A. Precedents Cited
The Court expressly relied on the classic statement of principle in
Attorney General's Reference (No 4 of 1989) [1990] 1 WLR 41, [1990] 90 Cr App R 366.
The cited passage (at Cr App R p.371) was used to reinforce two controlling points in unduly lenient sentence (ULS) jurisprudence:
-
High threshold: section 36 is aimed at cases of gross error, not fine differences of opinion, and leave should not be granted in borderline cases.
-
Residual discretion: even if a sentence is unduly lenient, the Court of Appeal retains a discretion whether to intervene.
Here, the Court exercised that discretion to intervene because the under-reflection of overall criminality across three victims and numerous incidents
placed the sentence outside the range reasonably open to the sentencing judge.
B. Legal Reasoning
i) The “unduly lenient” test applied to multi-count, multi-victim sexual offending
The Court reiterated the standard framework: trial judges are best placed to weigh sentencing factors; appellate intervention is confined to sentences
outside the reasonable range; and the jurisdiction is not for “borderline” correction. Nonetheless, the Court found the sentencing remarks did not
transparently explain how five years was derived for each victim given the agreed categorisation and scale of offending.
ii) Guideline structure, concurrency within victims, and the need for an “uplift”
The parties had agreed that all but one count were Category 1A under the relevant offence-specific guideline, with a stated starting point
of 5 years (range 4–10 years). The Court accepted that sentences within a victim’s set of offences could properly be made
concurrent around a lead offence, but emphasised that concurrency does not remove the need to reflect the additional criminality from multiple incidents.
The Court’s recalibration proceeded in three steps:
-
Single-offence baseline (pre-discount): for one offence against each victim, with multiple Category A aggravating features and limited mitigation,
the Court indicated 7 years would be merited before plea credit.
-
Within-victim uplift (pre-discount): because each victim suffered multiple offences (A: 12 incidents; B: 6; C: 4),
a proportionate uplift to 7 years 6 months (pre-discount) was required to ensure the concurrent package still represented the totality of offending
against that victim.
-
Plea credit and outcome per victim: applying the preserved 25% credit yielded 6 years 6 months as the appropriate
“block” sentence per victim.
iii) Totality across victims: consecutive blocks with moderation
The Court accepted (and maintained) the structure of consecutive sentences between victims, reflecting separate harm and criminality.
However, it also applied the totality principle a second time at the aggregation stage: adding three full “blocks” can produce an overlong overall term.
The Court stated that a reduction of one year in relation to the consecutive sentences for each victim was “just and proportionate”.
The practical result was that, rather than three blocks at 6 years 6 months plus some further uplift, the Court fixed
the final structure at 6 years 6 months per victim consecutively, totalling 19 years 6 months.
iv) Dangerousness and the extended sentence
The respondent argued that, if custody increased, the Court should reconsider whether an extended licence period remained necessary.
The Court rejected that: the respondent was properly assessed as posing a high risk of serious harm to female children and therefore met the
dangerousness rationale for an extended sentence under section 279 of the Sentencing Act 2020.
Accordingly, the 5-year extension period remained required for public protection.
v) A practical point about sentencing remarks
While acknowledging case-management pressures, the Court highlighted the difficulty created by the absence of a clear explanation for the chosen
custodial term per victim. The judgment underscores that in complex, multi-count sexual cases, sentencing remarks should reveal:
the lead-offence approach, the concurrency rationale, and how uplifts and totality moderation were quantified.