Sixth of the implications, and a coda — arguments built on the model, as distinct from the model itself. The mechanism post sets out the model; the coercion post describes the groups that engineer it. This one is about the word — and about a debt that limitation owes, because the sources it relies on were used to justify a great deal of harm.
Status. This post is mostly documented history and a named scholarly critique; the argument comes at the end. It names one group — MOVE — because its case is the clearest on record and because the harm done to it is the point. It diagnoses no group, and it is not a checklist.
Six posts have described a structure: a collapse, engineered, held by someone standing on the far side. The coercion post listed its mechanisms — exclusivity, an owned frame, an accelerated pace — and called the picture “coercive control.”
But there is a word for all of this, and it is cult. The series has been conspicuously not using it. That was deliberate, and this post is the reason, said out loud:
What happens when a description of harm becomes a verdict about a group?
Because it does something. The label is not a neutral summary of the mechanisms. It is a move in a game with a long, documented, and ugly history — one in which the people applying it have repeatedly done harm of their own, and in which some of the series’ own sources are implicated.
The sharpest formulation is Richard Kent Evans’, in MOVE: An American Religion:
“There is, to be clear, no such thing as a cult.” The category “cult” “does not reflect an empirical reality. There are no charges that can be leveled at ‘cults’ that cannot also be leveled at things we call ‘religions.’”1
And he names what it is instead: a “nomenclature of containment” — following Robert Orsi — “an implicit argument about what the observer imagines ‘true religion’ to be.”
Judith Weisenfeld puts the same point from the other side: “The label ‘cult’ tells us less about a group’s theology or members’ self-understandings than about the commitments of those who use the label.”2
This is not academic nicety. It is a testable claim about the word’s behaviour. Check it against any group you know: the things that brand something a cult — closedness, a charismatic founder, demands on members’ time and money, strong in-group norms, a claim to special truth — are all present, to some degree, in ordinary churches, firms, families, armies, universities, and political parties. The word is not sorting by what the group does. It is sorting by how much the group resembles the labeller’s idea of legitimate religion — which, as Evans says plainly, tends to mean “religion done right looks like moderate white Christianity.”
That has a consequence that is not theoretical.
Evans documents that critics attempted to discredit MOVE by accusing founder John Africa and his community of being abusive, disingenuous, irrational — that is, the abuse accusation functioned as a weapon in the lead-up to state violence. And the violence is among the worst on record: in May 1985 Philadelphia police dropped a bomb on the home of the Black-liberation community, let the fire burn, and killed eleven people including five children, destroying 61 houses and displacing 250 people.3
Weisenfeld’s point about the pattern is the general form of it: Black-led religious innovations, women-led innovations, and especially Black women-led innovations are disproportionately criticised as cults — and members of those groups “are often seen as less rational, weaker, and thus more susceptible to religious, as well as economic and sexual, manipulation.” The label arrives pre-loaded with an account of who is credulous.
So the word does not merely fail to describe. It has been an instrument — of surveillance, of civil-rights violation, of discrediting. Whatever your intentions, picking it up means picking up that history.
Here is the part that this series has to say about itself.
There was a movement in the 1970s and 80s — the anti-cult movement (ACM) — built on “rescuing” young adults from groups it called cults. Its central doctrine was “brainwashing.” Its practice was deprogramming: the forcible removal of adults from their communities, described in the literature as extra-legal abduction, defended initially as “a family problem involving unruly juveniles,” and eventually reined in by civil-liberties and NRM litigation — culminating in the collapse of the Cult Awareness Network in 1996.
Now the parts that are uncomfortable:
A distinction that is the whole lesson, so it must be stated precisely: Lifton and Singer did not run deprogrammings. They were scholars and clinicians — Lifton an academic psychiatrist whose fieldwork on thought reform was careful and whose criteria were descriptive, and Singer a clinician who worked with families in genuine distress. What the ACM took from them was the vocabulary: a set of observations about how coercive environments operate, converted into a warrant for action. The researchers supplied concepts; a movement supplied the inference — if a person’s agency has been overborne, then acting on their behalf without their consent is rescue — and the state supplied the force. Holding the two apart is not a courtesy to the scholarship. It is the entire mechanism by which good description becomes harm: not in the observation, but in the step from “this is what coercion looks like” to “therefore this person cannot choose.”
