BPO v R

The provisions of the Sexual Offences (Amendment) Act 1992 apply to these offences. Under those provisions, where a sexual offence has been committed against a person, no matter relating to that person shall during that person’s lifetime be included in any publication if it is likely to lead members of the public to identify that person as the victim of that offence. This prohibition applies unless waived or lifted in accordance with section 3 of the Act.

This is an appeal against sentence with the limited leave of the single judge. In so far as leave was refused, the Appellant makes a renewed application for leave.

In R v. Ahmed [2012] EWCA Crim 281 the Court established the principle that, when sentencing an adult for an offence committed whilst a child, the starting point is the sentence which was likely to have been imposed if sentenced shortly after the commission of the offence. This appeal raises the question whether, in applying that principle, the Court should take account of changes in early release provisions between then and now.

In June 2021 and May 2023, the Appellant was convicted of three counts of rape, contrary to section 1(1) Sexual Offences Act 1956. On 18 June 2021 in the Crown Court at Cambridge he (then aged 59) was convicted on count 1 of trial indictment 1. On 3 May 2023 in the same court he (by then aged 61), was convicted on counts 3 and 5 on trial indictment 2.

On 9 June 2023 the Appellant was sentenced as follows: on count 1 of trial indictment 1, 3 years imprisonment; on count 3 of trial indictment 2, 5 years imprisonment; and on count 5 of trial indictment 2, 3 years imprisonment. The sentences for the first and third of those offences are to be served concurrently to the sentence on count 3 of trial indictment 2. The total sentence was therefore one of 5 years imprisonment.

Following two trials the Appellant was convicted of a number of historic sexual offences against his two younger half-sisters, C1 and C2. Those offences were committed in the late 1970s and early 1980s.

In June 2021 he was convicted on count 1, a single incident offence of rape against C1, committed between 7 March 1977 and 7 March 1980. The jury in the first trial acquitted the Appellant of count 5, but were unable to reach verdicts in respect of the remaining counts.

Following a retrial in May 2023, the Appellant was convicted of counts 3 and 5 on a second indictment. Count 3 was a multiple incident count of rape against C1, committed on at least 5 occasions between 7 March 1977 and 7 March 1980. Count 5 was a single incident of rape against C2, committed between 12 February 1976 and 4 April 1981.

At the time of the offending C1 was aged between 8 and 11; the Appellant was then aged between 15 and 18. The offence against C2 was committed when she was between the ages of 7 and 13; and the Appellant was between the ages of 14 and 19. The Appellant was sentenced on the basis that he was under 18 at the time of all the offending.

The offences were committed at the family home and at a grandparent’s address, and whilst other family members were in nearby rooms. The Appellant attempted to persuade the girls that the offending was “normal” and that a friend of his did the same with his sisters. The Appellant also threatened that his half-sisters would be taken into care if they told their parents.

Following the death of their father in December 2017 C1 and C2 told their mother that the Appellant had sexually abused them as children. They reported the offences to police in the summer of 2018.

In his sentencing remarks the learned judge recounted the facts as we have set out above. The Appellant fell to be sentenced on the basis of six rapes of C1 and one occasion of rape of C2. He praised the two complainants for their considerable courage and bravery in maintaining their complaints and in their participation in the prosecution for many years. It was clear from the victim personal statements that the Appellant had done great lifelong damage to both of them.

The Appellant had known all along what he had done but had not had the decency and courage to admit his guilt. The offences were the rape of small children by a much older adolescent in their own family who the complainants should have been able to trust. The age difference between him and his half-sisters was significant.

The judge said that the Appellant could not be said to bear the culpability that an adult offender would have had, even though he was now a man entering his later years, but his culpability remained significant. The judge took account of the fact that he had not been in trouble throughout his adult life.

The judge referred to relevant guidelines, namely the guideline for the offence of rape of a child under 13; the guideline on overarching principles; the guideline on the imposition of custodial and community penalties; the guideline on sentencing children and young people; and the guideline on totality.

The judge then turned to the correct principles to apply. He applied, as he was required to, the decision in R vAhmed “notwithstanding the very real and very difficult practical challenges that poses”. Applying the principle in that case, the judge then identified that the starting point “must be the sentence that you would have likely received if you had been convicted of these offences very shortly after you committed them”. He proceeded on the assumption the Appellant had committed the offences when he was 16 years of age; and commented that that made two very big differences: first a 16 year old would not be treated as severely as an adult; and secondly the types of sentences that the courts imposed on sex offenders in the 1970s and early 1980s were “a world away from the sentences imposed now…. they were much, much lower.”

The judge added that he was entitled to look at the sentencing guidelines that apply to the modern equivalent of the offences of which the Appellant had been convicted, and to continue to make measured reference to those guidelines. The judge continued:

“I start with the type of sentence you are likely to have received, or indeed, not the type, the actual sentence you are likely to have received at the time of these offences.”

He then identified two types of sentences available to the court, had it dealt with the Appellant in the late 1970s when he was 16. These were, first, Borstal training where the maximum was a two-year sentence; and he noted that that was the equivalent of a present day four-year prison term; and secondly a sentence of detention, for any offence punishable, for an adult, with 14 years or more and where the maximum was life.

As regards Borstal, the judge rejected the submission that the Appellant would not have received anything more than Borstal training back in 1978 or 1979, concluding that he was confident that any sentencing judge at the time would have been appalled at his conduct and that the likely sentence that would have been imposed on him had he been dealt with for the offences before he turned 17 was one of five years’ detention.

