The London riots

Those involved in the English riots in the summer of 2011 have received far more stringent sentences than their crimes would normally have warranted.[1] The suspension of normal sentencing protocols and adoption of harsher punishment raises the issue of exemplary sentencing. Not only have sentences been more punitive, but there is also evidence that this was not a case-by-case response made by judges and magistrates, but rather followed an internal directive from public officials at the Courts and Tribunals Service that encouraged the abandonment of the usual guidelines for sentencing.[2] The question here, therefore, is whether exemplary sentencing can be justified ethically by the three major theories of punishment present in penal philosophy and theory. By applying normative theories to punishment in practice I shall show that exemplary sentencing under the current British penal system cannot be justified by any of the major strands of penal theory, and therefore that the rioters’ treatment was unwarranted.

There has not been, to my knowledge, a specific moral exploration of exemplary sentencing, though it is spoken to, and implicit in, many of the wider discussions of punishment. The three ideals of punishment that I will use to assess exemplary sentencing are consequentialism (or utilitarianism), retributivism and, more briefly, communicativism (or reformism).[3] These three are the most prominent theories, but also seem to adequately cover the character of Britain’s ambiguous penal code.[4] Dividing the theories like this is crude, as the debate has diversified and fractured over the last thirty years, but all penal theories, barring the extreme anti-punishment line taken by those such as Honderich,[5] can be located in relation to these three theoretical standpoints. I will use the ‘purest’ of each form (i.e. as free from crossover as possible) to evaluate exemplary sentencing.[6]

There are four assumptions which need to be made clear in order to limit the scope of this argument and to make it plausible as a response to public policy. The first is that the verdicts (as opposed to the sentences) were, in fact, correct. If the verdicts were incorrect then the issue of exemplary sentencing is, at best, secondary, as the debate surrounding the punishment of innocent people would take precedence. Secondly, the assumption that general punishment is justifiable on the basis of one, or a combination of the three theories that will be outlined below. Honderich’s conclusion that punishment itself is morally unjustifiable, although interesting, adds little to an assessment of actual public policy. Thirdly, the law, and in this case, British law, is just, and, as an extension of this, the normal sentencing guidelines are also just. That is not to say that this is the case in reality, but to sharpen the focus of the argument made here, which would be unduly complicated by questions of whether we have the right laws, the right sentences and so on.

Finally, as noted above, the sentencing guidelines were altered as a result of an explicit directive issued by public officials within the Ministry of Justice. The decision is therefore of public interest, as many, although not all, judges and magistrates who heard cases relating to the riots, judged in accordance with these revised guidelines. Although it is plausible that the judges ignored the guidelines and ruled harshly on the basis of individual evaluation, the issue of exemplary punishment as policy is made clearer if the personal sentiments of the judges are subsumed by an institutional decree. All judges, for the purpose of this argument, will therefore be assumed to have followed the guidelines, though this is likely to be untrue in reality. Although it cannot be denied that the directive was a guide rather than an order, it can reasonably be assumed that the guide shifted the sentencing limits above both the normal maximum and minimum thresholds, and thus signified a suspension of regular sentencing practices. It should also be noted that, at the time of writing, no one had been charged with rioting as a specific offence, but most commonly with burglary or violent disorder, offences which arise frequently in normal, non-riotous circumstances.[7]

So after those lengthy qualifications, and without the actual presentation of these arguments by proponents, I must present the strongest possible case that could be made by subscribers to each of the three theories of punishment and reveal the inadequacy of these justifications using their own internal standards of argument, which will prevent the possibility that one or more is legitimate. For example, if retributivism adequately refuted the consequentialist and communicative positions, but still remained susceptible to consequentialist attacks then a deadlock would exist in which disagreement over the theory of punishment would supersede the discussion of exemplary punishment. I will therefore show that each view fails by its own standards, thus avoiding this methodological difficulty.

The consequentialist position, broadly speaking, may best be explained as the belief that punishment is undeniably an evil, but justified by the prevention of greater evils through its incapacitative effect, individual and general deterrence.[8] The strongest argument in favour of exemplary sentencing from a consequentialist position would therefore run something like this:

whilst it is evil to inflict pain through punishment on anyone, and a greater evil still to impose an exemplary sentence (i.e. inflict greater pain) on person x (who committed burglary during the riots), the evil prevented, or conversely, good gained, by the imprisonment of person x through individual deterrence, general deterrence and the incapacitative effect, is greater than the evil done to person x through punishment. Consequentialism in its most basic form uses Bentham’s felicific calculator to measure whether exemplary sentencing does greater harm or greater good to the general happiness. In the case of the rioters, the consequentialist would argue three things:[9]

 I. Exemplary sentencing provides a greater individual deterrence than normal punishment by demonstrating the unpleasant consequences of criminal behaviour.

