Poulson & Ors: The “campaign of rape” benchmark, weighting of guideline factors, and limited scope for delay as mitigation in multi‑count sexual offending ([2025] EWCA Crim 1241)

Calibrating Multi‑Count Sexual Sentencing: the “campaign of rape” 20‑year benchmark, flexible guideline weighting, and limited delay mitigation — Commentary on R v Poulson & Ors [2025] EWCA Crim 1241

Introduction

This appeal arises from a large, multi‑defendant “gang grooming” prosecution centred on serial sexual offending against teenage girls, principally V1, in and around a Liverpool address described by the trial judge as a “den of iniquity.” The Crown Court trial (HHJ Medland KC) ran from October 2024 to February 2025; a two‑day sentencing hearing followed in April 2025.

Before the Court of Appeal (Criminal Division), Edis LJ (giving the judgment of the court) heard:

  • Nine applications by His Majesty’s Solicitor General under section 36 Criminal Justice Act 1988 (the unduly lenient sentence, “ULS”, regime);
  • Two defence applications for leave to appeal sentence (Poulson and Harwood);
  • Noted that Corley’s application for leave had been abandoned; and
  • Refused jurisdiction in an additional ULS application concerning Harvie Aspden because he had not been convicted of an offence to which the ULS regime applies.

All victims retain lifelong anonymity under the protective regime referred to in the judgment (termed the “Sexual Offences Act 1992” in the judgment; the protective regime is commonly understood to derive from the Sexual Offences (Amendment) Act 1992). The offences predominantly targeted children over 13 (primarily 14–15), and included multiple rapes, assaults by penetration, sexual activity with a child, and related sexual offences. The offending against V1 occurred in the first three months of 2018; the investigation was triggered when a co‑offender crashed a vehicle with V1 as passenger, prompting disclosures.

The central issues on appeal were:

  • How to sentence multi‑count sexual offending where offence‑specific guidelines provide limited direction on aggregation and totality;
  • The role of the rape guideline’s “campaign of rape” rubric as a benchmark for 20 years and above;
  • The proper treatment of delay (including pandemic‑era disruption) as mitigation;
  • The scope of appellate intervention in ULS references, and the deference due to trial judges who have lived with a complex case over many weeks;
  • The weighting of guideline culpability factors across adjacent categories (illustrated by a 1A/1B calibration in Turner); and
  • Mitigating factors including youth, cognitive limitations, serious ill‑health, guilty plea credit, and avoidance of double counting.

Summary of the Judgment

The Court:

  • Increased only one sentence: the sentence of Ashley Darbyshire, identified as the most culpable offender, was increased on three rape counts to concurrent terms of 18 years and 6 months (reflecting a revised pre‑mitigation term of 22 years, less 1 year for serious health consequences, then a little over 10% for very late guilty pleas).
  • Refused to interfere in all other cases, drawing a clear line between sentences that may be “lenient” and those that are “unduly lenient”. It found that most challenged sentences fell within the range of reasonable outcomes open to the trial judge.
  • Granted leave but dismissed the References in arguable cases (Poulson, Haslam, Turner, Fitzgerald).
  • Refused leave in Barrett, Harwood, Corley and Bainbridge‑Flatters.
  • Refused the offender appeals: Poulson’s application for leave to appeal sentence was refused as unarguable; Harwood’s leave application was refused for the same reasons that the ULS route failed.
  • Confirmed limited scope to treat delay as mitigating on these facts (scale of investigation; pandemic; defendants’ denials), with Corley as a notable exception due to early pleas and personal mitigation.
  • Clarified use of the rape guideline: “campaign of rape” is not a technical term; if engaged, 20 years is the bottom of the appropriate range and there is no upper limit; it may serve as a helpful benchmark to assess multi‑count cases.
  • Endorsed flexible, non–tick‑box application of guidelines: judges must assess the weight of culpability factors and may appropriately calibrate between adjacent categories (as in Turner).
  • Emphasised appellate restraint in ULS cases and urged greater selectivity in deciding which sentences to refer.