So the coercion post’s authorities — Lifton, Singer, and the thought-reform tradition — are also the intellectual supply of a movement that kidnapped people. That does not make the coercion post wrong. Lifton’s description of milieu control and confession can be accurate while the use made of it was harmful; the same lifeguard can be misused as a weapon. But a series that leans on these sources owes the reader this paragraph, and here it is.
The movement’s origins are worth holding too, because they are not villainous. The ACM was largely frightened parents — not operatives. Their adult children had joined groups they could not understand, and they wanted them back. The coercive-explanation was consoling precisely because it removed the blame from everyone: the child had not chosen this, the parent had not driven them to it. Sincere people, trying to help, built the apparatus. That is the uncomfortable shape of it — and it is the same shape as the coercion post’s own finding, that sincerity makes a harmful frame worse, because a sincere holder cannot see what a cynical one would.
And this pattern is not rare — it is the ordinary failure mode of a helping framework. The same decades produced the recovered-memory and ritual-abuse panic: clinicians, many of them sincere and some of them good, working from a theory that memories of abuse are commonly repressed and that recovering them is healing. The result was false memories, destroyed families, and a concept that is still radioactive. Nobody who started it set out to harm anyone. The theory was believable, it explained suffering, and it licensed an intervention — and that triad is the whole danger. A framework that is believable, explanatory, and actionable will be acted on, and its errors will be acted on at the same scale.
What saved the field in both cases was not better intentions. It was scholarship and courts: researchers who dismantled the brainwashing theory (Anthony, Robbins, and others), civil-liberties litigation, and clinicians who documented the harm done by their own profession’s practice. The same culture that generated the useful concepts generated their limits — slowly, and after the damage. That is the only self-correcting structure anyone has, and it is worth saying plainly in a series whose whole argument is that careful measurement beats confident assertion.
The obvious reply is: so never say it. Avoid the label entirely, stick to description, and the harm goes away.
That does not work either, and pretending otherwise is its own kind of harm. If “cult” is unsayable, then abuse becomes unchallengeable — because the accusation “you are calling us a cult” is available to any group as a shield. The No True Scotsman runs in both directions: “No real religion abuses its members” protects abusers, and “cult is just religion you don’t like” protects them too. The academic literature is clear that abuse in high-demand groups is real, that coercive control is a studied structure, and that people are genuinely harmed — sometimes severely, sometimes permanently.5
There are two failure modes, and they are symmetric:
| error | what it protects |
|---|---|
| Label freely | the labeller’s comfort; the state’s power; the reputation of the “respectable” |
| Never label | the abuser, who gains a shield from the vocabulary of tolerance |
Both of them leave the person inside the situation with no language, no warning, and no way out.
So the answer is not a better word. It is to stop sorting groups and start describing mechanisms — which is what the coercion post was already doing, and why it went out of its way to avoid naming anyone.
Look at what it actually offers. Not membership criteria or a vibe. Specific, checkable mechanisms:
Every one of those is a question about behaviour, answerable by an outsider, and none of them requires calling anything a cult. That is the whole point. A mechanism can be described, observed, and acted on. A label can only be agreed with or rejected — and rejecting it is cheap, because the group’s lawyers, defenders, and sincere members all get to say “we’re not a cult.” Nobody can say “we do not have an exclusive frame” if they do.
Mechanism-first preserves the ability to warn without the harms of labelling. It is more useful and less dangerous at the same time, which is rare, and it is why the coercion post reads the way it does.
The practical version, and it is short:
Describe what you have observed, specifically. Ask. Do not diagnose. “I’ve noticed you’ve stopped seeing X,” “your sleep has changed,” “what happens if you ask a hard question there?” — these are things a person can evaluate. “You’re in a cult” is not; it recruits the whole apparatus of the label, and it hands them a defensible denial.
Let the person decide. The mechanism has one structural result that governs this: the state cannot initiate its own exit, so recovery comes from outside — but “from outside” means a relationship, not a verdict. Someone who needs an exit cannot use your diagnosis. They can use the fact that you’re still there, still asking, still not conceding that the situation is normal.