II. It sends a message to others that those involved in criminal activity in connection to rioting will suffer similarly harsh punishment if they imitate x’s crimes.

III. There is an incapacitative effect on the offender, who is incapable of committing crimes whilst imprisoned, thus protecting people from the potential harms he (it is reasonable to assume the offender to be male in light of prison demography) might have inflicted.

 These three results will decrease crime in both the short and the long term, as people will be deterred from riotous behaviour by the exemplary sentence imposed person x, who will in turn, be less likely to offend again and forcibly prevented from doing so whilst imprisoned.

This line of thought is susceptible to various non-consequentialist arguments, but it is manifest that it is also fails by its own standards. Consequentialism, by its very nature, is bound up with empirics. The only way to know of an action’s consequence is to study its results empirically, and so we must turn to the likely consequences of a policy of exemplary sentencing. Only if these three reasons demonstrably outweigh the harm done to person x, can the justification be persuasive. Due to the lack of specific studies of sentencing in connection to the riots, I must examine the general evidence on punishment and deterrence (individual and general).

The idea that severe sentences discourage individuals from reoffending is poorly supported by evidence. One effect of shifting the guidelines up the punitive scale is that those who would have normally received non-custodial sentences are imprisoned. In Britain, current statistics show that recidivism for those who serve non-custodial sentences is between 5 and 9 percent lower than those who serve custodial sentences of less than 12 months. This fact has led the Ministry of Justice to conclude that the individual deterrent effect of severe sentences is inconclusive.[10]  Considering the lack of evidence, whilst acknowledging the general difficulties of statistics, it still seems far from clear that individual deterrence is a sound reason to justify exemplary sentencing, as studies in the US have agreed.[11] Furthermore, it could be argued that imprisonment, a likely result of exemplary sentencing, will further criminalise offenders, ultimately leading to increased crime and greater harm.

It has also never been persuasively shown that increased sentences in Western democracies provide any deterrent to potential offenders. Studies have actually suggested that it has no positive impact whatsoever.[12] Given this uncertainty, the theory of general deterrence therefore has little merit in defending exemplary sentencing, as there is no conclusive evidence to suggest that severe sentences decrease recidivism or deter potential offenders. This is supported by the common-sense assumption that crimes are committed for reasons that are unaffected by the severity of the penalty, such as poverty, addiction and opportunistic impulse; reasons which are unlikely to be invalidated by the numerical differences of hypothetical sentences.

Incapacitative effect looks to perhaps have a stronger claim than either form of deterrence, as it is generally true that locking up convicted criminals is likely to decrease crime. This argument justifies punishment, but not exemplary punishment. In fact, in terms of promoting the good, imprisoning individuals at great public expense (nearly £40,000 per prisoner per annum),[13] beyond the normal guidelines, could create public harm by diverting resources that otherwise could have been spent, for example, on improving public services, in order to punish person x over the legally recommended level. This does not undermine consequentialism’s justification of normal sentencing practices, as it can, and has, been argued that the expectation of some form of penalty deters people in some cases, although not all.[14] Consequentialism flounders in the face of the evidence upon which it relies, and without clear and demonstrable benefits consequentialism cannot persuasively defend exemplary sentencing.

Modern retributivism, like consequentialism, is diverse, so I have selected what seems to be a fairly uncontentious version, which correlates with the majority of the arguments in this broad school of thought. Retributivism is, in essence, the application of the notion of ‘deserts’ to the case of punishment. People deserve to be treated in the same way as they have voluntarily treated others and so criminals who wrong others (either personally or generally) deserve the wrong they receive in the form of punishment.[15] It might seem that an argument that is founded on the notion of ‘just deserts’ would have no place for exemplary sentencing, but ‘just’ could be considered flexible, and retributivists would require it to be so if they wish to support exemplary sentencing. The argument in favour of exemplary sentencing, with regard to the riots, would therefore be as follows:

