The long sentences passed on the offenders in the August riots should have re-opened the question of whether it can ever be right to pass deterrent sentences – but they didn’t. Deterrent sentences are substantially longer than those usually passed for the same offences and are intended to send a message to others not to commit them. The classic case was the execution of Admiral Byng in 1757 for ‘failing to do his utmost’ against the French, which led Voltaire to remark that in England they shoot admirals from time to time ‘pour encourager les autres’.

But what if the others are not encouraged? Or not many of them? In this post I ask if we have any evidence from which to determine whether deterrent punishments work at all; whether they work enough to justify the disproportionate sentences passed in the name of deterrence; and whether there can be a moral case for giving offenders more than their crimes call for in exchange for very uncertain results.

First, some legal background: Section 142 of the Criminal Justice Act 2003 requires Judges to ‘have regard’ to deterrence as one of the five principles of sentencing:

(a) the punishment of offenders,

(b) the reduction of crime (including its reduction by deterrence),

(c) the reform and rehabilitation of offenders,

(d) the protection of the public, and

(e) the making of reparation by offenders to persons affected by their offences.

The Lord Chief Justice (Lord Judge) based much of the reasoning in his long and fascinating judgment in the August 2011 riot appeals (R. v Blackshaw & Others) for upholding most of the sentences, citing Section 142. He also said

It is not, of course, possible now, after the events, for the courts to protect the neighbourhoods which were ravaged in the riots or the people who were injured or suffered damage. Nevertheless, the imposition of severe sentences, intended to provide both punishment and deterrence, must follow.

Note the elision of punishment and deterrence, and the simple, unargued imperative ‘must follow’: he takes it for granted that sentences for crimes committed in the course of public disorder ‘must’ be met with deterrent sentences. While he drew distinctions between the degrees of involvement in the disorder, the overriding need to pass severe deterrent sentences is a theme that recurs throughout the judgment. He was not impressed by defence arguments in the cases of the Warrington and Northwich Facebook non-riots: although no one actually turned up in answer to the call to riot (except the police) the offences still called for four-year sentences. It was the potential to cause disorder and the foreseeability of serious harm that counted, the fear that it engendered, and the ease of summoning rioters through social media websites was a serious cause for concern. The Lord Chief said

The [sentencing] judge was fully justified in concluding that deterrent sentences were appropriate. These offenders were caught red-handed. For the citizens of Northwich and Warrington that was just as well, because as we have explained, and the guilty pleas acknowledged, neither offender was joking when the Facebook entry was set up.

It would be hard to disagree with his view that relatively minor offences like shoplifting or receiving stolen goods became far more serious than usual when committed in the midst of widespread public disorder, and deserved greater punishment because each in its small way contributed to the spread of lawlessness and public fear. But the conceptual questions about deterrent sentences that are seldom asked, and were not in the riot appeals: how can you tell if a deterrent sentence works? And if you can’t tell, is it right to pass one?

No one knows if English admirals ‘did their utmost’ after Byng’s death because they thought they would be shot if they didn’t; or whether they did it anyway. No one knows today whether Blackshaw’s four years have stopped other Facebookers from inciting their friends to riot, or will do in future (although the odds are against Blackshaw doing it again – as they would probably have been if he’d got half the sentence.) What a judge is really saying to an offender when passing a deterrent sentence is: your offence is worth two years; but there’s too much offending like this and people are frightened of it; I have duty to do what I can to reduce crime; therefore you’re getting four years to send out a message to others.

So the offender given a deterrent sentence is being used as a penal billboard to advertise what will happen to others. Is it right to give people longer sentences that their offences really merit (if one concedes that offenders are people) as a means to an end, rather than just punishing them for what they have done? The argument has a long philosophical pedigree. A Kantian would say no, it is never right to treat people in that way, they must never be considered as instruments to achieve an end but as ends in themselves. They should be sentenced for what they have done – severely if necessary – and only for what they have done. But then few criminal lawyers are Kantians.

On the other hand, the utilitarian or consequentialist justifies deterrent sentences on the basis they increase the sum of human happiness (or at least reduce misery), even if that means using the offender instrumentally and giving him more than he deserves. It is a price worth paying. But the utilitarian has the onus of proving that it works: there’s no point in doing something does not verifiably produce the result you want. This is exactly the problem with deterrent sentences: what evidence do we have to show that they work? The problem is symptomatic of one of the larger problems associated with utilitarianism:  by what calculus do you measure actions that cause present pain against the future benefits they are forecast to bring? Anyone advocating a dose of present pain surely has the onus of proving at least two things:  (i) there is evidence or at least a compelling reason to believe that it will, in fact, bring future benefits; and (ii) such benefits outweigh the present pain by a wide enough margin enough to justify inflicting it.

In Admiral Byng’s day you could be hanged for many offences including theft, which was prevalent. Did hanging deter?  You can argue it three ways: no, because people kept on stealing; yes, because there would have been more stealing without it; maybe it worked a bit, and there was less stealing as a result. How does anyone know? And if you lack the evidence for a firm conclusion, is there any justification? My view is that in a matter as important as taking away liberty, the case for deterrence can only be made out if there is clear evidence that large numbers of potential offenders take notice of the billboard carried by people like Blackshaw, and desist from offending. Without that evidence, deterrence is just hit and hope.