The judge then turned to the current sentencing guidelines for the offence of rape of a child under 13. He placed the offences in category 3 harm and at the top end of culpability category B. He identified three specified aggravating factors: the Appellant had ejaculated on each occasion; the offences took place in the victims’ own homes or the homes of close relatives; and other children were often present. On that basis, for an adult, each rape would carry a starting point of 8 years and a range of 6 to 11 years imprisonment. Taking account of the fact that count 3 was a multiple incident count and applying totality, the judge concluded that if the Appellant had been an adult offender for recent offences, he would have imposed a term in the region of 20 years and, as a youth of 16 with an appropriate discount, that would have taken the term down to 12 years.

The judge then observed that that demonstrated the disparity between the approach taken today and the approach that would have been taken in 1978 or 1979. He asked himself, rhetorically, how the court was to square that circle. He could only rely on the guidance set out by the Court of Appeal in Ahmed and emphasised paragraph 32(vi) of the judgment which he then quoted (see paragraph 51 below). In the present case, he found that there were no reasons to depart from the starting point of the sentence which would have been imposed at or near to the time of offending and none had been advanced by the prosecution. He concluded:

“I therefore consider myself applying Ahmed bound to reach a conclusion that the total sentence I impose today should be no higher than the total sentence that adjudge would have been imposed in 1978.”

He then proceeded to impose the sentences which we have already identified, concluding that there was to be a total sentence of five years. The judge then stated that the Appellant would serve half of that time in custody and would then be released on licence.

After a short adjournment, counsel for the Prosecution drew to the judge’s attention the fact that the statutory provisions relating to early release had been amended by section 130 Police, Crime, Sentencing and Courts Act 2022, such that the Appellant would serve two-thirds of the sentence before release.

At that point Ms Bex KC for the Appellant contended that the judge should modify the term of imprisonment imposed so as to lead to the equivalent of half of the five-year term which the judge had just imposed (i.e. 30 months and not 40 months as would be the case under the modified new release provisions). In other words, Ms Bex effectively contended that the judge should reduce the term of five years to one of 45 months.

The judge rejected that submission, stating as follows:

“The release provisions are a matter for Parliament, not for the courts, and I have to sentence, and I regard this still as good law notwithstanding the changes wrought by Ahmed, I have to sentence in accordance with current sentencing law.… The circumstances do not take this case outside of the normal principal approach, namely that changes in release dates brought about by legislation should be put entirely out of a sentencing judge’s mind.”

By his original grounds of appeal, the Appellant contended, first, that the sentence of 5 years is manifestly excessive, as it was longer than would have been imposed in 1978 (“ground 1”). Secondly the Appellant contended, that even if he might have been sentenced to 5 years in 1978, he would have been eligible for release after serving one-third (e.g. 20 months) and so to impose a sentence that requires him to serve twice that, 40 months, is manifestly excessive (“ground 2”).

As regards ground 2, Ms Bex submitted that “when the judge calculated the appropriate sentence to reflect that which would have been passed in 1978, he expected that the Appellant would be released after serving half of his sentence and told the Appellant that he would be released after serving one half”. The judge fell into error in declining to adjust the sentence to reflect the change in release provisions. Whilst she accepted that it is often not inappropriate to decline to adjust a sentence to reflect a recent change in release provisions, in the present case it was wrong not to do so because the point of R v Ahmed is that the defendant is not meant to be more harshly punished than he would have been had he been sentenced in 1978.

The single judge granted leave to appeal limited to the issue of whether the judge should have adjusted the sentence to allow for the release provisions being harsher than he had thought when originally deciding on the sentence. i.e. ground 2. In relation to ground 1, Ms Bex renewed, before us, the application for leave to appeal.

By way of important background, we set out in a little detail sentencing provisions in force in 1978/79 of particular relevance to children and young persons.

First, as regards persons under the age of 17, imprisonment was prohibited: section 19(1) Powers of Criminal Courts Act 1973. Section 60 Criminal Justice Act 1967 (“CJA 1967”) governed release on licence for those adults serving a sentence of imprisonment. Release was at the discretion of the Secretary of State, on the recommendation of the Parole Board, where the person had served at least one-third of the sentence (or 12 months where the sentence was less than 3 years).

Secondly, by section 53(1) Children and Young Persons Act 1933 (“CYPA 1933”) a person under the age of 18 convicted of murder was to be detained at Her Majesty’s pleasure. By section 53(2) CYPA 1933 a young person (aged between 14 and 17) convicted of an offence punishable in the case of an adult with imprisonment for 14 years or more, was to be detained for a term set by the Court. In both cases detention was to be in such place and on such conditions as the Secretary of State might direct. The place of detention might be the person’s home, community homes, Borstals, young prisoner centres or special hospitals. For a 16 year old, in practice, he would be detained either in Borstal or at a detention centre. (For a person of a very young age, the Secretary of State might direct that the term was to be served at home). Wherever the Secretary of State directed the young person to live whilst detained, this constituted “legal custody”: see section 53(3). Once the young person attained 18, he could be directed to serve the remainder of the sentence in prison.

Release on licence of a child or young person sentenced under section 53 CYPA 1933 was governed by section 61 CJA 1967. The power to order such release was exercised by the Secretary of State, on the recommendation of the Parole Board. (In the case of section 53(1) detention there were further conditions). In the case of a section 53(2) sentence, the licence remained in force until the expiration of the sentence of detention. Unlike the position of an adult (under section 60 CJA 1967), it was thus not the case that a child or young person sentenced under section 53(2) was eligible for release on licence after serving one-third of the sentence. There was no time limit on early release; it could occur at any time; and indeed release on licence might not granted at all.

