Academics nowadays must pay careful attention to the impact of their published work, though “impact” is narrowly and rather strangely defined, for some purposes at least, as the number of times an article is cited by other articles. What brings this to mind is that I recently flunked the impact test in a big way, albeit in a political context. A few weeks ago, I gave a public lecture here at University of Galway where I argued, among other things, that sentencing should not only be consistent and proportionate, but moderate as well. (I am reliably informed that the lecture in question can be seen, heard and read at bit.ly/4553MOd, whatever that may be, and it can also be accessed through the Law School’s twitter account). The gist of my argument was that guidelines can be effective in promoting consistency and proportionality, but moderation calls for the involvement of other actors, mostly at political level. The severity or otherwise of a proportionate sentence depends on the legislatively prescribed maximum sentence. Ideally, therefore, we should have a general review of maximum sentences (a rather forlorn hope, I fear). Failing that, new maximum sentences should be set as low as possible, and existing ones should not be increased without a strong and compelling reason.
No sooner was this said than the then Minister for Justice announced that in addition to increasing the maximum sentence for assault causing harm from 5 years to 10 years, it was proposed to increase the maximum for assault on a member of the Gardai from 7 years to 12 years. This will apparently be accomplished in the Criminal Justice (Miscellaneous Provisions) Bill 2022, currently making its way through the Seanad. So much for impact! (Actually, I see the sense in increasing the maximum for assault causing harm to something more than 5 years because of the range of conduct the offence covers, but am not convinced of the need to go as high as 12 years for assaults on Gardai or ramming Garda cars.)
However, a further justification has now arisen for moderation in the use of prison sentences, only this time the courts are the key decision-makers. The justification to which I am referring is prison overcrowding.
According to the website of the Irish Prison Service which provides a daily statistical profile of the current inmate population and available bed capacity, there were 4,583 prisoners in custody on 2 June 2023 which meant that, overall, the prison system was operating at 102% capacity. Limerick Women’s prison was operating at 157% capacity (it was 175% a week ago) while Mountjoy Women’s Prison was operating at 111% capacity. On 31 January 2022, the total number of prisoners in custody was 3,835 (88% capacity) which means that, in less than 18 months, the prison population has increased by 750. For various reasons which I outlined in the aforementioned lecture, this upward trajectory in prison numbers is likely to continue. Some additional prison space may be needed, but this cannot be the full answer. Parsimony in the use of imprisonment and in setting prison terms is also essential for keeping the prison population within manageable bounds and for maintaining a humane and productive prison regime.
A recent decision of the English Court of Appeal underscores the legitimacy of treating prison overcrowding as a relevant sentencing factor. In R v Arie Ali [2023] EWCA Crim. 232, the appellant was sentenced to six months’ imprisonment on pleading guilty to assaulting an emergency worker (an officer in a prison where the appellant was then serving a sentence for an earlier offence). The assault was not particularly serious, and there were mitigating factors. The appellant had pleaded guilty, although belatedly. He wasn’t charged until 16 months after the assault and several months after being released from prison for the earlier offence. He had completed his licence period successfully and engaged well with the rehabilitation services. Moreover, because of Covid and industrial action taken by members of the criminal bar, there was a delay in the case finally coming to trial. The offence was committed on 20 September 2019, the appellant was charged in February 2021, sentence was imposed on 6 February 2023 and the appeal was heard and determined on 3 March 2023.
There were three grounds of appeal, but the Court accepted only one, namely, that the trial judge erred in imposing a sentence of immediate imprisonment. Normally, an offence of this nature would merit imprisonment, but the Court identified three factors that marked out this case as exceptional. First, there was the delay in charging and sentencing. Secondly, there was a realistic prospect of rehabilitation as evidenced by the appellant’s conduct since being released from prison in July 2020. Thirdly, and most relevant for present purposes, the appellant was being sentenced at a time when the prison population was very high.
Ireland is not the only country experiencing prison overcrowding right now. Courts in many countries are dealing with a post-Covid backlog of criminal cases which is one factor contributing to surging prison populations. In England, in November 2022, the Government had requested the use of 400 police cells to hold remand and sentenced prisoners (and there is a statutory basis for this arrangement). Further, less than three weeks after sentence was imposed in Arie Ali, the Deputy Prime Minister wrote to the Lord Chief Justice, saying:
“You will appreciate that operating very close to prison capacity will have consequences for the conditions in which prisoners are held. More of them will be in crowded conditions while in custody, have reduced access to rehabilitative programmes, as well as being further away from home (affecting the ability for family visits). Prisoners held in police cells under Operation Safeguard will not have access to the full range of services normally offered in custody, including rehabilitative programmes.”
The Court of Appeal in Arie Ali quoted this statement and also the following passage from R v Manning [2020] EWCA Crim. 592; [2020] 4 WLR 77; [2020] 2 Cr App R(S) 46, an important authority on sentencing during the Covid pandemic:
“We are hearing this Reference at the end of April 2020 when the nation remains in lockdown as a result of the COVID-19 emergency. The impact of that emergency on prisons is well-known…. The current conditions in prisons represent a factor that can properly be taken into account in deciding whether to suspend a sentence. In accordance with established principles, any court will take into account the likely impact of custodial sentence upon an offender and, where appropriate, upon others as well. Judges and magistrates can, therefore, and in our judgment should, keep in mind that the impact of a custodial sentence is likely to be heavier during the current emergency than it would otherwise be…. Applying ordinary principles, where a court is satisfied that a custodial sentence must be imposed, the likely impact of that sentence continues to be relevant to the further decisions as to its necessary length and whether it can be suspended. Moreover, sentencers can and should also bear in mind the Reduction in Sentence Guideline. That makes clear that a guilty plea may result in a different type of sentence, or enable a magistrates’ court to retain jurisdiction, rather than committing for sentence.”
