CQ No. 25 SA Crime Quarterly no 26 • December 2008 3 Punishment and deterrence Don’t expect prisons to reduce crime Lukas Muntingh Civil Society Prison Reform Initiative muntingh@worldonline.co.za The belief is thoroughly entrenched that prisons fulfil the triple function of punishing offenders, making society safer by removing dangerous individuals, and deterring potential offenders from committing crime. The fact that hundreds of years of evidence suggests that prisons do none of these terribly well has not had any effect on our belief in the utility of this institution. In this article the author suggests that we hold on to these beliefs because they serve the interests of politicians, who can appease their electorate with the clear and simple solution prisons purport to provide; and the private sector, as prisons support their commercial interests in a number of ways. What is needed, the author argues, is deeper discussion and clearer thinking about the value and function of prison. One of the most persistent traits in thinking around crime and safety is the belief that imprisonment will reduce crime. This belief is shared by the judiciary, ordinary citizens, and politicians. It is merely a ‘belief ’ as there is little (if any) evidence to support the idea that imprisonment can reduce crime on any significant scale anywhere in the world (Gendreau et al 1999). For the past 200 years the basic principles of imprisonment have remained essentially unchanged. Post-apartheid South Africa inherited a prison system built on these principles, yet we have not critically asked what the purpose of imprisonment is and whether prisons really do serve this purpose. To politicians1 and the private sector, prisons (and imprisonment) are very attractive, for a number of reasons. In this article it will be argued that the continued use of and support for imprisonment is a consequence of political and private sector interests that are served by imprisonment. Imprisonment also enjoys significant public support. It will be shown that a drastic decline in the number of offenders sentenced to imprisonment in the past five years did not result in a rapid increase in violent crime; in fact, imprisonment had no visible impact on the rate of violent crime. These observations raise questions both about the efficacy of imprisonment, and our expectations thereof, that require critical examination. The issues raised in this article are complex. It is impossible in the space of a journal article of this nature to deal fully with all the complexities of the argument, thus the argument, as presented here, may appear to be over simplified. This does not, however, negate the fundamental point that we (as a society) need to critically examine imprisonment, the use thereof, its appropriateness, and expectations when imprisonment is imposed. This is especially important given the current high rate of incarceration. Ultimately, a clear and precise understanding is sought of the purpose of imprisonment within a constitutional democracy. More specifically, we need to critically examine the strategic utilisation of imprisonment in an overall crime control and reduction strategy; assuming that such a strategy exists. OLD NEWS ABOUT PRISONS The prison, as we know and understand it today, is a relatively recent phenomenon that reached its full stature by the early 18th century in Western Europe, and was influenced by reformations in North America. However, by the early 19th century there was increasing evidence that prisons were not effective in reducing crime and a number of astute conclusions were drawn (in France) between 1823 and 1842 to this effect, including: • Freed inmates are subjected to conditions that necessarily condemn them to recidivism: they are under surveillance by the police; they are assigned to a particular residence, or forbidden others; they leave prison with a ‘passport’ that they must show wherever they go with details of the sentence they have served (1823) • Prison produces delinquents. It does so by the very existence it imposes upon its inmates — whether they are isolated in cells or whether they are given useless skills for which they will find no employment (1836) • Prison often indirectly produces delinquents by causing destitution among the inmate’s family. The same authority that sends the head of the family to prison reduces the mother to destitution, the children to abandonment, and the whole family to homelessness and begging. It is in this way that crime takes root (1836) • Detention causes recidivism; those leaving prison have an increased chance of returning (1837) • Prison enables, even encourages, the organisation of a milieu of delinquents, loyal to one another, with a particular hierarchy, ready to aid and abet any future criminal act (1839) • Prisons do not diminish the crime rate; whether they are extended, their numbers multiplied or institutionally transformed, the quantity of crime and criminals remains stable or, even worse, increases (1842) (Foucault 1977:265- 268) Prisons systems and advocates for prison reform responded to these challenges. Prison reform in Europe and North America attempted to make prisons