The exercise of this early release was based on policy. That policy could change from time to time. It was not, or not necessarily, based on culpability or harm but on the particular circumstances of the offender, and on the Secretary of State’s policy, at the time.

Thirdly, at the relevant time, Borstal training was governed by section 45 Prison Act 1952. Under that section a person sentenced to Borstal training was detained for a period to be determined by the Prison Commissioners. That period would be for a minimum of 9 months (unless directed to be released earlier by the Secretary of State) and for no more than 2 years. After release from detention, the person would be subject to supervision until the expiration of four years from the date of sentence; the supervision was to be under such person as specified by the Prison Commissioners, who had power at any time to modify or cancel the supervision terms. When sentencing a person to Borstal, the Court had no role in setting the actual period of detention in Borstal. This was decided upon by the Prison Commissioners and/or the Secretary of State.

Borstal training was a hybrid sentence itself because it involved a period of supervision. It was not a sentence which can readily be replicated now. It might be considered as a sentence of four years or a sentence of a maximum of two years custody, followed by supervision. There is no direct equivalent in terms of prison sentences. Moreover it is not possible now for the Prison Commissioners to release, because there are now no such persons.

In fact, on the basis that a sentence of Borstal training lasted overall for 4 years (and assuming detention for 2 of those years), for a case of rape such as in the present case, it is not possible today to replicate precisely such a sentence of Borstal training (as might have been imposed in 1978/1979). It is not mathematically possible to impose a sentence today which comprises 2 years custody followed by 2 years on licence. A sentence for rape today of 4 years imprisonment would qualify for release at the two-thirds point, and so would involve custody of 32 months (rather than 2 years) and a period of 16 months released on licence; and any sentence less than 4 years, even if automatic release took place at the half way point, would necessarily have a custodial element of less than 2 years.

We address the two principal cases relevant to the issue before us. Individually they address distinct issues. R vPatel [2021] EWCA Crim 231 [2021] Cr App R (S) 47 addresses the impact upon sentencing in general terms of a change in early release provisions between the date of conviction and the date of sentence. R vAhmedand others addresses the question of the correct approach to sentencing an adult for an offence committed whilst he or she was a child.

In R vPatel, the Court of Appeal considered the issue of whether sentencing judges should make an allowance for the fact that, because of delays between conviction and sentence, a change in the law regarding release provisions (adverse to the defendants) applied to their cases, resulting in them spending a longer period in custody. The issue arose in the context of the coming into force of a statutory instrument in April 2020 (“the 2020 Order”) which, for certain offenders, changed the point of automatic release from custody from one-half of the term to two-thirds of the term.

The Court of Appeal held that a sentencing judge should not ordinarily take account of early release provisions when deciding the length of a determinate sentence.

At paragraph 17 of the judgment, Dame Victoria Sharp P referred to the explanatory memorandum accompanying the 2020 Order setting out the policy reasons for the change and which states, at paragraph 7.8, that “releasing other serious sexual and violent offenders at the halfway point does not align with this more robust approach following the introduction of the EDS and the SOPC”.

At paragraph 22, the President pointed out that nothing in the legislative framework or in Sentencing Council’s guidelines requires or explicitly permits a sentencing court to take account of the impact of the early release provisions. She continued at paragraphs 23 and 24 as follows:

“It would defeat the statutory purpose of the early release provisions if their effect were ordinarily to be taken into account when passing sentence. The clear intention underpinning the 2020 Order (as is clear from the text of the Order itself and as is spelt out in the Explanatory Memorandum) is that, where it applies, the offender should, before being entitled to release, serve a further one-sixth of the sentence than was previously the case. If the sentencing judge reduced the length of sentence to reflect the harsher effect of the early release provisions then that would directly undermine the legislative purpose.”

Accordingly, the courts have consistently made it clear that a sentencing judge should not ordinarily take account of the early release provisions when deciding the length of a determinate custodial sentence…”

The President then continued at paragraph 24 to review the existing case authority. In particular at paragraph 24(3), she set out the following passage from the judgment of Hughes LJ VP in R v Round [2009] EWCA Crim 2667 at §§44 to 45 and 49:

“the general principle that early release, licence and their various ramifications should be left out of account upon sentencing is, as it seems to us, a matter of principle of some importance.

The wide possible range of regimes for early release and licence strongly reinforces the undesirability, never mind the impracticability, of courts being required to reflect the differences in their sentences.

…

Our clear conclusion is that it is not wrong in principle for a judge to refuse to consider early release possibilities when calculating his sentence or framing the manner or order in which they are expressed to be imposed. We are quite satisfied that it is neither necessary, nor right, nor indeed practicable, for a sentencing court to undertake such examinations. Ordinarily, indeed, it will be wrong to do so, although there may be particular cases in which an unusual course is justified” (emphasis added).

After conducting that detailed review of case authority on the issue, the President then commented:

“25.

This represents an extensive, consistent and binding body of authority, rooted in principle, that has been considered and endorsed by the Supreme Court. It is based on the different roles played by the judiciary and the executive. It recognises the different considerations that influence, on the one hand, individualised sentencing decisions, and, on the other hand, generally applicable statutory early release provisions that reflect broad government policy. The approach of leaving release provisions out of account when setting the sentence has been applied even where that might be said to cause a harsh effect in an individual case. It has been applied where the results are anomalous, and where (as in Dunn (Tony)) that is directly contrary to the intention of the sentencing judge and contrary to an expectation raised in the offender by the sentencing judge, and where (as in Francis) a delay to the sentencing hearing beyond the control of the Appellant has resulted in a change to the applicable provisions.” (emphasis added)

At paragraph 26, the President concluded that there is ordinarily no scope for sentencing judges to take account of the early release provisions.