And sometimes the ethical choice is to withhold a name you have. There are situations where the evidence is real, the harm is documented, and naming the group publicly would cost more than it buys — litigation, retaliation, the exposure of people who cannot afford to be identified, or simply the conversion of a live situation into a public position. Holding a name back is not cowardice or complicity. It is a legitimate judgement that the description is more useful than the verdict. This series makes that choice throughout, and it is a choice, defended here rather than left implicit.
The last move is the one the coercion post’s §7 already began, and it belongs here too, in full.
Every mechanism this series describes is available to be turned into a label. “Exclusive community,” “owned frame,” “accelerated pace,” “a teacher present at onset” — a checklist of those is a cult-detector, and cult-detectors are exactly what the anti-cult movement ran on. Give someone four boxes and a suspicious neighbour and you have reproduced the machine.
The safeguard is not a better checklist. It is a discipline, and it has to be re-applied every time:
That the vocabulary of depersonalization and dissociation was used to justify abducting adults is the sharpest possible warning, because it is this series’ own vocabulary. The lesson is not to stop describing. It is that a description is not a licence — and the distance between them is maintained by discipline, not by having the right words.
Because this risk is not hypothetical, and because the whole point of this post is that descriptions become warrants, the model at the centre of this series carries an explicit limit on its use. This is not a disclaimer. It is the boundary that the model’s own content enforces.
The model must not be used to:
What makes this enforceable rather than decorative is that every one of those misuses requires stepping past what the work actually says. There is no individual measurement in it, no validated mapping from model to person, and no result from the test that would be needed. Misuse here is not a subtle hazard; it is a category error, and it is only available to someone who has not read the bounds.
And the useful test is reflexive, so it can be applied to anyone invoking this work — including the author:
If someone cites this work to override a person’s account of their own life, they have left the work behind. They have reproduced the exact inference this post documents — from “this is what coercion looks like” to “therefore this person cannot choose.” That step is where the harm lives, and it is the one thing here that is flatly prohibited.
Marking: documented — Evans’ and Weisenfeld’s critiques; the MOVE bombing; the ACM’s use of brainwashing theory, deprogramming, and the Lifton/Singer apparatus; coercive control as the preferred frame. Argument — that mechanisms should replace labels; that the two failure modes (always label / never label) are symmetric and both leave the person unprotected; that this framework is itself label-able and requires discipline to be held safely.
Evans, R. K. (2021). MOVE: An American Religion. Oxford University Press. — “There is, to be clear, no such thing as a cult”; the category as a “nomenclature of containment” (after Orsi); the account of MOVE’s labelling and the 1985 bombing. https://therevealer.org/bombing-american-religion-to-save-it↩︎
Weisenfeld, J. (2016). New World A-Coming: Black Religion and Racial Identity during the Great Migration — “The label ‘cult’ tells us less about a group’s theology or members’ self-understandings than about the commitments of those who use the label”; the “religio-racial movements” framework. https://medium.com/bates-americanislam/what-does-race-have-to-do-with-cults-562671906093↩︎
MOVE — founded 1972 around John Africa’s teachings (Black liberation, natural living, rejection of technology); bombed by Philadelphia police in May 1985: eleven dead including five children, 61 houses destroyed, 250 people displaced. https://en.wikipedia.org/wiki/MOVE_(Philadelphia_organization)↩︎
On the anti-cult movement, “brainwashing,” and deprogramming — Young, E. A. (2012), “The use of the ‘Brainwashing’ Theory by the Anti-cult Movement in the United States of America, pre-1996”: no reliable scientific evidence for the brainwashing theory and “a great deal of research against it”; the term’s origin with Edward Hunter, “an undercover CIA propaganda and psychological warfare specialist,” in a documented disinformation campaign; deprogramming described as extra-legal abduction of adults; the ACM’s characterisation of victims as “abnormally depersonalised, dissociated and regressed to psychological infancy”; Lifton’s totalistic-community criteria and Singer’s trial testimony in the ACM’s apparatus; the collapse of the Cult Awareness Network in 1996. https://journals.openedition.org/zjr/387↩︎
The more defensible modern frame is coercive control — reliable for assessing how power imbalances are created and maintained in high-demand groups. Feliciano, S. E. (2023), An Application of the Coercive Control Framework to Cults. https://academicworks.cuny.edu/jj_etds/300↩︎