Detailed Analysis

1) Normative sources and “precedents” relied on

The Court did not cite case authorities by name. Its reasoning turned on:

  • Section 36 Criminal Justice Act 1988 (ULS regime) including its jurisdictional limits and the high threshold for intervention;
  • Victim anonymity under the protective regime referred to as the Sexual Offences Act 1992;
  • Sentencing Council guidelines:
    • Rape (with the “campaign of rape” rubric indicating that 20 years and above may be appropriate),
    • Assault by penetration,
    • Sexual activity with a child,
    • Sexual communication with a child,
    • Reduction for guilty pleas,
    • Sentencing children and young people, and
    • Totality.
  • Substantive offences referenced included the Sexual Offences Act 2003 (rape; assault by penetration; sexual activity with a child; causing/inciting sexual activity without consent; sexual communication), the Protection of Children Act 1978 (indecent images), the Criminal Attempts Act 1981 (attempted rape), the Misuse of Drugs Act 1971 (supply), and the Theft Act 1968 s.12A (aggravated vehicle taking).

The analytical substrate is therefore guideline‑ and statute‑led. The Court’s contribution is to clarify how those guidelines are to be used in the complex context of multi‑count sexual offending.

2) The Court’s legal reasoning and principles applied

a) Scope of ULS intervention and deference to the trial judge

ULS references are not an invitation to re‑sentence afresh or to substitute the appellate court’s view for that of an experienced trial judge who has managed a long, complex trial and a detailed sentencing exercise. The touchstone is whether the sentence falls outside the range of reasonable sentences — that is, whether it is unduly lenient, not merely “on the lenient side.”

Edis LJ stressed the unique vantage of the trial judge, who saw and heard victims and defendants over many weeks and carefully calibrated relative culpability across ten offenders. The Court signalled that it will accord real respect to such evaluative judgments. It also counselled the Law Officers to exercise greater discrimination when deciding which sentences to refer in sprawling multi‑defendant cases.

b) Handling delay: when does it mitigate?

The Court carefully separated three strands of delay in this case:

  • Complex investigations take time — particularly multi‑suspect, multi‑complainant grooming cases.
  • Pandemic disruption was a major factor.
  • Defendants’ choices (denials, late pleas) contributed to elapsed time; that reduces or removes any “delay credit”.

Accordingly, the mere passage of time between complaint and sentence was not mitigating for most appellants. The exception was Corley, who pleaded guilty relatively early, in a context where the complainant’s ultimate trial robustness was unknown, and who had positive personal mitigation. The Court endorsed the trial judge’s structured approach in Corley: start at the correct category, reduce for delay and mitigation, then apply standard plea credit.

c) Totality in multi‑count sexual offending and the “campaign of rape” benchmark

The Court underlined a practical gap: offence‑specific guidelines offer little direction on aggregation for multiple counts. The Totality guideline offers general guidance, but there is one notable exception: the rape guideline states that “offences may be of such severity, for example involving a campaign of rape, that sentences of 20 years and above may be appropriate.” The Court emphasised:

  • “Campaign of rape” is not a technical or defined term; it relies on judicial experience and common sense.
  • If engaged, 20 years is the bottom of the appropriate range; there is no upper limit.
  • This rubric can serve as a benchmark to cross‑check whether the totality properly reflects persistent, multi‑episode rape offending.

Practically, the sentencing technique approved here was to identify lead counts (often rape), place them correctly within the guideline, and then uplift those lead sentences to reflect the totality of additional counts, running the other sentences concurrently to avoid double counting.

d) Flexible weighting within guideline categories: not a tick‑box exercise (Turner)

In Turner, 1A culpability factors were present (offending “with others”; alcohol/drugs facilitated) but with limited weight on the facts. The sentencing judge pitched the sentence at the top of 1B, below the 1A range, explicitly recognising that the case did not fit “exclusively” within 1A. The Court approved this approach, underscoring that guideline application requires evaluation of the weight of factors, not mechanistic box‑ticking. This is a valuable clarification for cases straddling adjacent categories, permitting careful calibration between them when justified by the facts.