The issue in the 15 cases before the Court was whether exceptions to that general principle should be made on the particular facts of the cases on appeal, and in particular in the context of delays arising from the Covid pandemic.

After reciting the parties’ submissions, the President set out the Court’s conclusion of principle:

“37.

Nothing in the authorities explicitly rules out the possibility that there may be exceptional cases where it is appropriate to take account of the impact of early release provisions.”

However, the Court concluded that the effect of the Covid pandemic on the date of sentence with a resulting impact of the coming into force of the change in early release provisions was not such an exceptional case. The President continued:

“42.

A change in the early release regime is different. It is a legislative change that is introduced by Parliament (or by a Secretary of State with Parliament’s authority). The authorities we identify at [24] above show that the early release provisions may not ordinarily be taken into account by a sentencing judge - they do not amount to a mitigating factor for the purposes of sentencing. That is for the principled reason identified at [23] above. Mr Evans’ invitation to re-write art.5 of the 2020 Order so that it does not apply if the sentence hearing was (or perhaps should have been) first listed for hearing before 1 April 2020 (or to pass adjusted sentences that would have that effect) is directly contrary to that principle. The broader approach adopted by the advocates for all the Appellants is no different in effect to the exercise in legislating that Mr Evans asks us to undertake. That is because if the courts were to adjust the sentences imposed on offenders whose hearings were adjourned from before 1 April 2020 until after that date, then the courts would thereby change the intended impact of art.5 of the 2020 Order. The Secretary of State could have legislated to achieve the result that the Appellants seek. He chose not to do so (although he could still do so with retroactive effect). It is not open to the courts to thwart that legislative choice.

43.

If there is any exception to the principle that Hughes LJ identified in Round then the exception must, itself, be rooted in principle and consistent with the legislative framework that governs sentencing. The mere fact that the sentencing process has been delayed is not sufficient, as the authorities show. Nor is it sufficient that the process has been delayed for reasons that are beyond the control of the individual Appellant, as Francis shows. Nor is it sufficient that the reason for the delay was unforeseen or unforeseeable. …” (emphasis added).

R v Ahmed and others addresses the distinct question of the correct approach to sentencing an adult for an offence committed whilst he or she was a child. The Court of Appeal held, in summary, that in such a case, the starting point is the sentence which the Court considers were likely to have been imposed if the child offender had been sentenced shortly after the offence. In his judgment, Lord Burnett of Maldon CJ said:

“21.

… . Those who are under the age of 18 when they offend have long been treated by Parliament, and by the courts, differently from those who are adults. That is because of a recognition that, in general, children are less culpable, and less morally responsible, for their acts than adults. They require a different approach to sentencing and are not to be treated as if they were just cut-down versions of adult offenders. The statutory provisions in force from time to time have frequently restricted the availability of custodial sentences for child offenders, whether by prohibiting them altogether for those below a certain age or, more commonly, by restricting on a basis of age the type and maximum length of custody in all but grave cases. All such provisions are in themselves a recognition by Parliament of the differing levels of culpability as between a child and an adult offender: that is one of the reasons why we are respectfully unable to agree with the distinction drawn in Forbes between cases where no custody would have been available, and cases where some form of custody (however far removed from modern sentencing powers) would have been available. There is, in our view, no reason why the distinction in levels of culpability should be lost merely because there has been an elapse of time which means that the offender is an adult when sentenced for offences committed as a child.

22.

Section 59(1) of the Sentencing Code requires every court, when sentencing or dealing with an offender who was under the age of 18 at the time of the offending, to follow the Children guideline except in the rare case when the court considers it would be contrary to the interests of justice to do so. Paragraphs 6.1 to 6.3 of that guideline are relevant in such circumstances, and we are unable to see any justification in logic or principle for the submission that those paragraphs should only be followed where the offender has only recently attained adulthood. They remain relevant, and therefore to be followed, however many years have elapsed between the offending and the sentencing. That is because the passage of time does not alter the fact of the offender’s young age at the time of the offending. It does not increase the culpability which he bore at that time. We naturally hesitate to differ from the decisions in H and Forbes; but insofar as those cases adopted a different approach, it is our respectful view that the court did not have regard to the passages in the SGC Youth guideline which were to substantially the same effect as paragraphs 6.1 to 6.3 in the current Children guideline. In our view, the application of the Children guideline requires sentencers to adopt a different approach between sentencing for historical offending committed as a child and sentencing for historical offending committed as an adult. That difference, and the resultant difference (which may be substantial) in the respective sentences, is in accordance with principle and reflects the special approach to the sentencing of child offenders.” (emphasis added).

At paragraphs 23 to 30, the Court addressed a number of practical difficulties which had been posited as arising from such an approach, including the difficulty of ascertaining what would have happened in a sentencing process many years earlier; none of which the Court regarded as insuperable. In particular, Lord Burnett stated at paragraph 24:

“24.

