Totality and Consecutive Sentences for Threats to Kill Within a Pattern of Coercive Control

1) Introduction

In Southgate, R. v [2026] EWCA Crim 841, the Court of Appeal (Criminal Division) (Lady Justice Andrews, Mr Justice Bennathan, and His Honour Judge Picton) dismissed an appeal against sentence imposed in the Crown Court at Chelmsford.

The appellant was convicted after trial of (i) making a threat to kill arising from an incident on 9 November 2024, and (ii) controlling or coercive behaviour in an intimate or family relationship over roughly 2 years and 3 months. The complainant was the appellant’s partner at the time.

The appeal focused on: (a) whether the judge wrongly categorised harm—particularly for the coercive control offence; (b) whether the sentencing exercise involved impermissible “double counting”; and (c) whether the overall outcome was manifestly excessive, especially due to consecutive sentences and alleged inadequate allowance for totality.

2) Summary of the Judgment

  • The Court upheld the trial judge’s assessment that harm category 1 applied to both offences.
  • The Court rejected arguments that the judge “double counted” distress/panic (for threats to kill) or the domestic context (for totality).
  • The Court held the judge was entitled to impose consecutive sentences to reflect distinct criminality, while still making a sufficient reduction for totality.
  • The Court emphasised that the appellant was “fortunate” the threats to kill were not sentenced as higher culpability, given their prolonged and graphic nature against a background of coercive control.
  • The overall sentence of 5 years 5 months’ imprisonment was not manifestly excessive; the appeal was dismissed.

3) Analysis

A. Precedents Cited

The judgment does not cite prior appellate authorities by case name. Instead, it is anchored in:

  • Sentencing Council guideline frameworks for (i) threats to kill and (ii) controlling/coercive behaviour, including category-based culpability and harm assessment and the use of starting points and ranges.
  • Guidelines on Domestic Abuse, relied on to underline that offending “in a domestic setting” is treated as particularly serious.
  • Sentencing Council Guideline on Totality, applied to determine whether sentences should be concurrent or consecutive, and what adjustment was required to ensure proportionality.
  • Statutory aggravation of relevant previous convictions (including a prior domestic assault), treated as materially increasing seriousness and future risk.

The absence of case-law citation is itself notable: the Court treated the appeal as a conventional guideline-and-totality evaluation, giving weight to the trial judge’s advantage of having presided over the trial and assessed the complainant’s evidence first-hand.

B. Legal Reasoning

i) Harm categorisation: distress, fear, and real-world impact

The appellant argued that the coercive control offence should not be harm category 1 because there was no evidence of “serious psychological harm”, distress was not “very serious” for offending of this type, and fear of violence was not shown on many occasions.

The Court rejected that submission, relying on the complainant’s trial and ABE evidence and her victim personal statement, highlighting:

  • the prolonged controlling behaviour (monitoring, isolation from friends, demands for constant contact, online surveillance);
  • the escalation into face-to-face threats and the complainant’s extreme fear and panic response;
  • post-offence consequences (security measures, ongoing fear of being killed, and behavioural adaptations to manage risk).

The Court’s approach indicates that category 1 harm is not confined to clinically-diagnosed psychiatric injury. Where the evidence shows sustained intimidation, profound fear, and significant life changes (including safety planning), category 1 harm is sustainable.

ii) Culpability and the seriousness of threats made during coercive control

The threats to kill were sentenced as category B1. On appeal, the Court went further: it stated the appellant was “fortunate” the offence was not treated as A1, because the threats were prolonged, graphic, repeated, and delivered while the complainant was effectively trapped in her home and suffering a lengthy panic attack.

This reasoning matters in practice: the Court treated the manner, setting, and psychological domination as culpability-elevating features, even where there was no weapon produced and no physical violence used in the incident itself.

iii) “Double counting” arguments

Two double-counting complaints were advanced:

  • Threats to kill: that the complainant’s panic attack/distress was used both to place the offence in B1 and to justify further uplift. The Court did not accept that the judge’s approach was wrong; the sustained, varied, and sadistic nature of the threats, and the complainant’s acute vulnerability at the time, justified moving up within the range.
  • Totality/domestic context: that the “domestic” nature of the offences was counted twice when justifying consecutive sentences. The Court treated domestic context as a seriousness factor (per domestic abuse guidance) but still accepted that the sentencing structure could recognise distinct criminality.

The underlying message is that “double counting” is not established merely because the same background (e.g., domestic setting or victim distress) is relevant to multiple steps. The question is whether the judge has penalised the same feature twice for the same purpose, rather than using it legitimately for different analytical tasks (classification, positioning within range, and structuring for totality).

iv) Totality and the choice between consecutive and concurrent sentences

The central structural dispute was whether the threats to kill were “part and parcel” of the coercive control course of conduct such that concurrency was required. The judge imposed consecutive sentences but reduced the overall tariff by six months to reflect overlap/similarity.

The Court upheld that approach, emphasising:

  • the threats on 9 November 2024 were an extreme incident warranting “separate recognition”;
  • sentencing is not a single mandatory structure; different judges may structure differently, provided the final outcome is proportionate;
  • standing back, a lead-offence approach (threats to kill uplifted to reflect overall criminality, with a concurrent sentence for coercive control) could readily have produced a similar overall term.

In effect, the Court treated the imposed total sentence as a proportionate expression of overall criminality, and used the “alternative structure” comparison as a robustness check: if an equally justifiable concurrent-based structure could reach a similar total, the consecutive-based structure is unlikely to be manifestly excessive.

C. Impact

  • Category 1 harm can be evidenced by lived consequences: extensive safety measures, ongoing fear, isolation, and behavioural change may support the highest harm category even without formal psychiatric diagnosis.
  • Threats to kill within domestic abuse can approach higher culpability without weapons or injuries: repeated, graphic threats delivered in circumstances of control and confinement can justify movement towards the top of the guideline range and may even justify higher culpability classification in an appropriate case.
  • Totality is outcome-focused and structure-flexible: appellate review will often ask whether the overall sentence is proportionate and defensible, not whether there was only one correct way to arrange concurrency/consecutivity.
  • Trial judge advantage matters: where harm and impact are contested, the Court of Appeal may place substantial weight on the trial judge’s ability to assess the complainant’s evidence directly.

4) Complex Concepts Simplified

Culpability and harm categories (and “starting points”)
Sentencing guidelines often assign an offence a culpability level (how blameworthy the conduct was) and a harm level (the seriousness of the impact/risk). Each combination has a “starting point” and a range. Judges then adjust within the range for aggravating/mitigating factors.
Totality
When sentencing for multiple offences, the court must ensure the overall sentence is just and proportionate to the total offending. This can require concurrency, consecutivity, or a mixture, plus a final adjustment after “standing back”.
Concurrent vs consecutive sentences
Concurrent sentences run at the same time; consecutive sentences stack. The choice often depends on whether offences are essentially one episode/transaction, or whether each offence represents distinct harm/criminality needing separate punishment.
Double counting
This occurs when the same factor is unfairly used twice to increase punishment for the same reason. It is not automatically double counting for the same background feature (e.g., domestic context) to be relevant to more than one part of the sentencing analysis.

5) Conclusion

[2026] EWCA Crim 841 confirms a robust approach to sentencing domestic abuse cases where coercive control culminates in severe threats to kill. The Court upheld category 1 harm for both offences based on compelling evidence of fear, distress, and long-term practical impact, and it endorsed consecutive sentences where an acute threats-to-kill episode warrants distinct recognition—provided the judge properly applies totality by “standing back” and ensuring overall proportionality.