cancer, and Statistics". He noted that no investigator could produce evidence "in which a thousand children of teen age have been laid under a ban that they shall never smoke, and a thousand more chosen at random" were made to smoke. From the different trends in lung cancer among men and women, he argued that there was "no reasonable ground at all" to tie the disease's rise to smoking. Of that rise he wrote: "It is frightening. But it shouldn't be used to frighten people." He also called giving up smoking as insurance against an unknown danger "a perfectly rational attitude." Six years later, the 1964 Surgeon General's report, as the National Library of Medicine summarizes it, "estimated that average smokers had a nine- to ten-fold risk of developing lung cancer compared to non-smokers." A master of experimental method treated an experiment that could never be run as leaving the question open, and called a true warning a fright.

Source still being found. documented cases in which warnings about lead or opioids were dismissed at the time as panic or alarmism, with dates and quotations

Taking a panic apart

The grammar is undone not by deciding whether an alarm is a panic, which is the question each pole has already answered, but by asking smaller questions that both poles skip.

How big is the harm, and for whom? Not whether a harm exists, but how many people it affects, how badly, and whether it falls on everyone or on a few. A serious harm to a few and a slight harm to millions are different problems, and both differ from a harm that is mostly imagined.

Compared with what? A risk that sounds frightening on its own may be smaller than ones people accept every day, or larger. The comparison is where the judgment happens, and it is usually left out.

Who is the folk devil, and what happens to them? Every panic has a target, and the target pays a price whether or not the panic was justified. Ask what is being done to them, and whether it is being done faster than the evidence warrants.

What did the alarm predict, and did it happen? A panic almost always forecasts spread. A forecast can be checked later, and so can the counter-panic's forecast that nothing will happen.

Who is checking? Asking, years later, what became of the original claims is the step almost nobody takes, and the one that would teach the most.

None of these questions tells you in advance which way the answer will fall. A question that always ends with "it's a panic," or always with "it's real," is not a question.

Why it is so persistent

The grammar persists partly because stories travel faster than numbers. One tragedy with a name and a photograph reaches more people than a table of rates. The name for what follows is an availability cascade: a belief gains plausibility simply by being repeated, because what comes easily to mind seems likely.

Source still being found. the origin and research basis of the term "availability cascade," read in the original work

It persists because effect sizes are hard to feel. "Harmful" and "harmless" are easy to grasp; a small but real effect, weighed against everything else that affects the same outcome, is not. Without that sense, the choice collapses into a switch: either the thing is dangerous or the alarm is hysteria.

It persists because panics are useful to someone. In Cohen's outline, alarm becomes a profession, with people whose standing or audience depends on the problem staying large. The counter-panic has its professionals too, whose audience depends on every alarm turning out to be foolish.

Source still being found. research on how outrage, on either side, is rewarded by media and platform incentives

And it persists because the panic composts without an audit. When nobody grades the original claims, nobody learns how often this kind of alarm was right, and the next one arrives to a public with no memory of the last, or a memory edited to suit its side.

How the argument has been had before

This page traces the pattern through five earlier alarms: comic books, the role-playing game Dungeons & Dragons, the 1980s panic over satanic abuse in daycare centers, video games, and sexting. In this working draft each is a sketch of what the research should check: what the alarm got right and wrong, and what its dismissers got right and wrong in turn.

Comic books. The alarm was that comics were corrupting the young. The history should show what harm was claimed, on what evidence, and what was done to the industry and its readers as a result.

Source still being found. a sourced, dated account of the comic-book alarm, including what its leading advocates claimed in their own words and what action followed

Dungeons & Dragons. The alarm was that a fantasy role-playing game was leading players into the occult or toward harm. The history should show which cases were cited and what later scrutiny found.

Source still being found. a sourced, dated account of the Dungeons & Dragons alarm, with the specific cases cited at the time and what later scrutiny found

The satanic daycare panic. In the 1980s, accusations spread that children in daycare centers were being abused in organized satanic rituals. Here the folk devils were specific, named people, and the cost to them was concrete. It may also hold the hardest lesson for Pole B: child abuse is real, and the collapse of the ritual-abuse claims was not evidence that it wasn't. The history needs to show both.