First, we recognise that the approach set out in Limon and Priestley requires a sentencer to undertake what may be a difficult exercise in considering what would have been likely to happen in a sentencing process many years ago. In a case in which some form of custody would have been available for the child offender, advocates will accordingly need to research not only the statutory maximum sentence for the offence but also the types and lengths of custodial sentences which would have been available in the offender’s case. However, judges are experienced in grappling with the various difficulties which can arise in the context of sentencing for historical offences. In any event, if principled application of the law requires difficulties to be confronted, then they must be.” (emphasis added).

After dealing with these general points, at paragraph 31, the Court addressed an issue which arose on the specific facts of two of the cases before it, where at the time of the offence, the former sentence of Borstal training would or might have been imposed:

“31.

We would add that the approach to a sentence of Borstal training available at the time of offending became common ground before us. In determining what length of custodial sentence should now be imposed to reflect the sentence which was likely at the time of the offending, a sentence of Borstal training (which would have comprised detention for up to two years, followed by supervision for a further two years) can properly be reflected by a sentence of up to four years’ imprisonment. That would reflect current early release provisions.” (emphasis added).

Then at paragraph 32, the Court set out its conclusions of principle:

“32.

We therefore answer as follows the question posed at the start of this judgment:

i)

Whatever may be the offender’s age at the time of conviction and sentence, the Children guideline is relevant and must be followed unless the court is satisfied that it would be contrary to the interests of justice to do so.

ii)

The court must have regard to (though is not necessarily restricted by: see (v) below) the maximum sentence which was available in the case of the offender at or shortly after the time of his offending. Depending on the nature of the offending and the age of the offender, that maximum may be (a) the same as would have applied to an adult offender; (b) limited by statutory provisions setting a different maximum for an offender who had not attained a particular age; or (c) limited by statutory provisions restricting the availability of different types or lengths of custodial sentence according to the age of the offender.

iii)

The court must take as its starting point the sentence which it considers was likely to have been imposed if the child offender had been sentenced shortly after the offence.

iv)

If in all the circumstances of the case the child offender could not in law have been sentenced (at the time of his offending) to any form of custody, then no custodial sentence may be imposed.

v)

Where some form of custody was available, the court is not necessarily bound by the maximum applicable to the child offender. The court should, however, only exceed that maximum where there is good reason to do so. In this regard, the mere fact that the offender has now attained adulthood is not in itself a good reason. We would add that we find it very difficult to think of circumstances in which a good reason could properly be found, and we respectfully doubt the decision in Forbes in this respect. However, the point was not specifically argued before us, and a decision about it must therefore await a case in which it is directly raised.

vi)

The starting point taken in accordance with (iii) above will not necessarily be the end point. Subsequent events may enable the court to be sure that the culpability of the child offender was higher, or lower, than would likely have been apparent at the time of the offending. They may show that an offence was not, as it might have seemed at the time, an isolated lapse by a child, but rather a part of a continuing course of conduct. The passage of time may enable the court to be sure that the harm caused by the offending was greater than would likely have been apparent at that time. Because the court is sentencing an adult, it must have regard to the purposes of sentencing set out in section 57 of the Sentencing Code. In each case, the issue for the court to resolve will be whether there is good reason to impose on the adult a sentence more severe than he would have been likely to have received if he had been sentenced soon after the offence as a child.”

(emphasis added)

The Court then turned to consider the individual appeals. The first appellant was Nazir Ahmed (judgment, paragraphs 35 to 47). He had been convicted of one offence of buggery of C1 aged 9 or 10 (when the appellant was 14) and two offences of attempted rape of C2 aged 4 or 5 (when he was 16). The judge imposed a sentence of 3 years 6 months for the offence of buggery and, consecutively, a sentence of 2 years concurrent for each offence of attempted rape, making a total of 5 years 6 months. In relation to the buggery, the judge was sure that a court at the time would have imposed a custodial sentence of more than 6 months (paragraph 38). At paragraph 39, the Court addressed the two offences of attempted rape:

“As to counts 3 and 4, the judge bore in mind that they were offences of attempt. He assessed them as falling within category 2B of the guideline for rape of a child under 13, because C2 was particularly vulnerable due to her extreme youth. At the time of the offending, the maximum custodial sentence which the applicant could have received was two years’ borstal training. The judge considered whether this was one of those rare case in which a longer sentence than that maximum should be imposed but concluded that it was not.”

At paragraph 43, the Court identified the issue at the heart of Mr Ahmed’s appeal was whether the judge was correct to conclude that the sentences likely to have been imposed at the times of the offending were a custodial term exceeding six months on count 1, and Borstal training on counts 3 and 4. At paragraph 44 the Court concluded that the judge had fallen into error in relation to the offence of buggery. If the applicant had been sentenced shortly after the count 1 offence, the court would have been dealing with a child of 14, with no previous convictions, for a single offence of buggery. They disagreed that such a court would have decided that neither custody for up to six months, nor any other lawful way of dealing with the applicant, was suitable. Whilst the case might have been regarded as coming close to the borderline, a custodial sentence of six months would probably have been regarded as a suitable penalty.

On the other hand, the Court concluded that the judge had been correct in relation to the two offences of attempted rape, stating at paragraph 45:

“45.

Counts 3 and 4 were two offences committed by a teenager of 14 against a very young victim. Although they were attempts, it appears that the applicant came close to penetrating C2’s vagina. In those circumstances, we are not persuaded that there can be any criticism of the judge’s decision that the likely punishment at the time would have been Borstal training. On the other hand, we do not think it necessary to increase the sentences imposed on the basis that a sentence of Borstal training could be treated as equivalent to four years’ imprisonment.”