e) Youth and maturity discounts (Barrett)

Barrett was 17 at the time of multiple rapes. The judge adopted a mature‑adult comparator of 20 years (having regarded the offending as akin to a “campaign of rape”) and then applied a 40% youth discount in line with the guideline on sentencing young people, arriving at 12 years. The Court endorsed this methodology and outcome. It also recognised the relevance of his dysfunctional home environment (the offending locus was his father’s house), though it did not excuse the offending.

f) Serious ill‑health, late pleas, and avoiding double counting (Darbyshire)

Darbyshire was the prime mover: 19 offences against five victims, including multiple rapes, targeting very young teenagers and “grooming” V1 for himself and others. The trial judge had applied substantial reductions to a 19‑year aggregate for three rapes, including a large discount for very late guilty pleas after a serious brain haemorrhage had prevented him from standing trial with others.

The Court increased the lead rape sentences to 22 years pre‑mitigation, then made a one‑year reduction for serious health consequences, then applied “a little in excess of 10%” for the very late pleas, to reach 18 years and 6 months concurrent on each of the three rape counts. In doing so, the Court:

  • Reinforced that health‑related mitigation and plea‑related credit must be treated separately; and
  • Warned against double counting either way.

g) Jurisdictional limits of the ULS regime

The court refused jurisdiction over the ULS application concerning Harvie Aspden because he had not been convicted of an offence to which the ULS regime applies. This is a concise reminder that section 36 CJA 1988 does not confer a general power to refer any sentence: it is confined to specified offences.

h) Why the other sentences stood

The Court applied the above principles to each remaining offender, consistently distinguishing between what might be “lenient” and what crosses the threshold into “unduly lenient.” Illustrations include:

  • Poulson (17 years): Multiple offences including two rapes; the judge did not categorise as a “campaign of rape” attracting 20+, but uplifted to 17 years. The CACD found this within the reasonable range; leave to refer granted but Reference dismissed. Poulson’s own application to appeal sentence was “simply unarguable.”
  • Harwood (10 years): One rape, one attempted rape, one sexual activity with a child, with violence and injury, sentenced at 10 years (2A starting point of 10; judge indicated 12 years if a mature adult without cognitive issues; reduced for age 18 and cognition). Both the ULS and Harwood’s own appeal failed.
  • Haslam (16 years): Seven counts including three rapes; judge saw it as broadly similar to Poulson but “less prolific,” applying a smaller uplift. Lenient but not unduly so; leave to refer granted, Reference dismissed.
  • Turner (2 years): Two counts of sexual activity with a child in his own home on one occasion; borderline 1A/1B treatment expressly justified by reduced weight of 1A factors; the CACD endorsed this flexible calibration; leave to refer granted, Reference dismissed.
  • Fitzgerald (66 months): One count of assault by penetration; sentence below the starting point but firmly within the range. The judge accepted defence mitigation and the pre‑sentence report contained positives. While the reduction could have been explained more fully, the outcome was within range; leave granted; Reference dismissed.
  • Corley (28 months): Early guilty pleas (October 2023) to two counts of sexual activity with a child, with personal mitigation and positive report; correct application of category 1A starting point (five years), reduction for delay and mitigation, then 20% plea credit. Leave refused.
  • Bainbridge‑Flatters (7 years): Four counts of sexual activity with a child plus drug supply (to V1) and aggravated vehicle taking; uplift for multiple sexual counts and consecutive terms for the non‑sexual offences was proper totality practice. Leave refused.