person y, who took part in the riots and its associated crimes, committed offences that wronged other individuals and y therefore deserves the wrong he in turn receives through imprisonment. This is the case for normal sentencing, but it can be extended for the case of the rioters and the exemplary sentences that they received. Committing a crime in normal circumstances is not the same as committing a crime in the context of a riot. The latter is a greater wrong than the former, because it contributes to social malaise and the damage done in a riot outweighs the damage done by the sum of its parts. For example, stealing one bike from Evans Cycles wrongs the shop owner (to some degree), but by being part of a larger theft, if only constituently, the shop owner is harmed more greatly, for example by losing her entire stock and thus being unable to trade for several weeks. The single crime committed by person y is therefore part of a more serious offence, for part of which y is responsible. An increased, or exemplary sentence is therefore justified because y deserves to be punished for the actual harm committed to the storeowner.

The above argument appears convoluted due to its pure retributivist nature. Many retributivists would incorporate an element of consequentialist deterrence to their argument, but since this has already been deemed insufficient to justify exemplary sentencing, I have stripped retributivism to its essentials in order to make this hypothetical argument as strong as possible. Retributivists, such as Rachels, have attempted to define under what circumstances the deserts an offender receives are, in fact, just. Of his four principles of justice: guilt, equal treatment in law, proportionality and excuses, the application of exemplary sentencing fails to meet at least two.

In one case, two men were jailed, each for more than four years for creating an online group that encouraged rioting on a social media site, despite no rioting occurring as a result of this incitement.[16] Although the two were guilty of incitement to riot, the sentences they received comply with neither the equal treatment nor the proportionality criteria laid out by Rachels. Four years imprisonment for inciting a riot that never occurred is not only disproportionate, but fails to treat the same offences in the same way, as such a crime would have surely have been treated with greater leniency, if not gone unnoticed, outside the context of the riots. If the law ceases to treat like for like cases in the same manner then the deserts that it sanctions cease to be just. Considering the offences of an individual as part of a broader offence, to which they did not necessarily autonomously subscribe, fails to treat them as ends, but rather as constituent means, thus breaching a core Kantian principle from which retributivism is, in part, derived.[17] The reality further supports this conclusion, as no rioters have been charged with rioting, meaning that the charges that they have faced are comparable and equivalent to their non-riotous counterparts and should therefore be subject to the same legal consequences.

The final branch, communicative or reformist theory, is less formally distinct than either of the theories discussed previously and is often incorporated into retributivist or, more commonly, utilitarian arguments to strengthen their positions. Duff, for example, who advocates a communicative theory, adds a concept of penance to a communicative notion of desert.[18] I will, however, use Duff’s account as the basis, as a purely communicative account neither exists, nor would be plausible, but ignore its retributive element which has been refuted above. Because its role in British law is secondary to the previous two theories and its refutation is so simple, it can be quickly discredited. Briefly then, communicative theory suggests that punishment is the ‘communication of deserved censure’ (backward-looking retributivist element) and serves to persuade offenders to repent the wrongs they committed for its own sake, rather than instrumentally (forward looking communicative element).[19] The reformist argument for exemplary sentence would be as follows:

taking part in crimes in the context of a riot signifies a greater social disregard and a more profound problem for the individuals than offences committed outside this context. That z committed burglary during the riot, suggests that he requires further rehabilitation and must experience greater repentance than someone who committed burglary under non-riotous conditions. The longer sentence is therefore justified because it is required in order to successfully rehabilitate the offender and enable z to repent fully. In addition, it communicates censure to the population by condemning this type of offence.

Once again, distilled from the admixture of actual penal reality, this argument appears somewhat alien, but perhaps also logically sound. Although the communication of censure holds true, the actual punishment of the individual cannot be defended. When applied to the treatment of the rioters communicativism fails on logical and practical grounds. If the aim of punishment is for convicts to recognise their wrongdoing then time is not a factor in enabling repentance. Offenders may repent immediately, and thus fulfil the aim of punishment without imprisonment, requiring only the retributivist account of deserts to be fulfilled. Equally offenders may never repent and therefore indeterminate sentences would be required to meet the communicative criterion of punishment. Exemplary sentences would not increase the level of penance and may in fact increase resentment as a result of perceived injustice. Secondly, the British penal system offers little in the way of reform, as the recidivism rates have already shown. Only in a truly rehabilitative system could Duff’s theory operate successfully, and considering the mixed motivations of British punishment and the poor provision of rehabilitation, it seems that exemplary sentencing cannot be justified by a communicative theory of punishment.