Thus, the sentence in relation to the offence of buggery was reduced to 6 months; the sentences for the attempted rape offences were undisturbed. On that basis the total sentence was reduced to 2 years and 6 months.

The fifth appellant was Peter Hodgkinson. He had been sentenced to a total special custodial sentence for an offender of particular concern made up of a custodial period of 6 years and a further licence period of 2 years for six offences of indecent assault contrary to section 14(1) of the 1956 Act. Those offences had been committed between 1972 and 1977 when he was between 16 and 21 and where the complainants were between 9 and 14 and 7 and 12 respectively. The Court allowed the appeal, stating at paragraph 125 as follows:

“125.

The maximum sentence for indecent assault during the indictment period was five years. … the offence was not one for which long term detention [i.e. section 53(2)] would have been available had the appellant been sentenced at the time of the commission of the offences. Rather, the available custodial sentence would have been Borstal training. We have already considered the effect of such a sentence in our determination of the appeal of Ahmed. The effect of a sentence of Borstal training is the same in the appellant’s case as in the case of Ahmed. Counsel for this appellant concedes that the combination of the maximum initial period of custody of two years coupled with a period of supervision expiring four years from the date of the imposition of the sentence corresponds to a determinate term of four years’ custody, with release on licence at the half way point of the term. Applying the principles we have set out earlier in this judgment, the appellant should not have been sentenced to a custodial term in excess of four years. There is nothing exceptional about his case which would justify a departure from those principles.”

In relation to ground 2, Ms Bex in written submissions and in oral argument, submitted as follows.

In not taking account of the release provisions as they applied in 1978/1979, the judge erred because it is impossible to apply the letter or spirit of Ahmed without doing so. The effect of paragraph 31 of Ahmed is that in calculating the proper sentence, the court should make allowance for the time actually served rather than simply pass the same “headline” sentence. Paragraph 31 requires the judge to pass a sentence that is as close as possible in both form and effect as that which would have been passed at the time of the offending.

The need to ensure that an offender is not being more harshly punished than he would have been, had he been sentenced near the time of the offending, necessarily requires regard to the early release provisions. In the present case an offender who is required by a sentence to spend at least 40 months in prison before release is being treated more harshly than one who is required to spend at least 20 months in custody before being eligible for release. The fact that the regime for release back in 1979 was discretionary, rather than automatic, does not mean that release is so speculative as to justify it being ignored. In oral submissions, Ms Bex accepted that 20 months was not the correct comparison, since the one-third release provisions then applicable did not apply to a section 53 sentence for a young person (see paragraphs 30 and 32 above). However, even if release would have been “at large”, given his background, the Appellant would have been likely to have been near the front of the queue for early release.

This is not a case of seeking to mitigate the effect of early release provisions by imposing a lower sentence. What is required here is to punish the offender in the same way that the court would have done had he been sentenced at the earlier date. It is clear from paragraph 31 of Ahmed that in order to achieve that it is necessary to have regard to both past and current release provisions. In applying Ahmed, the court is not limited to seeking to identify how culpability and harm would have been reflected by the court at the earlier date. The whole point of Ahmed is to ensure that an offender is “not more harshly punished”. The current government’s more punitive approach to release provisions reflects their view of an offender’s culpability and the harm caused. To ignore those provisions is to fail to reflect how culpability and harm would have been reflected at the material time.

The Court arrived at the figure of 4 years in paragraph 31 of Ahmed because it had regard to the current release provisions. This demonstrates that the Court was working on the basis that in order to impose a sentence that was no harsher, one necessarily had to have regard to the current release provisions.

The Appellant’s case is not contrary to Patel. Paragraph 23 of Patel indicates that the legislative purpose of the changes to the release provisions reflected the “more robust approach” to sentencing for qualifying offences. However the point made in Ahmed is that it is not appropriate to apply a more robust approach to offenders sentenced as mature adults for offences committed when they were children themselves. Therefore, a departure from Patel is both required and entirely appropriate in order to achieve the objective of Ahmed. Ahmed makes it clear that the sentencing of an adult for an offence committed whilst a child is an exceptional class where it is necessary to have regard to culpability and harm at the time. The only sensible interpretation of paragraph 31 of Ahmed is that the Court did have regard to early release provisions.

Finally, the exceptionality to be applied is not specific to the Appellant’s case but to the class of cases that fall within Ahmed i.e. cases of sentencing adults for offences committed as a child. Those cases are rare.

Mr Mably KC for the Respondent submitted, in the Respondent’s Notice and in oral argument, that the judge’s approach to the sentencing exercise was correct. The judge was not required to calculate the likely sentence based on 1978/79 release provisions (i.e the way in which a sentence of detention would have been served in practice at that time) or to recalculate his sentence to take account of the fact that the Appellant would have to serve two-thirds, rather than one-half, of his sentence in custody.

First, there was nothing to indicate that the judge took into account either the release provisions as applied in 1978/79 nor those that apply now. His reference to automatic release after half the sentence was the standard judicial comment made in all sentencing remarks. When it was drawn to his attention that the Appellant would have to serve two-thirds of the sentence, there was nothing to correct in terms of sentence length. The judge was correct not to take into account the recent changes in release provisions, consistent with the decision in Patel.

Secondly, the relevant statutory scheme in 1978/79 did not provide for automatic early release, and was dependent on the Parole Board. The precise time that the Appellant would have been released on licence cannot therefore be determined, and so the basis of the Appellant’s case is inherently speculative.

