Masayoshi Shido (The candidate): He committed the assault, and then decided which of the two would be the criminal

A man is assaulting a woman in the street outside a bar. A teenager intervenes, the man falls, and by morning the teenager has been charged with assault, convicted, placed on probation and expelled from his school in his home town. The man is a politician on his way to the leadership of a governing party, and his name does not appear in any of it. Everything that follows in the boy's life — the transfer to Tokyo, the guardian who was told he is dangerous, the teachers who were briefed before he arrived, the classmates who already know — flows from a document, not from the night. Criminology has a framework built for exactly this, and its central claim is that the document is the more important of the two.

// The theoryDeviance is not what you did, it is what was successfully said about you

Howard Becker's Outsiders (1963) contains the most quoted sentence in modern criminology: social groups create deviance by making the rules whose infraction constitutes deviance, and applying those rules to particular people; deviant behavior is behavior that people so label. The point is not the relativist one it is often flattened into — that nothing is really harmful — but a procedural one about where the quality of being deviant resides. It is not in the act, because identical acts are processed differently depending on who commits them; it is in the outcome of a social process with several stages, each of which can fail. Frank Tannenbaum had described the first stage in 1938 as the dramatization of evil: the community shifts from defining the act as bad to defining the person as bad, and the tagging becomes a way of stimulating the very traits complained of. Edwin Lemert supplied the mechanism in 1951 with the distinction that still carries the theory. Primary deviance is the original act, which may have any cause and usually has no effect on identity. Secondary deviance is behavior produced by the societal reaction to the first — the person reorganises their life around the status assigned to them, because that status has closed the ordinary routes and opened others. Becker added the part that matters here: the capacity to make a label stick is unevenly distributed, and the study of deviance is therefore in large part the study of who gets to do the naming.

// The subjectThe labeller is the one person the label cannot reach

What makes this subject worth the file is that he occupies both positions in the same incident. He performs the primary deviance — an assault on a woman in a public street, witnessed — and he also controls the machinery that decides what gets processed as a crime, because he has standing, a party, a prosecutor who wants a future, and a boy with none of those things. The conviction is not a miscarriage in the usual sense of an error. It is the process working exactly the way Becker described, at the point where the two accounts of the night meet and one of them has more weight behind it.

His political program extends the same operation to a country. The plan is to engineer public disorder, let the blame settle on whoever is available to carry it, and present himself as the man who will restore order — which is the labeling process run deliberately and at scale, the manufacture of a category of dangerous people in order to be elected against them. It is also the clearest possible statement of the theory's limit condition. Labelling requires a labeller with more power than the labeled; a person who accumulates enough of it stops being reachable, not because his conduct improves but because there is nobody left above him to apply the tag. That is why he closes this hub rather than opening it. Kamoshida spends the authority of a school, Madarame a professional convention, Kaneshiro the structure of his victims' lives, Okumura the position of an employer — and Shido spends the state itself, which is the one instrument that defines what the others did as offenses at all.

// The crackThe theory that explains the second offense and not the first

Labelling theory was the dominant framework of the 1960s, was badly damaged in the 1970s and has recovered more ground than its critics expected. The objections are worth stating precisely. First and most serious: it does not explain primary deviance — Lemert conceded this himself, since the first act happens before any label exists, and a criminology with nothing to say about why offenses begin is half a criminology. Second, the empirical record was poor for two decades: studies comparing labeled and unlabelled offenders often found no effect, which is what let the theory be written off. That verdict has since been partly reversed by longitudinal designs — Bernburg and Krohn, and the Sampson and Laub life-course work — which find that official intervention in adolescence reduces later educational attainment and employment, and that reoffending follows from those blocked routes rather than from the label directly. The effect is real, indirect and mediated by opportunity, which is a more demanding claim than the original and better supported. Third, the theory can slide into denying the harm: a framework centered on the reaction can read, in careless hands, as though the victim of an actual offense were merely a participant in a social construction, which is both false and a betrayal of the people the theory was meant to protect. Fourth, its policy child — radical non-intervention, the recommendation to leave well alone — produced real damage when applied without discrimination, as diversion is powerfully protective for minor first offenses and negligent for serious and repeated ones. What survives, and it is a great deal: formal processing has costs of its own that must be weighed against its benefits; those costs fall hardest on the young, the poor and the already-marked; and the question of who is in a position to label whom is not a side issue but part of the explanation of the crime rate itself.