Rolf Harris has been sentenced to five years and nine months in prison for sexual offences he committed at various points in the 60s, 70s and 80s. There has been public outrage at the supposed leniency of his sentence, which will now be reviewed by the Attorney General to determine whether it will be sent to the Court of Appeal.
Why is it shorter than might be expected?
Part of what makes the sentence appear shorter than might be expected – given the number of offences, each of which attracted a custodial term – is the application of the ‘totality principle’. This is a principle set out in the sentencing guidelines of England and Wales that directs judges when they are sentencing an offender for more than one offence. The totality guideline states that:
All courts, when sentencing for more than a single offence, should pass a total sentence which reflects all the offending behaviour before it and is just and proportionate. This is so whether the sentences are structured as concurrent or consecutive. Therefore, concurrent sentences will ordinarily be longer than a single sentence for a single offence.
Although there is no hard-and-fast rule for when sentences should run consecutively or concurrently, the guideline indicates that concurrent sentences will usually be appropriate where offences arise out of the same incident or facts, or when there is a series of offences of the same or similar kind, especially when committed against the same person. In such cases, ‘the sentence should reflect the overall criminality involved.’ Accordingly, some of the sentences imposed by Mr Justice Sweeney are intended to run consecutively and others concurrently.
Whilst the concept of concurrent sentences is theoretically challenging – What does it mean to serve two sentences at the same time? Why should it make a difference if it is the same or a different victim? How do we determine the ‘overall criminality involved’? – I will not address this issue here. Instead, I want to consider the other reason that led to Harris’ overall sentence being less than we might have anticipated: the application of historical sentencing levels to historical offences.
Historical sentencing maxima
Mr Justice Sweeney explained that had Harris committed the very same offences today, he would have faced a significantly higher penalty. For example, five of the 12 counts for which he was sentenced would today attract a starting point of around eight years’ imprisonment (given the psychological harm caused in the particular case), and a maximum sentence of life imprisonment. This is in significant contrast to the historical maximum sentence for these particular five counts of two years’ imprisonment. The Sexual Offences Definitive Guideline states that, although the court ‘should not seek to establish the likely sentence had the offender been convicted shortly after the date of the offence’, the sentence is ‘limited to the maximum sentence available at the date of the commission of the offence’.
What is the justification for using historical sentencing maxima?
Driving at least part of the public outcry might be a difficulty to see why Harris should not be sentenced in line with our current practices. After all, changes to sentencing legislation and guidelines are supposedly made in the interests of better serving justice. Why appeal to out-dated sentencing practices, which are now seen as inadequate or unjust?
The first potential justification is analogous to the ‘legal certainty’ or ‘fair warning’ argument. So that people can be equipped to chose between lawful and unlawful conduct, laws must give the ordinary person a reasonable opportunity to know what is prohibited, so that he may direct his actions accordingly. If the law does not make it clear that a certain act is illegal, no person performing that act can be punished for it. This is the principle that prevents people from being punished for acts that have become crimes, but that were not prohibited at the time the acts were committed.
The same sort of argument has been made with respect to increases in the maximum penalties attached to specific offences: people have to know what ‘price’ is attached to the conduct they are considering engaging in. To increase the penalty for a particular offence and then impose this greater penalty on those who committed the offence prior to the increase is thought to be unfair – they had not been warned that they might incur such a great cost for their actions. Drawing on such arguments, sentencing policy in England and Wales is committed to the principle that changes in the law that create new offences, or increase the maximum penalties for existing offences do not apply retrospectively to crimes committed before the change in the law.
Is it convincing?
But do these arguments convince? The principle that people should not be convicted for acts that were not illegal at the time they were committed seems to be convincing. After all, it seems that people ought to know the boundaries of proscribed conduct in order to regulate their conduct accordingly. Furthermore, we might also observe that the possibility of retrospective criminalization could present significant opportunities for abuse of power.
However, the principle that updated sentencing tariffs should not be applied retrospectively is more controversial. The first argument that could be made is that although most people have a good idea of what is legal and illegal, they do not have a thorough knowledge of the sentencing guidelines and of all the factors that aggravate or mitigate a sentence. Thus, the argument might be that we in fact don’t give citizens fair warning about the likely penalty they would face for committing some offence (and so perhaps this is not a principle we’re committed to here). The sentencing decision is a complicated one, inevitably involving some degree of subjective judgment despite attempts to rein in undue scope for judicial discretion. Judges themselves are often under scrutiny for not delivering consistent sentences, so how can we assume that citizens know what they would be sentenced to were they to commit an offence? However, although precise sentence levels are complicated to determine, the maxima are clearly specified and strictly adhered to. There is at least certainty over this.
Asymmetry with decreased sentencing maxima
A further objection might point to the apparent asymmetry, which could be seen to indicate an inconsistent approach. Sentences that were harsher in the past are not routinely reverted to in cases where the offence was committed at the time of the harsher punishment. An extreme example would be the abolition of the death penalty for murder in England in 1965. After this date, the maximum sentence for murder was life imprisonment, regardless of whether the murder occurred prior to the Murder (Abolition of Death Penalty) Act 1965.