Thirdly, there is nothing in the authorities, and in particular in Ahmed, to indicate that in cases such as the present a judge ought to have regard to release provisions when fixing the length of a sentence. It would be wrong in principle to seek to mitigate or sidestep the effect of early release provisions by imposing a sentence different from that which is otherwise appropriate. In looking back to the historic sentencing regime and practice, the sentencing court today is seeking to identify how culpability and harm would have been reflected by the court at the material time. The way that such a sentence historically would have been administered thereafter is irrelevant to that issue.

Fourthly, the reference to early release provisions in Ahmed at paragraph 31 was made in the context of an historic type of sentence which had a particular structure (involving a period of custody and a period of supervision) and which has no direct equivalent in the case of an adult offender. The Court’s consideration of current early release provisions in that specific case considered in paragraph 31 was not a statement of general principle that a comparison of such provisions was necessary when reflecting historic youth detention with a current term of imprisonment.

Fifthly, as to whether the Appellant’s case could fall into an exceptional category where release provisions should be taken into account (as suggested at paragraph 37 and 43 of Patel), any exception to the ordinary rule must be based either on the exceptional nature of the particular offence or offender or upon the exceptionality of a class of case. Neither applies in the present case.

In conclusion as a matter of principle there is no reason why the appropriate “headline” sentence reflecting culpability and harm, identified by a sentencing judge should be altered to take account of the way that sentence would, or may, have been administered by the executive.

First, the starting point of the “sentence likely to have been imposed … if sentenced shortly after the offence” referred to in paragraph 32(iii) of Ahmed is the Court’s sentence i.e. the sentence which would have been imposed by the Court at that time and regardless of the regime governing release. The judge should not impose a sentence more severe than that likely to have been imposed at the time. That means not more severe, in terms of the assessment of culpability and harm. That is given effect to by the Court’s - “headline” - sentence. In that exercise there is no reason to consider the subsequent administration of that sentence. There is no indication in Ahmed that that exercise involves analysis of the early release provisions or the particular form of the sentence. The underlying rationale for the decision in Ahmed (set out at paragraphs 21 and 22) was to recognise the sentencing principle that children are less culpable and less morally responsible than adults and therefore a different approach to sentencing is required. Ahmed expressly gave effect the Sentencing Council’s Guideline on Sentencing Children and Young People dealing with cases where offenders pass through a significant age threshold. Under that guideline the starting point is the sentence likely to have been imposed at the time of offending. That sentence is the headline sentence and not the headline sentence attenuated by the particular way that the executive has from time to time chosen to administer such sentences.

Secondly, the established principle confirmed by Patel is that early release provisions may not ordinarily be taken into account by a sentencing judge. As a matter of sentencing principle, a judge should not have regard to release provisions. That reflects the constitutional distinction between the judge (imposing sentence in the individual case to reflect culpability and harm) and, on the other hand, the administration of sentences by the executive: see Patel at paragraph 25. The judge should not cut across that distinction by mitigating executive policy.

Thirdly, at paragraph 32 (i) to (vi) of Ahmed, the Court sets out in detail the principles to be applied when sentencing an adult for an offence committed when he was a child. There is no reference in those principles to early release provisions in general nor to Patel. (Indeed there is no reference at all to Patel in Ahmed). Paragraph 24 makes no reference to the release provisions at the time. Rather the court should consider what the sentence would have been by reference to two factors: the type of sentence and the likely length of sentence at that time.

Fourthly, the Court’s sentence - then and now - is based on an assessment of the harm and culpability of the particular offence and the particular offender. Even if the purpose of an increase, or change, in early release provisions, is to impose a more harsh or “robust” punishment (see para 7.8 of Explanatory Memorandum at Patel paragraph 17), that is in respect of the Government’s view of the harm and culpability arising from the commission, in general, of the offences in question. We do not accept Ms Bex’s contention that to ignore the release provisions is to fail to reflect how culpability and harm arising from the defendant’s particular offence would have been reflected at the time of offending.

Fifthly, it is significant that back in 1978/9 in general there was no “automatic” early release; rather release was at the discretion of others – namely the Secretary of State or the Prison Commissioners. In the case of children and young persons, under section 53(2) CYPA, release on licence was wholly at the discretion of the Secretary of State (on the Parole Board’s recommendation) and there was no minimum or maximum period of detention. In the case of Borstal training, release was never automatic, but rather at the discretion of the Prison Commissioners (or the Secretary of State). If required to take this into account when sentencing now, the judge would be required to speculate, on an entirely hypothetical basis, as to whether and when - many years earlier - the defendant in question might have been recommended for release by the Secretary of State or the Prison Commissioners. At paragraphs 24 to 30 in Ahmed the Court addressed - and met - the possible practical difficulties that would arise where sentencing by reference to the time of the offending. However it was not envisaged that those practical difficulties would include problems arising from the approach, from time to time, to release on licence. It is not necessary, nor indeed possible, to have regard to speculative discretionary factors such as the particular form of custodial setting or when the Appellant may have been released. At that time good behaviour and, in the case of Borstal, performance in educational training, would have been relevant. Neither could be assessed now – hypothetically and retrospectively.

Sixthly, the only reference in the statement of the general principles in Ahmed to early release provisions is at paragraph 31. In our judgment, but for the Court’s observations there, which deal with the specific case of Borstal training, it is clear that in sentencing now by reference to the sentence at the time of the offence, no account should be taken of any change in early release provisions between then and now. We consider that, on analysis, paragraph 31 does not undermine that conclusion.

