GW v R [2026] NICA 6 — Sentencing Must Reflect a Narrow Interpretation of Specimen-Count Verdicts (and Credit Unexplained Delay)

1) Introduction

GW v R is a Northern Ireland Court of Appeal sentencing decision arising from convictions for historic child cruelty / ill-treatment / wilful assault offences contrary to section 20(1) of the Children and Young Persons Act (Northern Ireland) 1968. The appellant (anonymised to protect the identity of his children) was convicted after a long procedural history involving multiple trials. He ultimately stood convicted of five counts relating to three child victims: two sons (A and B) and a stepson (C).

The Crown Court imposed an aggregate sentence of eight years’ imprisonment by making consecutive terms across victims. The appeal contended the sentence was manifestly excessive because (among other matters) the sentencing judge treated “specimen” counts as proof of repeated conduct, made insufficiently cautious use of victim impact material, and failed to allow for extreme investigative/prosecutorial delay.

The central legal issue was not whether custody was justified (the defence accepted the custody threshold was crossed), but whether the sentencing exercise remained faithful to the jury’s verdicts on the indictment—especially where counts were framed as specimen counts and where the accused had been acquitted of more serious allegations.

2) Summary of the Judgment

The Court of Appeal (Keegan LCJ, O’Hara J and McLaughlin J) allowed the appeal and substituted a total sentence of four years six months’ imprisonment (reduced from eight years). The Court held that errors of principle infected the original sentence, chiefly because:

  • The judge did not sentence on a narrow interpretation of the verdicts and did not proceed on the version most favourable to the defence where the verdict basis was uncertain.
  • The judge impermissibly treated conviction on specimen counts as evidencing multiple further incidents of the same type (contrary to fundamental principle).
  • The judge did not sufficiently “cast a critical eye” over victim impact statements which appeared to address the total alleged history, including matters on which the appellant was acquitted.
  • The judge failed to make a proper deduction for delay, despite an extreme overall delay and a conceded period of unexplained inactivity.

Re-sentencing, the Court treated the relevant specimen counts as reflecting single incidents (and, where appropriate, “two incidents” only where the structure of the counts justified it), imposed revised terms, preserved the legitimacy of consecutive sentencing to reflect multiple victims, and then deducted six months for delay.

3) Analysis

A. Precedents Cited

(i) The “specimen count” sentencing prohibition

The Court anchored its analysis in a well-established line of authority (summarised through Blackstone’s Criminal Practice 2025 D20.57) that limits sentencing courts to the offence(s) of conviction and prevents them from inflating sentence by treating conviction on a sample/specimen count as proof of other uncharged or unconvicted instances.

  • Burfoot (1990) 12 Cr App R(S) 252: cited for the proposition that, following a contested trial, an offender should not be sentenced as if guilty of other offences not on the indictment merely because the prosecution described the count as “specimen”. The underlying concern is protection of the accused’s right to jury trial: omitted offences cannot be “smuggled back” at sentence.
  • In Clark [1996] 2 Cr App R(S) 351 (following McKenzie (1984) 6 Cr App R 99): relied on for the “weight of authority” that a sentencer must not proceed on the basis of further similar offending unless the defendant admits it.
  • In Canavan [1998] 1 Cr App R 79: central to this appeal. The Court in GW v R treated R v Canavan as stating a fundamental principle: a court cannot base sentence on offences not forming part of the offence for which the defendant is being sentenced. GW v R applies this to specimen counts: conviction on a specimen count does not entitle the judge to sentence for a wider course of conduct.
  • In Clifford [2014] EWCA Crim 2245: used to reinforce the boundary between conviction facts and other trial material: it is inappropriate to increase sentence based on evidence admitted for other purposes (there, bad character), rather than the offence of conviction.

(ii) Indictment clarity, specimen counts, and what a verdict means

The Court emphasised that specimen-count practice requires particular care because the indictment and directions must allow: (a) the defendant to know the case to meet, and (b) the sentencer to know precisely what the jury has found proved.