Source still being found. a sourced, dated account of the satanic daycare abuse panic, including at least one specific case, what happened to the accused, and how the claims were later assessed

Video games. The alarm was that violent games produced violent children. Here effect sizes matter most, because research was done, and the argument has been over how large an effect it found, if any.

Source still being found. a sourced account of the video-game violence debate, including what the research found and how large the effects were, read in the studies themselves

Sexting. The alarm was that teenagers sending each other explicit images was a new and dangerous epidemic. The history should show what was claimed and whether the response harmed the young people it was meant to protect.

Source still being found. a sourced, dated account of the sexting alarm, including what was claimed about its prevalence and what legal responses followed

Example still being found. a historical case in which a warning was dismissed as a moral panic and later turned out to be right, sourced and dated, to stand beside these five as a historical example of Pole B's failure

If the sources bear it out, a finished version of this section should show each episode right about something and wrong about the rest. Abuse does happen; teenagers do get hurt online; whether children imitate what they see is a real research question. The error was in letting the size of the alarm stand in for the size of the harm, and, on the other side, in letting the alarm's excesses stand in for proof that there was no harm at all.

How each pole fails

The two poles fail in mirror-image ways.

The panicker starts from a case and arrives at a trend. A single story, often true, is told as if it were typical; a possible harm is treated as a probable one; spread is forecast without checking how often such forecasts have come true. Policy gets made at the speed of the story, and the folk devil pays before the evidence arrives.

Example still being found. a real, quoted, dated present-day instance of Pole A from the political right, in which a harm is asserted from anecdote without regard to its scale

Example still being found. a real, quoted, dated present-day instance of Pole A from the political left, in which a harm is asserted from anecdote without regard to its scale

The panic-caller starts from a pattern and arrives at a verdict. Because many alarms were overblown, this one must be; because people with an agenda raised it, it must be false; because the evidence is imperfect, nothing can be concluded. The tell is that the dismissal would have arrived whatever the evidence said, and its own forecast, that nothing will come of this, goes as unchecked as the panic's.

Example still being found. a real, quoted, dated present-day instance of Pole B from the political left, in which "moral panic" or a similar label is used to dismiss a concern without engaging its evidence

Example still being found. a real, quoted, dated present-day instance of Pole B from the political right, in which "moral panic," "hysteria" or a similar label is used to dismiss a concern without engaging its evidence

Both poles share one move. Each settles the question of size before measuring anything, and each treats the other side's alarm as a symptom while treating its own as a finding.

Where it borders other grammars

This grammar rarely travels alone.

With G1: Purity and Contamination it shares the folk devil as a pollutant. When one exposure is said to be enough to do the damage, dose has stopped mattering.

With G5: Decline and Golden Age it shares the young as evidence. Many panics are about what the next generation is becoming, measured against a remembered childhood held up, unexamined, as the healthy standard.

With G6: The Hidden Hand it shares the search for an organizer. A panic that begins with scattered cases often ends by positing a network behind them, and the counter-panic's reply, that there is never any network, is G6's opposite pole.

With G2: Liberty versus Safety it shares the policy fight that follows: demands for protection on one side, the language of freedom on the other, and the size of the harm lost between them.

When naming the pattern becomes a weapon

This grammar's own name is this site's standard example of a vocabulary turned into a weapon: "moral panic" used to end an inquiry rather than discipline it.

The phrase carries a verdict inside it: that the alarm is out of proportion to the harm, said before anything has been measured. Sometimes that is right. But the phrase saves the work of checking. A worried parent, a community that sees a pattern in its own streets, a researcher who has found a real effect, can each be waved away with one word, and the person waving gets to feel like the calm one in the room.

The weapon also points the other way. Panickers have their own vocabulary for those who question them: they don't care about children, they are covering for something. Suggesting that whoever asks "how big?" must be on the side of the harm is the panicker's conversation-ender, as "moral panic" is the panic-caller's.