The reference to early release provisions in paragraph 31 arises only in the context of the particular historic structure of the former sentence of Borstal training. That arose on the specific facts of two of the cases before the Court, where a possible sentence available at the time of the offending was Borstal training.

In our judgment, at paragraph 31, the Court in Ahmed was not setting out a statement of general principle (effectively, by way of an exception to the general approach in Patel) that it is necessary to compare early release provisions. Rather it was dealing with the historic, and different, structure of a sentence of Borstal training and giving a summary of the broad effect of Borstal training. The maximum length of that sentence was 4 years, but only half of that period might involve a custodial element. No question of early release was involved. The question was how a sentence structured in that way could be reflected in the case of a sentence of imprisonment. The Court stated that a sentence of up to four years’ imprisonment would “reflect” a sentence of Borstal training. In fact, we consider that it is not possible in a sentence imposed today to replicate the contents of a sentence of Borstal training. In the latter case, the judge did not set any term of custody; at the time of sentence, there was no indication of the custodial period and the period on licence; that was wholly a matter for the discretion of the Secretary of State or the Prison Commissioners (applying relevant policy from time to time). Moreover, as explained in paragraph 36 above, it is not possible to impose a sentence today for the present offences which would replicate the four year term of a Borstal training sentence (up to 2 years in custody followed by 2 years on licence). In our judgment, paragraph 31 of Ahmed is not authority for the proposition that it is necessary when considering the sentence that would have been imposed at the time of the offence, to take account of the comparative position as to release on licence.

For these reasons, we conclude that, in considering the sentence which was likely to be imposed if sentenced shortly after the offence under the principle in Ahmed, a judge should not take account of the provisions for release on licence applicable either now or at that earlier time. In the present case, the judge correctly applied the principle in Ahmed and further was right to refuse to make any adjustment to take account of release provisions, in line with the principle in Patel Accordingly Ground 2 fails.

In renewing the application for leave, Ms Bex submitted that, regardless of the position in relation to early release, the sentence of 5 years was too long. In 1978 the Appellant was aged 15 or 16. Prior to conviction he was of good character. This was a first-time offence. Ms Bex relied upon a comparison of the present case with the specific facts relating to Nazir Ahmed in the Ahmed case (and in particular at paragraphs 39 and 45), where a sentence of 5 years 6 months was reduced to 2 years 6 months. She further relied on paragraph 125 relating to Peter Hodgkinson. R v Billam [1986] 1 WLR 349 (relied upon by the Respondent) was a case in 1986 and does not inform the view of what would have been the position earlier, in 1978. The judge should have concluded that if sentenced then, the Appellant would have been sentenced to Borstal training and not to section 53 detention and thus that the sentence now should be no more than four years’ imprisonment.

Mr Mably submitted it cannot be said that a sentence of 5 years in this case was manifestly excessive. The judge was entitled to conclude that this case was too serious for Borstal and that the Appellant would have been sentenced to section 53(2) detention. In R v Billam the Court of Appeal indicated, in 1986, that, in most cases, a young person convicted of rape would be sentenced to detention under section 53(2) CYPA 1933. Moreover, whilst the particular facts of other cases are of limited relevance, in any event the present case was more serious than the case of Nazir Ahmed in Ahmed.

In applying the Ahmed approach, the judge correctly identified the two types of sentence available in 1978 – namely Borstal training and section 53(2) detention. On the basis that the Appellant had been convicted of raping his younger sisters and on no fewer than 7 occasions, he was not persuaded that a court in 1978 would have imposed only Borstal training and was confident that such a court would have been “rightly appalled”. First, R v Billam (at 352D-E) gives an indication that in 1986 (and in 1978), in most cases, the appropriate sentence for rape would have been section 53(2) detention. The Lord Chief Justice expressly said (at 350B) that the Court was provided with an opportunity to “restate principles” to be applied in sentencing rape cases.  He went on to cite some statistics to show that some sentences being passed had been far too low, and said (at 350H-351A) that judges “may need reminding what length of sentence is appropriate”.  Ahmed does not require a court in 2023 to replicate errors made by some judges prior to Billam in passing what the higher courts had always thought to be inappropriately lenient sentences.

Secondly, in so far as factual comparison with other cases is helpful, in any event, we consider that the present case is substantially more serious than the facts of Nazir Ahmed: in the present case, there were at least seven instances of the completed offence of rape perpetrated against members of the Appellant’s own family in the family home. In Nazir Ahmed, there were two offences of attempted rape (even if in one case, it was close to the complete offence). Thirdly, even in the case of Mr Ahmed, it appears (from the last sentence in paragraph 45) that the Court of Appeal considered that the judge could have imposed a sentence now of 4 years to reflect the historic 2 year maximum for Borstal training (rather than the two years which the judge did impose).

In the present case, the judge did not fail to have regard to any matter and in our judgment it is not arguable that the judge was wrong to conclude that, had he been sentenced in 1978/79, the Appellant would have been sentenced to section 53(2) detention (rather than Borstal training). A sentence of 5 years was not arguably manifestly excessive. For these reasons we refuse leave to appeal on ground 1.

There is one further issue. The Appellant is entitled to receive credit for half the number of days on curfew if the curfew qualified under the provisions of section 325 Sentencing Act 2020. On the information before the Court, the Appellant was subject to a qualifying curfew for 720 days and, accordingly, the credit for the purposes of section 240A Criminal Justice Act 2003 is 360 days. But if this period is mistaken, this Court will order an amendment of the record for the correct period to be recorded.

For these reasons we refuse leave to appeal on ground 1 and the appeal on ground 2 is dismissed.