  • R v M [1999] NI 45: treated as “of continued utility” and quoted for general observations on specimen counts, including: identifying specified incidents where possible, jury directions requiring satisfaction of at least one occasion per count, charging a sufficient number of counts to reflect extent, and recognising that categories of offending should be reflected by different counts. GW v R uses this guidance to underline why the sentencing judge must not assume repeated conduct unless the verdict and drafting justify it.
  • R v Rackham [1997] 2 Crim App R 222: referenced within the R v M discussion, supporting the principle that the indictment must be drawn so that the defendant understands the case and the judge understands what has been proved for sentencing.

(iii) Interpreting the verdict in the defendant’s favour where uncertain

  • R v King [2017] EWCA Crim 128: applied for a practical sentencing rule: if there is uncertainty about the interpretation of the jury’s verdict, the judge must sentence on the basis most favourable to the defence. In GW v R, the Court found the sentencing judge did not apply that approach and instead over-estimated the “extent and level” of offending.

(iv) Contextual sentencing principles in child cruelty/neglect cases

The Court placed its re-sentencing within broader principles on child cruelty sentencing, stressing fact-specificity, deterrence, protection of children, and the limited utility of “tariff comparisons”.

  • R v W [2014] NICA 71: identified as the only directly cited authority in the “area” by counsel; the Court noted it concerned significantly more serious offending (including sexual offending) and therefore did not offer a direct comparator tariff.
  • R v CD [2021] NICA 45: quoted for its restatement of the principles summarised in R v W, including the need to consider entire factual context and that repeated actions are more serious than a simple incident.
  • The R v CD quotation itself incorporates: R v Orr [1990] NI 287 (protection and deterrence), R v Bereton [2002] 1 Crim App Reports (S) 63 (limited sentencing authorities; fact-specificity), R v Durkin [1989] 11 Crim App Reports (S) 313 (punishment and deterrence), and Attorney General's Reference (No 105 of 2004) [2005] 2 Crim App Reports (S) 42 (immense variety of facts; seriousness varies). GW v R uses these not to justify sentence inflation, but to locate the proper (reduced) sentence within a principled, fact-sensitive approach.

B. Legal Reasoning

(i) Specimen counts, multi-incident counts, and the Crown Court Rules (Northern Ireland) 1979

A key analytical step is the Court’s distinction between specimen counts and multi-incident counts, and how those concepts interact with Northern Ireland procedure:

  • The Court stated that the Crown Court Rules (Northern Ireland) 1979 do not allow for a multi-incident count “in the same manner as in England and Wales”.
  • Even where English rules permit “multiple count” drafting, the Court noted that absent careful drafting such counts may be presumed to signify conviction based on two incidents rather than one.
  • Crucially, the Court reaffirmed the deeper principle: even if drafting mechanisms could encompass more than one incident, a convicted defendant must not be sentenced for other alleged offences for which he was not indicted or convicted.
  • The Court crystallised a practical rule: a specimen count signifies for sentencing purposes one instance, whereas a multi-incident count signifies two incidents unless specified otherwise.

This is the decision’s core operational guidance to sentencing courts: do not convert “specimen” evidence into a de facto “course of conduct” conviction by the back door.

(ii) Failure to resolve the factual basis dispute and the “narrow verdict” approach

The Court criticised the sentencing process because there was “a level of disagreement as to the substance of the counts” and yet the judge declined to make detailed factual findings. That omission mattered because it created space for the sentencing exercise to drift away from the verdict’s lawful boundaries.

On appeal, having examined the particulars of the counts, the Court held the judge was not entitled to treat the counts of conviction as if they constituted convictions for “multiple other offences of the same type of behaviour”. The Court also found the prosecution “went too far” in maintaining repeat offending for sentencing on counts that did not lawfully support that characterisation.

Applying R v King [2017] EWCA Crim 128, the Court held the judge failed to sentence on the version most favourable to the defence, and therefore committed errors of principle by overestimating the offending.

(iii) Victim impact statements: relevance limited to convicted conduct

The Court accepted the seriousness of the harm described in the victims’ statements, but emphasised a legal constraint: victim impact material must be evaluated against the offences of conviction. Here, the statements appeared to “reflect the views of the victims in relation to the totality of the allegations”, including allegations that led to not guilty verdicts.