Example still being found. a real, quoted, dated instance of someone who questioned the scale of an alarm being accused of not caring about the victims or of sympathizing with the folk devil

Both weapons do the same thing. They answer a question about size with a judgment about character.

Catching it in yourself

A few questions help.

Which alarms do I take seriously on the strength of one story? Think of the last alarm you shared or repeated. Did you know how common the thing was, or only that it had happened?

Which alarms do I dismiss without looking? The last time you said "this is just a moral panic," did you check the evidence, or recognize the people raising the alarm and decide from that?

What did I predict last time? Recall an earlier alarm you believed, or mocked. If you can't remember what you thought, or have quietly revised it, that is the composting at work in your own memory.

Who pays if I'm wrong? If you are wrongly alarmed, someone is treated as a danger who isn't one. If you are wrongly dismissive, someone is harmed while you look away. Which cost have you been counting?

Would I accept the same evidence from the other side? If an alarm you shared rested on the kind of evidence you mock in theirs, your standard is your side, not the evidence.

Tools that help

• B1: The Effect-Size Reflex. Ask how big a real effect is, for whom, and compared with what. It is the question both poles skip: neither "it's dangerous" nor "it's hysteria" says whether a real harm is small or large. Its weaponized form is demanding ever more precision as a way of never acting on a harm that is real but hard to measure.
• C4: The Prediction Ledger. Write a prediction down, date it, and grade it later. Every panic makes a forecast, and so does every counter-panic; going back to check what each said would happen turns the composting into an audit. Its weaponized form is grading opponents on outcomes and allies on intentions.
• A3: The Newsfeed Inversion. How often something appears in your feed measures how well it performs, not how common it is. A week of viral stories can make a rare thing feel like an epidemic; the remedy is the actual trend line. Its weaponized form, waving away a trend that is real and rising, is the panic-caller's move.
• D2: The Rage Audit. Strong emotion is information about how a message was designed, not about the world. Panics run on fear and anger, and so do the mocking replies to them; before sharing either, ask whether the enraging detail does real work in the argument. Its weaponized form is dismissing a true claim because it arrived angry.

Four further tools work against every grammar, this one included: A4: The Mirror Check, C2: The Steelman Protocol, E5: The Precision Covenant and F4: The Scout Report. The full set is on the tools page.

Where it shows up now

As of October 2026, the following live arguments are listed as places to look, not as findings. Listing an argument here does not mean the site has judged which side, if any, is panicking.

• Smartphones and teenage mental health. The fight over Jonathan Haidt's claim, in The Anxious Generation, that smartphones caused a teen mental-health crisis. The evidence is genuinely contested, which makes it a useful case for both poles. Not yet examined.
• AI doom and AI "slop." Alarm about catastrophic risk from artificial intelligence and alarm about low-quality AI-generated content, running in parallel. Not yet examined.
• Fentanyl exposure stories. Reports of police officers overdosing from touching fentanyl. Not yet examined.
• Human trafficking statistics. Disputes over how large the figures for trafficking are and how they are arrived at. Not yet examined.
• Drag performances and library books. Alarms about drag events and about books in school and public libraries. Not yet examined.
• The "loneliness epidemic." Including the claim that loneliness is as deadly as smoking fifteen cigarettes a day. Not yet examined.

Related concepts

• Moral panic. A wave of public alarm about a group or practice that is out of proportion to the harm it actually causes; the model of how such waves rise and fall is associated with the sociologist Stanley Cohen.
• Folk devil. The group, person or practice a moral panic singles out as the source of the problem.
• Counter-panic. Alarm about other people's alarm, which can be just as out of proportion in the opposite direction.
• Effect-size literacy. The ability to ask not just whether an effect is real but how large it is, and whether that size matters for any decision.
• Availability cascade. A self-reinforcing process in which a belief seems more plausible the more often it is repeated.
• Slippery-slope reasoning. Arguing that a small step will lead, by stages, to a much larger and worse outcome.
• Telling a live hazard from a folk devil. The practical skill this grammar asks for: sorting a real, sized danger from an alarm whose loudness has outrun its evidence.