[A shorter version of this post appears in  The Justice Gap –]



  1. I’m curious about a piece of logic in your argument.

    If “deterrent sentencing” means selecting a sub-group of offenders and giving them longer sentences than will be given to all other offenders in comparable circumstances – in other words, “the next offender can expect a shorter sentence” – I can see the problem, both morally and philosophically.

    But if “deterrent sentencing” means that, for some classes of crime, the tariff for all offenders must include a deterrent element, I think it becomes less obvious that the policy is objectionable. This second interpretation of “deterrent sentencing” suggests a need for more careful thinking.

    You agree with the court’s view (and mine) that “relatively minor offences like shoplifting or receiving stolen goods became far more serious than usual when committed in the midst of widespread public disorder”. But if (hypothetically) we could know that there will never again be another riot, can we not say that extra seriousness of the shoplifting is less than if we face the fear that others may riot in future?

    In other words, I’m positing three levels of seriousness, in ascending order: (1) isolated shoplifting; (2) shoplifting in the world’s last ever riot; and (3) shoplifting in a riot which could be followed by other riots. Can we not say that the sentence for (3) should exceed that for (2)?

    Generalising away from riots: You ask “whether there can be a moral case for giving offenders more than their crimes call for” and answer the question (later in the piece) with your view that “[offenders] should be sentenced for what they have done – severely if necessary – and only for what they have done.” I can agree with that, but I am left asking: how do we determine what a crime “calls for” and what is the sentence “for what they have done”?

    Why shouldn’t the principles we adopt for answering that question include an element for deterrence, so long as it is applied equally in all (comparable) cases?

    1. Simon: thank you for this comment. I think that characterising a sentence passed on a particular offender as deterrent is different from raising the tariff across the board. (Another feature of Blackshaw was the Lord Chief expressed disapproval (in the nicest possible way) of a mere Circuit Judge (even if he was the Recorder of Manchester) presuming to set tariffs for the various offences committed during the riot. The LCJ rightly pointed out that only the CoA has the responsibility for setting tariffs.) In more normal times, Judges will say they are passing deterrent sentences because, for example, a particular crime is prevalent in the local area. If the incidence goes down, so does the sentencing. There is not usually a ratchet effect.

      So I think the objection remains. But even if there was a generalised increase, what I regard as the fundamental problem would still be there. Take murder – always met with a life sentence, or even more starkly in parts of the USA – with the death penalty for certain types of murder. The tariff could not be higher – but how to measure the effect? How many murders are prevented? How many unprevented murders must be committed before one doubts the efficacy of deterrence?

      As for what non-deterrent sentence a crime may call for: the simple answer is that precedent or common consent or statute or more and more the guidelines issued by the Sentencing Council set the tariffs. That’s the legal answer. On the philosophical level – more complex of course. Back to Thrasymachus?

      I don’t have an answer – just thought the questions were worth asking.

      1. 1 Like you, I was asking questions, rather than presuming to answer them. I have no expertise in criminal law or sentencing (as you may have gathered), but sometimes an outside view, from first principles so to speak, can add something (or maybe not, in my case?).

        2 I can certainly see the force of an argument against a court giving without warning higher sentences for crimes that have started to increase locally. The first offenders to suffer the higher extra-deterrent* punishment seem to have been treated on a different footing to the precedessors. But I can also see a counterargument which says that part of the deterrent effect is the practice of courts increasing the tariff (by a proportionate amount) when the crime levels rise.

        * I say “extra-deterrent” because I would argue that all sentencing includes a deterrent effect. The temporary, localised increase is an “extra”.

        3 It is, of course, valid to ask whether the deterrent effect is working. But I don’t subscribe to the view that a rule or regulation is permissible only if you can prove that it works. (Although, if one could prove that it doesn’t work, the practice should stop.) Some things can’t be proved either way and a decision has to be taken, nonetheless.

        4 Yes, obviously, precedent, statute and/or common sense should feature in the determinantion of what constitutes an approriate sentence. I think my (implicit) point was that, if precedent and/or common sense dictate that deterrence is a relevant factor (and if statute permits it) the notion of what a crime “calls for” automatically includes the deterrent element. I would go further: even if statute banned deterrence, I think that could only apply to what I called above “extra-deterrence”. I don’t think you can remove normal deterrence from sentencing, unless you abandon all punishment.

        5 I have never been in favour of capital punishment. It has no place in my argument.

        I am not aware of anything in my first submission above which suggested a ratchet effect, so I am not sure why you introduced the negative of that.

  2. Just realised that I had completely overlooked the second sentence of your reply (“I think that characterising a sentence passed on a particular offender as deterrent is different from raising the tariff across the board.”) If that is right, it seems to drive a coach and horses through much of what I have written! I may need to think again … but I shan’t use up the hospitality of your blog to do so.

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 )

Google+ photo

You are commenting using your Google+ 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 )


Connecting to %s