A settlement in a Colorado winter is starving. Its leader is a former teacher who preaches, keeps order without visible force, and is addressed with something close to affection by people who have nothing. He has made two decisions that the community lives inside: that they will eat what is available, and that what they are doing is not what it is. Nobody in the group is coerced into either. They are supplied with a way of seeing themselves, and the way of seeing comes with scripture, with a paternal voice and with the promise that everything happens for a reason. The criminology that explains this arrangement is a footnote in most textbooks and should not be.
// The theoryYou do not need bad company. You need a perspective
Daniel Glaser's “Criminality Theories and Behavioral Images” (1956) begins from the gaps in Sutherland. Differential association holds that criminal behavior is learned in intimate personal groups, which fits gangs and apprenticeships and fits very badly a great deal of real offending: the solitary offender, the person radicalized by material rather than by people, the one whose associates are entirely law-abiding. Glaser's reformulation keeps the learning and changes the unit. A person pursues criminal behaviour to the extent that he identifies himself with real or imaginary persons from whose perspective his criminal behaviour seems acceptable. Three things follow. First, contact becomes optional — identification can be with a figure encountered in a book, a broadcast or a scripture, which is why the theory travels to phenomena Sutherland could not reach. Second, the mechanism is perspective-taking rather than technique-transfer: what is acquired is a vantage point from which the act is describable as something other than what it is. Third, it explains change without relocation: a person can shift identification while keeping the same friends, job and street, which the association model struggles with.
A community held together by a description
What David supplies is not food and not safety, both of which he is manifestly failing to provide. It is a description. The group is not eating people; it is surviving, being provided for, being held together by a shepherd during a trial. Every element of the framing is doing identifiable work. The religious register makes the act a sacrament rather than a transgression. The paternal role makes obedience a filial virtue rather than submission. And the promise of meaning—everything happens for a reason—converts the accumulating evidence of what they have become into further confirmation of the plan. Glaser's point is that a person who has taken this vantage point does not need to be taught anything else, because from inside it there is nothing to refuse. The group's cohesion is therefore extremely brittle in one specific way: it depends on nobody applying an outside description. David's reaction to Ellie is so extreme precisely because she is the first person in the settlement who names what is happening in ordinary words.
// The subjectThe offer made to Ellie is the theory operating on one person
The sequence with Ellie is worth reading carefully, because it is the mechanism applied individually and it is also, unambiguously, grooming. David does not threaten her. He offers, in order: recognition (he tells her she has a capacity for violence that he admires), belonging (a place, a role, protection), a shared identity (we are the same), and finally a future in which what he wants is the natural consequence of all three. Every step is an invitation to adopt a perspective from which the next step is reasonable. Researchers who study grooming describe exactly this progression — the establishment of a special relationship, the reframing of the adult's interest as uniquely perceptive, and the incremental redefinition of what is normal between them — and the reason it works is not that the target is credulous but that the perspective arrives before the request does. Ellie refuses, and what she refuses is not the proposal. It is the description of herself that would have made the proposal make sense.
That places him against the other files in this hub rather precisely. Marlene reasons from a calculation she states openly. Kathleen is driven by a grievance she never stops naming. Joel operates on an attachment he does not examine. David is the only one whose method is to install the premises, which is why he is the most dangerous person in the series to the fewest people and why the story he tells about himself never once cracks.
David
Get the map of theories for free
Subscribe to the criminal archive —one case a week, no spam— and I'll send you The map of theories: the great currents of criminology on a single sheet.
// The crackAn elegant idea that almost nobody tested
Differential identification has a specific historical problem: it was correct, influential and then absorbed. Akers's social learning theory incorporated imitation and definitions as two of its four mechanisms, and in doing so covered most of what Glaser had proposed within a framework that came with measurement instruments attached. The consequence is that Glaser's formulation was never independently tested at scale, and the honest position is that it is plausible and under-evidenced rather than established. Three further problems. Measurement: identification is a subjective state, typically assessed by asking people who they admire or whose opinion matters to them, and those answers are shaped by the same self-presentation pressures as everything else in self-report. Circularity: if we infer identification from the behavior it is supposed to explain, nothing has been demonstrated, and escaping this requires establishing the identification first — which almost no study has done prospectively. Direction: the selection problem that haunts differential association applies here too, in a sharper form. People choose whom to identify with, so a prior disposition may be producing both the identification and the conduct, and Glaser's framework has no internal way to separate them. What survives is the correction itself, and it matters more now than when he wrote it: a theory of criminal learning that requires face-to-face intimate association cannot account for radicalization at distance, for imitation from media, or for the solitary offender who has assembled a worldview from material rather than from people. Those are not edge cases any more.
The point of view did the rest
David is a predator who uses a congregation as cover and who attempts to coerce a fourteen-year-old, and the theory here explains his method rather than diminishing any of it. The observation worth carrying out of the file is about his followers rather than about him. None of them was recruited by force, none was taught anything technical, and most would have described themselves as decent people doing what was necessary. What they accepted was a vantage point, and the vantage point did the rest. That is a description of how a great many people end up participating in things they would have refused if the request had arrived before the perspective did.
Three key ideas
- Glaser (1956) replaced Sutherland's association with identification: a person offends to the extent that they identify with real or imaginary figures from whose perspective the act seems acceptable. Contact becomes optional, which lets the theory reach solitary offenders and learning from media.
- What David distributes is a description, not a technique: the religious register makes the act a sacrament, the paternal role makes obedience filial, and “everything happens for a reason” converts evidence into confirmation. The offer made to Ellie is the same mechanism applied to one person, and it is grooming.
- The theory was absorbed by Akers before being tested independently, identification is hard to measure, the inference is circular unless established in advance, and people choose whom to identify with. The correction survives: intimate association cannot explain radicalization at distance.
Sources · to keep reading
- Glaser, D. (1956). “Criminality Theories and Behavioral Images”, American Journal of Sociology, 61(5).
- Sutherland, E. H. (1947). Principles of Criminology, 4th ed. Lippincott.
- Akers, R. L. (1998). Social Learning and Social Structure. Northeastern University Press.
- Craven, S., Brown, S. y Gilchrist, E. (2006). “Sexual Grooming of Children”, Journal of Sexual Aggression, 12(3).
- Bandura, A. (1977). Social Learning Theory. Prentice Hall.
- Warr, M. (2002). Companions in Crime. Cambridge University Press.
Case-file questionsFrequently asked questions
Who is David?
David (“El predicador”) is a villain from The Last of Us, analysed on KRIMINIS through the lens of criminology. He never threatens anyone in the group.
Why does David (The preacher) become a villain?
David and differential identification: the preacher who turns cannibalism into communion and predation into fatherhood. The file on the theory that explains offending without bad company.
Which criminological theory explains David?
David's case is analysed through Differential identification. Daniel Glaser proposed an elegant correction to Sutherland in 1956. Saying criminal behavior is learned “by differential association” leaves a lot of people unexplained: those who offend without a single offending friend, those who learn from a figure they have never met, those who change without changing company. Glaser proposed replacing association with identification: a person pursues criminal behavior to the extent that they identify with real or imaginary figures from whose perspective that behavior seems acceptable. David is that theory in a dog collar. He trains nobody and teaches nobody a technique: he distributes a perspective, and from inside it everything his group does has a different name.