more humane and directed the focus at religion, fostering habits of industry, and rehabilitation. However, Foucault also reminds us that despite numerous attempts at prison reform, reform efforts were predicated upon the same seven universal maxims of the good ‘penitential condition’ which have remained unchanged for nearly 200 years: • Imprisonment ‘must have as its essential function the transformation of the individual’s behaviour’ – the principle of correction (1847) • Prisoners must be classified in respect of the gravity of the offence, age, mental attitude and the stage of their transformation – the principle of classification (1850) • It must be possible to adjust penalties, based on the degree of liberty, according to the behaviour or performance of the prisoner, the progress that has been made or relapse – the principle of modulation (1838) • Work is an essential element in the transformation of the sentenced prisoner – ‘it must enable him to learn or to practice a trade, and to provide the prisoners and his family with a source of income’ – the principle of work as an obligation and right (1857) • Prisoners must be educated ‘in the interest of society and [education is] an obligation to the prisoner’ – the principle of penitentiary education (1838) • Prisoners must be supervised and supported by ‘specialised staff possessing the moral qualities and technical abilities required of educators’ – the principle of technical supervision of detention (1850) • Imprisonment must be followed by supervision and support to ensure that the rehabilitation of the offender is completed – the principle of auxiliary institutions (Foucault 1977:269-270) 4 Institute for Security Studies Unsurprisingly, the seven maxims are also found in the White Paper on Correctional Services (2005). For nearly 200 years prison systems have been subject to reform and yet the same fundamental principles are repeated when a new reform process is announced (Foucault 1977:269- 270). We can thus conclude that the basic principles underpinning the purpose and function of prisons, and thus their fundamental nature has remained much the same for nearly 200 years. In addition, it was already known from about the middle of the 19th century that prisons did little to contribute to a safer society; rather they go a long way towards achieving quite the opposite effect. This necessarily begs the question: why do we have prisons and what can we expect of them? To answer this, we need to look at who stands to benefit from prisons – and the answer is simple: politicians and the private sector. WHY PRISONS ARE ATTRACTIVE TO POLITICIANS AND THE PRIVATE SECTOR Prisons have symbolic value; they communicate the message that government is tough on crime and is willing and capable of legally depriving citizens of their liberty because they have committed a crime and offended society. Prisons symbolise the state’s power over its citizens. More importantly, they communicate the willingness of the state to use its coercive power. In establishing and maintaining a prison system, the state goes to enormous lengths, perpetuating a huge bureaucracy consuming vast quantities of resources, in order to give effect to its coercive power. The exercise of coercive power is important to the state – it symbolises and demonstrates authority. The deprivation of liberty and the other, often incidental, ills associated with imprisonment also speak to a very human attribute, namely the need for retribution. Retribution is backward looking; it imposes pain as punishment for something that was regarded by the state (and society) to be wrong. We are led to believe that in imposing and executing this punishment, satisfaction is to be found. Concomitantly, the harsher the punishment, the more satisfaction we should experience. Imprisonment, as the harshest punishment (in the absence of the death penalty) that can be imposed, results in the highest level of retribution and must therefore result in the maximum amount of satisfaction. The deprivation of liberty, especially for long periods, is intended to have deterrence value. The legal provision for minimum terms of imprisonment for serious offences is largely motivated by the deterrence argument: would-be offenders will be deterred from committing such offences if they know the punishment will be severe, and the offender who is punished in this way will be deterred from committing such a crime again. From the politician’s perspective, the deterrence value of imprisonment is important, for it links with other values such as retribution and simplicity (see below). Prisons also remove people from society. They cannot commit any further crimes as long as they are in prison. Incapacitation gives a sense of security and a statement such as ‘we put criminals behind bars’ speaks to this sense of security. The subtext being that the state is in control: the state arrests criminals, prosecutes and convicts them, and removes them from society so that they can cause no further harm. It also communicates that the state takes care of law-abiding citizens and removes law-breakers from the benefits of being a free citizen. Prisons, as buildings, also have