(What I have quoted here is the Weekly Law Reports version rather than the garbled version that appears in Arie Ali).
Manning arose from an undue leniency application in respect of a 12-month suspended prison sentence imposed on an adult male for sexual offences. The Court of Appeal found this sentence to be unduly lenient and substituted a two-year sentence which it also suspended in full.
Both Arie Ali and Manning reflect the same principle: custodial conditions may affect sentencing decisions. They might, as in those cases, justify a penalty other than immediate custody or else a reduced custodial term. As Lyndon Harris points out in his comment on Arie Ali ([2023] 6 Crim. L. R. 431), there is nothing new about this approach which, as he says, is reflected in earlier authorities such as R v Bibi [1980] 1 W.L.R. 1193 and R v Seed and Stark [2007] EWCA Crim. 254; [2007] 2 Cr. App. R. (S.) 69. In the latter case, Lord Phillips C.J. had said:
“In particular, when considering the length of a custodial sentence, the court should properly bear in mind that the prison regime is likely to be more punitive as a result of prison overcrowding.”
In fact, in the aftermath of Arie Ali, the Chair of the English Sentencing Council issued a statement referring to “well-established sentencing principles” which “include a requirement for the court to take into account the likely impact of a custodial sentence on the offender and, where appropriate, on others such as children or other dependants.” (I am indebted to Lyndon Harris’s comment for this information also).
That well-established principle applies in Ireland also, quite independently of the English authorities just mentioned and which are obviously not binding here. The severity of any prison sentence is a function not only of its length but also of the prisoner’s subjective experience while serving it. Thus, we readily accept that a given term of imprisonment will be more onerous for a seriously ill, disabled or infirm person than for a fully able-bodied person. (The classic piece of academic writing on this topic is Adam Kolber, “The subjective experience of punishment” (2009) 109 Columbia L.R. 182). That is why serious illness and disability are treated as mitigating factors (See, for example, People (DPP) v O’Brien [2015] IECA 1 and R v Hall [2013] EWCA Crim 83; [2013] 2 Cr. App. R. (S.) 68). The same consideration should apply where detention will be more onerous than is usual or acceptable because of overcrowding and its consequences.
Irish courts have rarely had to confront this question directly. In People (DPP) v Hegarty [2013] IECCA 67 the Court of Criminal Appeal held that if prison conditions are to be a relevant factor at sentencing (and the Court did not deny that they could be), there should be evidence of the conditions in which the offender is likely to be held. One difficulty associated with this requirement is that, at the time of sentence, it cannot always be predicted where an offender will serve the sentence (and it may be in different institutions over time). However, the data now available on the Irish Prison Service website means that prison populations can be identified in real time, so to speak. If, for example, a court were sentencing an offender on 1 June 2023, it could discover from that website that the previous day, every prison establishment in the country was operating either very close to 100% capacity or substantially above it . On 31 May 2023, Mountjoy Men’s Prison, for example, was operating at 106% capacity while the Women’s Prison was operating at 111% capacity. Wheatfield was the lowest at 93% capacity. All others were well above that. (To get this information, log on to https://www.irishprisons.ie; click on “Information Centre”, then on “Statistics & Information”, then on “Daily Prisoner Population”).
In practical terms, reliable information about prison overcrowding may tip the balance in a so-called cusp case in favour of a suspended or non-custodial sentence, as opposed to immediate custody. As it happens, both Manning and Arie Ali, because of the particular facts, could be regarded as cusp cases. Had they been significantly more serious, neither Covid nor prison overcrowding would have saved the offenders in question from imprisonment. However, even where imprisonment must be imposed, likely conditions of detention remain a relevant factor. In Arie Ali, the Court of Appeal said (Para. 22):
“This factor [prison overcrowding] will principally apply to shorter sentences because a significant proportion of such sentences is likely to be served during the time when the prison population is very high. It will apply only to sentences passed during this time.”
In Ireland, unfortunately, prison overcrowding is likely to be with us for quite some time. It is therefore a factor which courts can take into account, at least when imposing short to medium-length prison sentences. Such sentences, where they must be imposed, can legitimately be reduced because the custodial experience will probably be more severe or onerous than normal. The “punitive bite” will be equivalent to that of longer sentences served in better custodial conditions. Additionally, making sentences as short as possible will make some contribution, in the medium to long term, in addressing overcrowding.
In Seed and Stark (above), the English Court of Appeal observed that the “numbers of those in prison are a product of the number of custodial sentences imposed and the length of those sentences.” True enough, but prison numbers are also determined by the length of time for which prisoners are detained. An effective parole system is another essential piece of the jigsaw but, as outlined in a previous post on this blog, that is something we do not currently have.
Sentencing, Crime and Justice 