Research Highlights

Research Highlights top

The restorative justice toward the era of decreasing birthrate and aging populationNew and Old Debates over Models of Offender Treatment: From the Medical Model versus the Justice Model to the RNR Model versus the GL Model

writer: AIZAWA, Ikuo (Ritsumeikan Global Innovation Research Organization, Senior Researcher) published: 2016-12


When it comes to how offenders are to be treated, in the past there was a conflict between the medical model and the justice model, while today there continues to be a debate between the RNR model and the GL model. This debate is occurring in English speaking countries, but it can provide many hints and suggestions when we consider how offenders should be treated in Japan.

The medical model versus the justice model

The medical model of offender treatment is a way of thinking that views crime as an expression of some sort of illness or maladaptation, sees criminals as people who are ill, and considers their treatment to be the provision of medical care. This model has been strongly influential in the United States since the late 1930s. This is not unrelated to social circumstances. The economic crash changed people’s view of crime from “a crime committed by an individual” to “a response to powerful forces beyond a person’s control.” Psychoanalytic theory and social work, both of which showed significant progress during the same era, raised expectations that people who commit crimes could in fact be treated.
Entering the 1960s, however, criticism of the medical model was developed by both liberals and conservatives. This was a way of thinking that came to be called the “justice model.” Conservatives such as Ernest van den Haag emphasized the punitive and deterrent aspects of punishment rather than its utility. They believed that an approach that aimed to rehabilitate criminals and return them to society was too easy on them, and that a harsher punishment that would achieve justice was required. Liberals such as Francis A. Allen, on the other hand, claimed that the idea of rehabilitation itself infringed on the rights of those being punished and gave rise to injustice. One facility, for example, subjected young prisoners to cruel treatment in which they were sprayed with water from a fire hose in a process called “hydro therapy.” In principle the medical model was also a system of “indeterminate sentence” in which a prisoner was not released from a facility until their treatment was complete, and it was pointed out that this could lead to deprivations of freedom that exceeded responsibility. The medical model was thus caught between the criticism of both liberals and conservatives.
During the same period, the medical model was also dealt a fatal blow in the so-called “what works” debate by thinkers such as Robert Martinson. These critics reviewed all of the literature on the treatment of criminals that had been published in English from 1945 to 1967, and concluded that aside from a few exceptional cases efforts at rehabilitation had no effect whatsoever on recidivism. Their evaluation was criticized for being too harsh because it applied the standard of a program being successful right from the time it was announced, but the rejection of the effectiveness of medical treatment, a premise of the medical model, lead to a loss of faith in offender treatment itself.

The RNR model versus the GL model.

In contrast to this pessimistic view of offender treatment, what came to prominence in the midst of the trend toward evidence-based practice that began in the 1990s was the RNR (Risk Need Responsivity) model developed by, among others, Donald A. Andrews and James Bonta. This model is based on three principles. First, the density of treatment must be concentrated on those with a high risk of recidivism (the Risk principle). Second, treatment must be restricted to criminogenic needs (the Needs principle). Third, treatment must increase the responsivity of the person in question, mainly by employing cognitive behavioral therapy (the Responsivity principle). It has been shown that the more faithfully these principles are followed the lower the risk of recidivism. The RNR model emerged as the savior of criminal treatment, which had been viewed with great skepticism since Martinson, and beginning with Britain it has influenced criminal policy in various countries around the world.
There has been criticism of this approach, however, by people such as Tony Ward. These critics maintain that the RNR model focuses only on managing risk, and has difficulty eliciting the motivation and cooperation of the people to whom it is applied. Moreover, as the effectiveness of the RNR model provides a foundation for the justification of intervention, there is a danger of extreme methods of intervention (from castration to capital punishment) being justified because they are viewed as effective. The approach Ward and others have advocated based on these criticisms of the RNR model is called the “GL (Good Lives) model.” This model holds that human beings naturally pursue “goods” of some sort (called “primary human goods”), and criminal behavior results from attempting to obtain these goods by inappropriate means. For example, there are people who employ violence in an effort to obtain the “good” of intimacy with other people. The GL model therefore holds that rather than simply managing risk, the aim of treatment should be to have people obtain “goods” (agency, friendship, inner peace and creativity, etc.) through empowerment. It is thought that for the person in question, the pursuit of a “good life” in this sense will lead to a “(morally) good life” without crime.

What sort of offender treatment is most desirable?

In Japan, too, against the backdrop of a recent focus on preventing recidivism, there are increasing demands for thorough risk management in offender treatment. The RNR model does indeed have an advantage when it comes to results and the efficient distribution of resources. If adopting it leads to the conclusion that any treatment is acceptable as long as it lowers the risk of recidivism, or that any treatment that does not lower this risk should be rejected, however, then treatment will become both too dangerous and too narrow. In this regard, I think there is a lot to be learned from the GL model that pursues good lives for criminals from a broader perspective. I also believe that the conflict between the RNR and GL models is not as fundamental as that between the medical and justice models, because both of these newer models accept the necessity of criminal treatment. Ward himself maintains it is possible to combine the RNR model with the GL model. I believe the question of whether risk management and the pursuit of good lives can be compatible with each other is a topic that requires further study.


AIZAWA I., (2015), “Social inclusion and offender treatment: from the moments of Justice and Utility”, Ryukoku corrections and rehabilitation center journal 5, pp. 16-35.
Ward T. and Maruna S., (2007), Rehabilitation: beyond the risk paradigm. London: Routledge.


Ritsumeikan Journal of Human Sciences


Find us on facebook


Institute of Ars Vivendi. Ritsumeikan Univ.