What prison can do to a person — and what the evidence cannot tell us

A person wakes to a light and the sound of a key. Rules of the institution shape waking, movement, meals, contact with family and the evening routine far more than personal choice.

Imprisonment restricts autonomy by definition. Its precise psychological effect cannot be inferred from bars alone. Safety, treatment, privacy, relationships, meaningful activity, duration, health and life before imprisonment all matter.

1. The “total institution” is an analytical framework

Erving Goffman described institutions in which major parts of life occur in one place under a common authority and formal schedule. The concept names a concentration of power and routine. It is not a biological mechanism or an experiment predicting the same personality change in every prisoner.

Prisonisation likewise refers to adopting norms, habits and strategies associated with prison life. It may involve greater dependence on institutional structure, vigilance, mistrust or emotional distance. “May” is essential: this is not a diagnostic checklist, and adaptations differ.

2. How strong is the evidence?

Much of the literature consists of observational studies, interviews, administrative records and expert syntheses. These can identify associations between prison conditions and well-being but have difficulty separating imprisonment itself from prior health, poverty, trauma, substance use, sentence length and differences between institutions.

The US National Research Council concluded that restricted movement, lack of privacy, uncertainty and fear are important sources of stress for some incarcerated people. It also stressed the high pre-existing health and social burden in the prison population and the difficulty of separating causation from selection.

A 2026 systematic review found recurring associations between restricted autonomy, privacy, predictability or relationships and psychological distress across prison and hospital settings. The studies were heterogeneous and mostly observational; the authors explicitly rejected simple causal conclusions and direct quantitative comparison.

3. An adaptation that works inside may fail outside

A strategy useful in a dangerous or highly controlled environment may lose its function after release. Vigilance may complicate trust, reliance on fixed routine may complicate independent planning, and emotional distance may affect family relationships. This is a plausible framework described in the literature, not the certain path of every person.

Housing, work, debt, healthcare, family ties, stigma and access to support also shape re-entry. A poor outcome cannot automatically be attributed either to “damage caused by prison” or to the person’s character.

4. What follows for institutions

The Nelson Mandela Rules state that prison systems should not aggravate the suffering inherent in deprivation of liberty and should support reintegration. This is a normative standard, not evidence that a particular programme works.

The practical question is not only how to maintain control, but which responsibilities and skills a person can safely practise before release. Each programme must be judged by its implementation and outcomes, not by the word “rehabilitation” in its name.

5. What this article claims — and what it does not

6. What would weaken the conclusion?

It would be weakened by high-quality longitudinal evidence across different prison regimes that, after accounting for prior health and social conditions, found no association between everyday autonomy, safety, relationships or meaningful activity and psychological well-being or re-entry. Current evidence is too heterogeneous for an absolute claim, but it is not empty.