visibility value. We can see the buildings, even from a distance, and know that this is where people who have offended are kept. As monuments to law enforcement they have simultaneously a comforting value and a deterrent value. It is easy to scare children with a prison building: ‘this is where you go if you are naughty’. Occasionally we are allowed to see inside prisons when photographs and video footage are released, and these confirm the austere appearance of prison buildings. It is indeed when we see what people are doing inside prisons and what they have to endure that prisons attain their maximum visibility value. SA Crime Quarterly no 26 • December 2008 5 6 Institute for Security Studies Prisons create jobs for a large number of civil servants. The Department of Correctional Services (DCS) currently employs about 42 000 officials with plans to increase the number to 45 000. As an employer it is one of the largest in the civil service. Politically this is important and the previous regime knew this and used it extensively to create employment in rural areas; the number of prisons in the Free State and Boland bear testimony to this. Prisons consume vast amounts of resources to feed, clothe, care for and maintain prisoners. This is continuous. The resources are largely supplied by the private sector and paid for by the taxpayer. Securing contracts with the prison system can be extremely lucrative for private companies. There are, therefore, good reasons for the private sector to support prisons, or at least not to oppose them. When private companies start operating prisons, as is the case with the two private prisons in South Africa, profits can be of a significant magnitude. Prisons are also very costly to build. Building prisons requires specialist skills, high quality materials and far more rigorous standards than a residential building or other forms of accommodation. The increasing integration of information technology into prison architecture has created new features such as keyless prisons, biometric security and so on. Financially, construction and related companies can benefit greatly from prison construction, especially when the state embarks on a large-scale prison construction programme. Perhaps the most attractive attribute, from the view of politicians, is that prisons have simplicity value. By stringing together the retributive value, deterrence, the value of incapacitation and visibility value of prisons, it is becomes fairly easy for any politician to explain, in a convincing manner, why we need prisons and why we should have even more prisons. According to politicians and the private sector, prisons work. It is far more beneficial to politicians and the private sector to promote imprisonment than to tell the public that prisons have no or very limited value in so far as making society a safer place. The public has also, in general terms, been quite receptive to the use and expansion of imprisonment. For a South African politician it would be extremely risky to be perceived as being ‘soft on crime’. What ‘soft on crime’ actually means is not clear, although being ‘tough on crime’ is generally accepted to mean doing more of the same thing (i.e. long sentences and high imprisonment rates) and expecting a different result! PRISONS DO NOT REDUCE CRIME There is a growing body of evidence to indicate that large-scale imprisonment makes the situation worse (Clear 2007). Even the incapacitation argument is unconvincing. It has been shown that in order to affect a ten per cent reduction in crime in the UK, using a general incapacitation approach, would require a doubling of the prison population (Piquero et al 2007:17). One can only speculate about what this means in the South African context. Other research has also confirmed that imprisonment does not contribute to reducing recidivism (Gendreau et al 1999). Even when controlling for risk profiles, those offenders who were sent to prison had a higher re-offending rate than those who received a community-based sentence. Higher recidivism rates are also associated with longer prison terms. In short, this implies that imprisonment per se increases the recidivism rate and the longer the term, the worse the impact. It is also reported that imprisonment increases recidivism for low-risk offenders. From a policy perspective it is concluded that: Prisons should not be used with the expectation of reducing future criminal activity . . . therefore the primary justification for the use of prisons is incapacitation and retribution, both of which come with a ‘price’, if prisons are used injudiciously (Gendreau et al 1999:20-21). South Africa’s current imprisonment rate is around 340 per 100 000 of the population, which is one of the highest in the world for a democracy. We also know that the imprisonment rate is not SA Crime Quarterly no 26 • December 2008 7 evenly distributed across the population in respect of gender, age, race and socio-economic status. There is also reason to believe that certain geographical areas contribute disproportionately to the prison population.2 In the past 15 years we have also seen how sentence lengths have increased dramatically following