A more recent example is the significant reduction in the severity of the sentence for so-called ‘drug mules’ – those, usually naïve, individuals who have been exploited by organised criminals to carry drugs. In revising the sentencing guideline for drug offences, the Sentencing Council felt that, for this group of offenders, sentencing levels were sometimes disproportionate to the level of culpability involved. In most cases, they do not really understand the consequences of what they are doing and are treated as ‘disposable’ by their exploiters. The Council’s aim with this specific group of offenders was to ensure a downward shift in sentences to ensure they are proportionate to those received by other drug offenders.
This change to the sentencing levels was intended to come in to force as soon as the guideline was implemented and, in fact, the number of outstanding cases (i.e. offenders who were waiting to be sentenced) provided motivation to implement the changes as soon as possible, so that their sentences would be more lenient despite the offences having occurred at the time of the harsher sentence levels.
So, it seems that there is an asymmetry: when the maximum sentence for a particular offence goes up, it is not applied retrospectively, but when the maximum sentence goes down, it is. The inconsistency, if defensible, must be based on an argument that it is not permissible to punish offenders more harshly than they bargained for, even if this is now considered to under-punish, but it is permissible to punish offenders less than they might have expected, even though they took the risk of receiving a higher punishment. Justice should be done where possible – over-punishment will be avoided – but sentencing offenders more harshly than they were warned is indefensible – there must, regrettably, be under-punishment in such cases.
Why do sentence levels change?
Perhaps reflection on the reasons why changes are made to the severity of sentences can aid discussion. It seems there are two possible reasons why sentences change: 1) it could be that some offences actually become more (or less) serious and/or 2) we now have a better understanding of the seriousness of different offences and realize that the existing sanctions are inadequate.
The first possibility invites the idea that the seriousness of some or all wrongdoing is culturally relative. This sort of view is encapsulated in the ‘but it was different in the 60s’ argument that has occasionally been voiced in relation to Rolf Harris’ offending. The idea is that groping and other low level sexual assaults were ‘part of the culture’ or even considered borderline acceptable, and that this to some extent excuses the perpetrators of these offences in that era. (We should note that not all of Harris’ offences were at this lower level.) But how could it be that the same kinds of offences committed today – the same acts, with the same intent – are more serious than they were when committed in the 60s? It would have to be the case that the offences committed in the past caused less harm to the victims or that the perpetrators were not as blameworthy for committing the offences as they are these days. That the offences committed in the 60s were less harmful is extremely doubtful. The victims’ testimonies speak of shattered lives and debilitating anxiety. Cultural differences (if they really were that big) would only make the harm of sexual abuse worse: on top of the terrifying and humiliating experience, victims may have felt they were expected to endure the abuse or even that they in some way deserved it. A culture that didn’t take sexual abuse as seriously is not one in which it harms the victims any less, but in fact compounds the harm by dismissing the incidents as petty. Sexual abuse was not less serious in the past even if it was, regrettably, taken less seriously.
The second, more likely reason sentences change is that we learn more about the severity of the harms caused or about the factors that lead people to commit certain offences (e.g. in the case of drug mules). In the case of sexual abuse, it is not just attitudes that have changed but the psychological harm it causes is better understood. Essentially, we now know that sexual abuse is more serious than historical sentences indicated, and sentencing practice has been updated to reflect this.
So should historical sentences be imposed for historical crimes?
It’s not the case that Harris’ crimes were less serious at the time they were committed. The sentences that were imposed in the 60s and 70s were unduly lenient – the seriousness of the offences had not been fully appreciated. So why should manifestly inadequate sentences be passed nonetheless? If part of the role of imposing a sanction is to communicate to the offender and society how bad – how blameworthy – the offence was, surely, we might think, the assessment should be made by this society at this time. To impose historical sentences is to convey an erroneous message from the past. Even if Harris was somehow less culpable because of attitudes at the time of the offence (although this is strongly contestable), this would also provide a reason to communicate to him and the general public the actual seriousness of his offences – to set the record straight.
However, the fair warning principle might generate an argument along the lines: ‘if those involved in passing sentencing legislation didn’t get the assessment of seriousness right 50 years ago, then why should we expect offenders to have done so? Citizens should not be expected to be better assessors of ordinal or cardinal proportionality than judges and penal theorists.’
We now, rightly, take sexual abuse more seriously, recognizing the immense psychological harm that it can precipitate. Principles conflict: communicating the proportionate level of disapprobation to the offender and society is a central aim of punishment; giving citizens fair warning about what is illegal and what penalties attach to proscribed conduct is a central requirement for the rule of law. If Harris’ case does get referred to the Court of Appeal, judges will have to decide which of these principles should prevail.