3) Likely impact on future cases

  • Anchoring multi‑count rape sentencing: The Court’s characterisation of “campaign of rape” as a non‑technical descriptor with a 20‑year floor and no ceiling gives sentencing judges a principled benchmark when offence‑specific guidelines run out of road on aggregation.
  • Flexible guideline application: Endorsement of weighting within and across categories will encourage nuanced, fact‑sensitive calibration, particularly where 1A factors are present but only lightly (Turner).
  • Delay: Pandemic and investigative complexity limit the scope for “delay credit”; defendants’ choices are relevant. Early admissions (as with Corley) will be rewarded; very late pleas (as with Darbyshire) attract heavily curtailed credit.
  • ULS discipline: The judgment signals that the CACD will be slow to re‑enter the sentencing arena where the trial judge’s decision is within the reasonable range, and invites the Law Officers to be more selective in sprawling, multi‑defendant references.
  • Youth discounts: The Barrett analysis confirms robust application of the young person guideline, including large percentage reductions from an adult comparator where a 17‑year‑old is involved, even in extremely grave offending.
  • Health mitigation and double counting: The Court’s structured recalibration in Darbyshire offers a template for separately valuing serious health impacts and late plea credit without overlap.

Complex Concepts Simplified

  • Unduly Lenient Sentence (ULS) Reference (s.36 CJA 1988): The Law Officers can ask the CACD to review certain sentences thought “unduly lenient.” The Court will only interfere if the sentence falls outside the range of reasonable options. “Lenient” is not enough; it must be unduly lenient.
  • “Campaign of rape” benchmark: In the rape guideline, if the offending amounts to a persistent “campaign,” sentences of 20+ years may be justified. It’s a common‑sense descriptor, not a rigid category. If engaged, 20 years is the starting floor, not a cap.
  • Totality: When sentencing multiple offences, the court must ensure the overall sentence reflects the full criminality without double counting. Often, a judge increases the lead count(s) and runs other counts concurrently; sometimes consecutive terms are added for genuinely separate criminality (e.g., unrelated drugs or vehicle offences).
  • Guideline categories and weighting: Culpability/harm categories (like 1A vs 1B) are not boxes to tick. Judges should assess the real weight of factors; a case can be calibrated at the border between categories where appropriate.
  • Delay as mitigation: Delay may mitigate in some cases (e.g., where the defendant has moved on positively; anxiety; prosecutorial fault). But where delay is due to complex investigations, pandemics, or the defendant’s denials, it usually carries little or no weight. Early pleas can change that analysis.
  • Credit for guilty pleas: The later the plea, the smaller the discount. Very late pleas (post‑trial) typically attract around 10% unless there are exceptional circumstances.
  • Avoiding double counting: Courts must not count the same feature twice, whether aggravating or mitigating. For example, if health issues already reduce a term, they should not also inflate plea credit.
  • Anonymity of victims: Publication of any material likely to identify complainants in sexual cases is prohibited for life under the relevant statutory regime. That protection is non‑negotiable unless lawfully lifted.

Conclusion

Poulson & Ors provides clear, practical guidance for the most challenging area of sexual sentencing: multi‑count, multi‑offender cases where offence‑specific guidelines do not fully address aggregation. The decision:

  • Recasts the rape guideline’s “campaign of rape” language as a usable benchmark — with a 20‑year minimum anchoring point and no upper limit — to calibrate sentences where multiple rapes and persistent conduct must be reflected in the totality;
  • Confirms that guideline application is evaluative, not mechanical: judges may and should weigh culpability factors and, in borderline cases, position sentences across adjacent categories where justified (Turner);
  • Restricts the mitigating effect of delay where caused by investigative complexity, pandemic disruption, or defendants’ litigation choices, while rewarding truly early pleas (Corley);
  • Demands discipline in ULS references and preserves a healthy deference to the fact‑sensitive evaluations of experienced trial judges; and
  • Demonstrates structured mitigation assessment — separating health impacts and plea credit, and avoiding double counting — in revising the sentence of the principal protagonist (Darbyshire).

The net effect is a strong, workable framework for future sentencing in group grooming and similar multi‑count sexual cases. It enhances coherence (via the “campaign of rape” benchmark), respects judicial discretion in nuanced categorisation, and clarifies the limited circumstances in which mere passage of time will mitigate. For appellate practice, it re‑affirms that only sentences outside the reasonable range will be recast — and that the appellate lens is one of principled restraint.

Case Details

Year: 2025
Court: England and Wales Court of Appeal (Criminal Division)

Comments