the increase in the sentencing jurisdictions of district and regional courts, and the introduction of minimum sentences. South Africa is indeed using imprisonment injudiciously and we will continue to pay the ‘price’ referred to by Gendreau, Goggin and Cullen above. PRISONS DON’T MAKE A DIFFERENCE TO THE VIOLENT CRIME RATE Logically it follows that if more offenders, especially violent offenders, are imprisoned, the rate of violent crime would decrease. This will hold true if one is able to imprison especially those offenders who are most likely, if not guaranteed, to be repeat violent offenders. Similarly, it follows that if fewer offenders were imprisoned, the rate of (violent) crime would increase. While advocates of the incapacitation approach support this logic, reality does not support it. The following will attempt to illustrate this. For the purposes of this analysis the number of violent crimes reported annually to the SAPS will be compared to the number of offenders admitted annually to prison to serve sentences for the period 2003 to 2008.3 Violent crime is defined as murder, rape and aggravated robbery and used collectively as an indicator of violent crime in South Africa. Prison admissions refer to all sentenced admissions. The two data sets are presented in Figure 1. During the period 2003 to 2008, violent crime remained fairly stable; deviating no more than 6 per cent from the full term average in any one year. The number of sentenced admissions to prisons, however, dropped from 178 569 in 2003 to 94 566 by 2007/8; a drop of 47 per cent. These figures are startling in their simplicity and have a number of implications for our understanding of criminal justice. This discussion will not go into the possible reasons for the drop in admissions as they are complex and relate to, amongst others, the quality of crime detection, docket preparation, effectiveness of prosecutions and so forth. The comparison being made is restricted to the results. Firstly, there appears to be no link between the number of violent crimes reported annually and the number of sentenced offenders admitted to prison. The number of offenders imprisoned does Figure 1 Violent crime and imprisonment, 2003-2008 250 000 200 000 150 000 100 000 50 000 0 2003 2004 2005 2006 2007 2008 Prison admissions Violent crime 8 Institute for Security Studies not appear to have any direct and observable impact on the rate of violent crime. If this was not the case, the rate of violent crime should have climbed steadily during the period because fewer offenders were imprisoned, but this did not happen. Secondly, the effect of the so-called minimum sentences became visible in sentence profiles from around 2000 (Giffard and Muntingh 2007), but the intended deterrent effect of this legislation is also not visible in the number of reported violent crimes. If it were having the intended effect we should have seen a reduction in violent crime. Even if it is assumed that the decreases (in 2006 and 2008) in t he violent crime rate could be attributed to the deterrent effect of long prison sentences, increases in the violent crime rate of similar proportions are also observable in 2004 and 2007. What is clear is that imprisonment has a minimal (if any) deterrent effect and that the minimum sentences introduced in 1998 have similarly had no deterrent effect. There is simply no evidence of this. Thirdly, if the imprisonment rate did not have an impact on the rate of violent crime, then something else must have had. What this is, is not clear, but may relate to socio-economic conditions, the general character of a violent society, the quality of policing, the effectiveness of prosecutions and so forth. Given the low conviction rate for reported violent crimes (between 8,9 per cent and 12,6 per cent for the three offences used above) (SAPS 2008:114), it seems unlikely that imprisonment could have an impact on the rate of violent crime. Fourthly, whether the imprisonment rate is comparatively high (as it was in 2003) or low (as it was by 2008), it appears that imprisonment is being used injudiciously. Whoever are imprisoned do not appear to be the individuals driving the high violent crime rate, even if only a small proportion of them have ended up in prison in the past five years. WHAT NOW? It has been argued that the case for imprisonment is made by politicians and the private sector not because imprisonment reduces crime but because it is politically valuable and supports commercial interests. Society has in general been receptive to this message. It has also been shown that imprisonment, especially on the scale happening in South Africa, makes the situation worse and not better. Any argument stating that (increased) imprisonment will reduce crime must be viewed with extreme caution, as there is little evidence to support such a position. For example, the so- called ‘three-strikes-and-you-are-out’ legislation of California4 had no greater impact on serious or petty crime trends than what were predicted based on pre-existing trends (Piquero et al 2007:23). This harsh sentencing legislation was also found to be without deterrent effect (Piquero et al 2007:24). Using