Having dealt with each individually it might appear plausible that a combination of the three, as exists in the ambiguous British legal system, could provide a reasonable justification for exemplary sentencing and thus rebut my case. It seems, however, strange at best and absurd at worst, to believe that supposedly coherent theories of punishment which cannot defend exemplary sentencing on the basis of their own standards, would be better equipped to justify exemplary sentencing when muddled together in unknown quantities. The ad hoc and historical nature of the British legal system leaves this open to doubt, but I see no good reason to assume that such a gallimaufry would succeed where coherent programmes have failed.

On the basis of this account then, the exemplary sentences used against the English rioters were ethically unjustifiable within the context of the British penal system. Consequentialism and communicativism are unable to bear the weight of evidence and reality, whilst retributivism inherently struggles to justify the injustice of disproportionality and differential treatment. A further argument, for which there is little space here, is made by Carlen, who suggests that sentencing must be sensitive to social conditions and inequality in order to ensure feasibility and justice.[20] Whilst the argument I have made discredits the justification for exemplary sentencing, Carlen’s account may perhaps offer reasons beyond parity of treatment and towards leniency for the rioters, who after all, hailed not from Hampstead, Westminster and Chelsea, but Hackney, Walthamstow and Croydon. Exemplary sentencing in Britain is morally unjustifiable under the present system. A more socially sensitive and coherent response might consider clemency over severity as an ethically sound response to the related problems of social and criminal unrest.


[1] Owen Bowcott, ‘Facebook cases trigger criticism of ‘disproportionate’ riot sentences’ in The Guardian, 17 August 2011.

[2] Bowcott, ‘Magistrates were told to send rioters to Crown court, emails show’ in The Guardian, 14 September 2011.

[3] Anthony Duff and David Garland, ‘Introduction’ A Reader on Punishment (New York: OUP, 1994), pp.6-8.

[4] Nigel Walker, Why Punish? (Oxford: OUP, 1991), p.8 and H.L.A Hart, ‘Prolegomenon to the Principles of Justice’ in Punishment and Responsibility (New York: OUP, 2008), p.2.

[5] Ted Honderich, Punishment: the supposed justification revisited (London: Pluto, 2006).

[6] Matt Matravers, ‘Introduction’ in Punishment and Political Theory (Oxford: Hart, 1999), pp.1-4.

[7] ‘UK riots: the demographics of magistrate cases and convictions’ in The Guardian, 18 August 2011.

[8] Joel Feinberg, ‘The Classic Debate’ in Feinberg and Coleman (eds.) Philosophy of Law, 7th Ed. (London: Wadsworth, 2004), p.799.)

[9] C.L. Ten, ‘Fantastic Counterexamples and the Utilitarian Theory in Feinberg (ed.) Philosophy of Law, pp.804-805.

[10] ‘2011 Compendium of re-offending statistics and analysis’, Ministry of Justice, 10 May 2011.

[11] Blumstein, Cohen and Nagin (eds.), Deterrence and Incapacitation, National Academy of Sciences, Panel on Research on Deterrent and Incapacitative Effects (Washington, 1978), p.66.

[12] For the case against the deterrent effect of sentence severity see, Anthony Doob and Marie Webster, ‘Sentence Severity and Crime: Accepting the Null Hypothesis’ in Crime and Justice, vol. 30 (2003), pp.143-195, for the opposed view see, D.E. Lewis, ‘The General Deterrent Effect of Longer Sentences’, in British Journal of Criminology, Vol. 26, No.1 (1986).

[13] House of Common Debates, 25 November, 2008, Col. 1307W.

[14] Nigel Walker, ‘Modern Retributivism’ in Gross and Harrison (eds.), Jurisprudence: Cambridge Essays (Oxford: Clarendon, 1992), pp.74-75.

[15] James Rachels, ‘Punishment and Desert’ in Hugh LaFollette  (ed.), Ethics in Practice, (Oxford: Basil Blackwell, 1997), pp.473-474.

[16] Helen Carter, ‘Pair jailed for four years for using Facebook to incite disorder’ in The Guardian, 16th August 2011.

[17] Walker, Why Punish?, p.7.

[18] Duff, Punishment, Communication, and Community (New York: OUP, 2001), p.30.

[19] Ibid.

[20] Pat Carlen, ‘Preface to Carlen, ‘Crime, Inequality and Sentencing’ in Duff, Punishment, pp.306-308.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s