The Court’s reasoning is not a criticism of victims; rather, it enforces the integrity of acquittals in sentencing. A judge must ensure that an offender is not sentenced for acquitted allegations under the label of “impact”.

(iv) Delay as a mitigating feature requiring express allowance

The Court accepted there was “extreme” delay (11–17 years overall), with a particularly problematic and unexplained period between 2013–2017. The prosecution conceded that some allowance should have been made.

In re-sentencing, the Court deducted six months from the recalculated total to reflect the delay. This confirms that, where delay is exceptional and not attributable to the defence, a sentencing court should confront it explicitly and quantify an allowance where appropriate.

(v) Re-sentencing structure: consecutive terms, but on a corrected verdict basis

The Court endorsed the legitimacy of consecutive sentencing to reflect offending against multiple victims, especially where there is a “serious breach of trust”. But it recalibrated each component by applying the narrow verdict basis:

  • For A: the 25 June 2007 incident was already treated by the judge as “lower end” (one year). The Court held count 16 should have attracted a similar sentence, substituting one year consecutive (making two years for A-related matters).
  • For C: verbal abuse with humiliation across two counts, treated as two incidents; the Court reduced the total for counts 8 and 9 to 18 months concurrent with each other.
  • For B: the pellet gun incident was a one-off and denied; the Court imposed 18 months.
  • Total before delay: five years; after delay deduction: four years six months.

C. Impact

GW v R is significant in Northern Ireland sentencing practice for historic child cruelty/assault offences because it:

  • Reasserts (and operationalises) the rule that specimen counts are not gateways to sentence for unindicted or unproven repeated conduct, aligning NI practice with the R v Canavan principle.
  • Emphasises the judicial obligation to settle the factual basis where disputed, and where uncertain to adopt the most defence-favourable interpretation (R v King [2017] EWCA Crim 128).
  • Provides practical guidance on how NI courts should conceptualise specimen versus (where applicable) multi-incident counts when translating verdicts into sentencing facts.
  • Reinforces that victim impact must be tethered to the convictions and not become a vehicle for sentencing on acquitted allegations.
  • Signals that unexplained investigative/prosecutorial delay in historic allegations can warrant a quantified reduction.

More broadly, the judgment is likely to influence charging and indictment strategy: if the prosecution wishes a sentence to reflect repeated conduct, it must be achieved through proper charging (and lawful drafting) rather than reliance on “specimen” labelling at sentence.

4) Complex Concepts Simplified

Specimen count
A charge presented as an example of alleged wider misconduct. After a trial, conviction on a specimen count generally permits sentencing only for the one instance proved by that count—not for all the other alleged similar incidents—unless those further incidents are separately charged and proved, or admitted.
Multi-incident count
A count drafted to encompass more than one incident. GW v R explains that (absent specification) such a count may be treated as signifying two incidents; but, critically, it still does not authorise sentencing for other incidents beyond the count’s proper scope.
Narrow interpretation of the verdict
When translating a jury verdict into sentencing facts, the judge must not assume aggravating features not necessarily decided by the jury. If there is uncertainty, the judge must adopt the interpretation most favourable to the defendant.
Totality
The principle that when sentencing for multiple offences, the overall sentence must be just and proportionate to the total offending. Consecutive sentences can be used (especially for different victims), but the final total must not be excessive.
Victim impact statement (VIS)
Evidence describing the harm suffered by a victim. The court may consider it, but it must relate to the offences of conviction; it cannot be used to punish the defendant for allegations that resulted in acquittal.
Delay deduction
A reduction in sentence to reflect exceptional and unjustified delay in the progress of the case, particularly where it is not the defendant’s fault.

5) Conclusion

GW v R [2026] NICA 6 reaffirms a strict boundary between what the jury convicted of and what the court may sentence for, particularly in the common but hazardous territory of specimen counts in historic child cruelty cases. The Court’s insistence on a narrow, defence-favourable interpretation of uncertain verdict foundations, careful handling of victim impact material, and a concrete allowance for unexplained delay collectively strengthens sentencing fairness without undermining the seriousness of child cruelty or the legitimacy of consecutive sentences for multiple victims.