recent statistics it was shown that in the past five years the number of offenders sentenced to imprisonment dropped nearly 50 per cent but that that the rate of violent crime remained fairly stable. It was concluded that imprisonment does not impact on violent crime and that there are other variables affecting the rate of violent crime. It was similarly concluded that the intended deterrent effect of long prison sentences did not materialise. The solution does not lie in abolishing prisons because they are failed institutions, but rather to use them selectively, judiciously and with a clear understanding of their purpose and what they can realistically achieve. This requires a clear understanding by stakeholders, but especially the judiciary, of the place of imprisonment in a comprehensive and coherent crime control and reduction strategy. Regrettably such a coherent strategy has not emerged. The point of departure must be that imprisonment should be used as a measure of last resort. This means that all other options, not only penal sanctions, need to be assessed and exhausted before a person is deprived of his or her liberty (Muntingh 2007:7). This requires the highly selective and intelligent use of imprisonment, and above all, avoiding the SA Crime Quarterly no 26 • December 2008 9 over utilisation of imprisonment, especially on offenders who may benefit from other interventions or much shorter terms of imprisonment. Imprisonment must be regarded as the most severe sanction to be imposed when no other sanction would have been reasonably able to achieve the same results intended by the court in a less restrictive manner. Courts also need to be guided by an ‘imprisonment policy’ that defines the purpose of imprisonment in relation to other sanctions, the overall function of prison in society, the place of imprisonment in a constitutional democracy, the known risks of imprisonment, and what can realistically be expected as the outcomes of imprisonment. It has been noted by others that the size of prison populations is determined more by political sentiments and penal policy than by actual crime trends. The use of imprisonment as penal sanction should therefore not be regarded as the unrestrained and uncontrollable result of the interaction between crime trends and law enforcement.5 The opposite is in fact more accurate – the use of imprisonment is based on policy decisions emanating from particular philosophical positions. Imprisonment should therefore be used highly selectively to ensure that only those offenders for whom imprisonment will reduce the risk of re-offending, or who in reality do need to be incapacitated, are in fact imprisoned. How these offenders are selected is indeed another field of investigation, but that we need to depart from current practice is without question. REFERENCES Clear T 2007. Imprisoning Communities – How mass incarceration makes disadvantaged neighbourhoods worse. New York: Oxford University Press. Foucault M 1977. Discipline and Punish – the Birth of the Prisons. London: Penguin. Gendreau P, Goggin C and Cullen FT 1999. The Effects of Prison Sentences on Recidivism. Canada: Public Works and Government Services. Giffard C and Muntingh L 2007. The effect of sentencing on the size of the prison population. Cape Town: OSF(SA). Muntingh L (2007) Prisons in South Africa’s Constitutional Democracy, CSVR, Johannesburg. Piquero AR, Farrington DP & Blumstein A 2007. Key Issues in Criminal Career Research. Cambridge: Cambridge University Press. SAPS 2008. SAPS Annual Report 2007/8. Pretoria. ENDNOTES 1 ‘Politicians’ in t his article refer to the broad mass of political leadership in t he country. They are understood to play a formative role, through various means, in shaping the discourse on crime and punishment. They are regarded as critically important in shaping people’s views; either by providing innovative ideological leadership or by affirming already held beliefs. 2 As at the end of September 2008, 98 p er cent of the total prison population were males; nearly 37 p er cent were under the age of 25 years and Coloured people are imprisoned at double the rate of the national norm (Figures supplied by the Judicial Inspectorate for Correctional Services). Prisoners are also predominantly from the lower socio-economic strata. 3 Figures on prison admission were supplied by the Judicial Inspectorate for Correctional Services. Figures for the second half of 2008 were projected as these were not yet available. 4 It should be noted that California’s imprisonment is only slightly above South Africa’s at 474/100 000 o f the population. Wagner P 2003. The Prison Index: Taking the Pulse of the Crime Control Industry. Available at http://www.prisonpolicy.org/prisonindex/variation.html [last accessed 21 November 2008]. 5 Dünkel F and Van Zyl Smit D 2001. ‘Conclusion’ in F Dünkel and D Van Zyl Smit 2001. Imprisonment Today and Tomorrow, Kluwer Law, The Hague, p. 811; Gardner D 2002. ‘ Finland is soft on crime’, New Politics Vol XI No. 3 and Tonry M 2006. K eynote address, Conference on Sentencing in South Africa, Hosted by the Open Society Foundation (SA), C ape Town, 25-26 October 2007, C ape Town.