As indicated earlier, law enforcement is often assumed to serve a dual pur-pose, one that is both repressive and preventive. This dual objective can be traced back to two different theories. The retributive theory suggests that punishment (output) is necessary to restore the legal order as well as society’s trust in the rule of law (outcome). Retribution is achieved by hav-ing offenders suffer, as their victims and society as a whole have suffered because of them. The harm inflicted on the victim must also be relieved or compensated. Hence, offenders are mainly punished because they deserve to be punished.
The utilitarian or instrumental theory focuses on crime prevention via two separate mechanisms. Specific prevention occurs when punishment put) deters the culprit from overstepping the mark again in the future (out-come). At the same time, law enforcement (input) creates the opportunity (output) to rehabilitate offenders (outcome) as part of their sentence. And in the case of detention (output), offenders are of course physically pre-vented from reoffending for the time being (outcome). General prevention is also thought to come about as a result of deterrence (outcome), but this time involving potential perpetrators among the general population rather than the actual convicts. Individual punishment then sets an example and confirms society’s norms and values (De Keijser 2004). Countries vary in the emphasis they place on the different functions of law enforcement, as revealed for example in the types of punishments imposed, treatments
available and day-time programs for rehabilitation and after care. Evidence on general prevention is mainly found in economic literature. For evidence on specific prevention, we rely on an extensive international review by Van Noije and Wittebrood (2008, 2009) of – often criminological – impact evaluations. This review included only studies which met the criteria of a quasi-experimental design. Reference is made to the original review for a list of these underlying international studies, which will not be mentioned individually in the following sections.
General prevention
The impact on general prevention is expected to derive from both the prob-ability and the severity of punishment. The probprob-ability of punishment is firstly the result of the probability of being arrested, and secondly, the probability of being convicted. The severity of punishment is determined by the type of punishment (e.g. fines versus imprisonment) and its severity (level of fines, length of prison sentence). These two variables determine the poten-tial costs of criminal activities, which are weighed against the potenpoten-tial gains and legal alternatives. This logic is firmly rooted in rational theory and is most elaborately developed in economic literature. The underlying assumption, introduced by Becker in the 1960s, is that criminals are utility maximisers just like anyone else. If a potential perpetrator judges that the potential costs outweigh the potential gains, he or she is assumed to refrain from illegal activity: general prevention, in other words (Eide et al.
2006).
Eide et al. (2006) depict the distinction between economists and crimi-nologists by their perspective on crime: it is considered rational behav-iour by the former, but deviant behavbehav-iour by the latter. Broadly speaking, according to economists, individual preferences are more or less constant and the impact of changes in environmental incentives (such as sanc-tions and wages) are therefore highly predictable (since they are rational).
Criminologists, on the other hand, assume that preferences are influenced by an individual’s social environment, social norms and other personal characteristics. As preferences vary between individuals, their reaction to incentives such as the probability and severity of punishment is also highly personal and less predictable. They do not necessarily deny the rational impact of punishment, but add uncertainty in the precise responses to it.
The evidence from empirical economic (criminometric) literature is most convincing regarding the effect of the probability of punishment, especial-ly for juvenile crime. The evidence is less consistent regarding the severity of punishment. Myopia, a lack of risk aversion and imperfect knowledge about their criminal ability among offenders may explain its often modest effects (Eide et al. 2006; Pyne 2012).
Specific prevention
At the front line of law enforcement are the police. Apart from the incon-sistent effects on general prevention of an increase in the number of police officers, effects which seem largely limited to police deployment
inproblem areas, police arrests are a means of deterring the individual arres-tee. This objective is certainly not achieved for all types of arrestees and offences. Once caught – minors especially, but also those who have com-mitted less serious offences – have been found to be more prone to commit subsequent offences. Hence, arrests are not always an effective instrument to reduce recidivism.
There is also no empirical support for the common policy assumption that imposing quicker punishments would enhance the effect of the imposed sanc-tion on recidivism. There is even some preliminary evidence to the contra-ry: an elongated threat of punishment was found to reduce the probability of recidivism among minors.
As regards types of punishment, the evidence is quite unambiguous about incarceration: detention alone is ineffective in reducing recidivism. Rather, adverse effects have been known to occur, undermining both general and specific prevention. This is not to say that detention is also ineffective if it is meant to prevent convicts from offending during the time they are locked away; detention may also serve to restore the legal order (retribu-tion). Likewise, alternative confinements that restrict the physical freedom of movement of convicts emerge as ineffective if no guidance and treatment are offered. For a specific group of offenders – drug addicts – who typically move in and out of prison due to frequent but relatively minor offences, the chances of rehabilitation are better if detention is sufficiently long to enable addiction care and social support to be offered.
Once detention has been imposed, a primary objective is to prevent recidi-vism after release. The road towards desistance is a highly personal process and one-size-fits-all interventions do not work. Nonetheless, cognitive behavioural therapy and social learning or social skills training appear to have the best chance of reducing recidivism (see also Sapouna et al. 2011). For young people these programmes were found to be more effective in ambulatory than in residential settings, however: the share of young people relapsing is higher in the latter. Often, and preferably, such programmes include guidance to improve their future prospects in terms of schooling or jobs.
Poor results are obtained by authoritarian institutions, camps or expeditions for re-education, where young people are temporarily removed from their familiar surroundings. They are expected to modify their behaviour under the influence of discipline and peer pressure. Ironically, critics point to peer pressure as one of the factors that stands in the way of behavioural improvements in these contexts.
Although as of yet no evaluation has demonstrated their actual effective-ness, attempts at resocialisation by guiding them towards schooling or suitable employment, and help them develop prosocial networks also emerge as promising for adult prisoners (see also Sapouna et al. 2011). This also applies to programmes in which prisoners are allowed the opportunity to acquire outside work experience during their period of detention, and to
continue after detention. As a rule of thumb, key events in offenders’ lives such as a stable relationship, parenthood, employment and re-integration in the local community enhance offenders’ motivation to stop reoffend-ing, as do interventions which facilitate such events.
Finally, aftercare is generally found to be indispensable to the success of rehabilitation programmes. Most knowledge about effective aftercare con-cerns young people, which again indicates the importance of guidance and support alongside mere surveillance. In general, if intensive supervision merely involves reporting to social work offices, and is not accompanied by some form of support in addressing criminogenic causes, it is unlikely to lead to desistance (see also Sapouna et al. 2011).