Section 2(9) Terrorism Act 2006 defence: universally available, objectively assessed, and proved by the defendant

Case: R v Hossen Neutral citation: [2026] EWCA Crim 1065 (CA (Crim Div), 7 August 2026)
Court below: Crown Court at Kingston (HHJ Lodder KC)
Judgment: Norton J (with Edis LJ (VP) and Choudhury J agreeing)

Key holdings (new guidance):
  • Availability: the statutory defence in s.2(9)–(10) Terrorism Act 2006 is not confined to journalists, academics, librarians, or any occupational class; it is fact-specific and turns on whether the defendant can satisfy both limbs.
  • Limb (b) is objective: “clear in all the circumstances” in s.2(9)(b) requires that it be objectively clear that the publication did not express the defendant’s views and lacked endorsement—mere denial will often be insufficient.
  • Burden: “it is a defence … to show” imposes a legal (persuasive) burden on the defendant on the balance of probabilities, not merely an evidential burden.
  • Application: the defence was rightly not left to the jury because there was no sufficient evidential basis for the objective limb; convictions and sentence were upheld.

1. Introduction

The appellant was convicted of five counts of disseminating terrorist publications contrary to s.2(1) Terrorism Act 2006 after sending two agreed “terrorist publications” (“The Book of Jihad” and “44 Ways to Support Jihad”) to a small group including his future spouse and others (identified by initials in the judgment). The live issue at trial was mental element: whether dissemination was done with intent to encourage terrorism (directly or indirectly) or, alternatively, with recklessness as to that effect.

During deliberations, the jury asked for guidance about how “reckless” and “indirect” should be understood, posing hypotheticals about downstream effects (e.g., recipients developing a supportive mindset, further disseminating, or donating to terrorist groups). In response to that note, the defence for the first time invited the judge to leave the statutory defence in s.2(9)–(10) to the jury. The judge declined, reasoning that the defence was intended only for a limited class (such as academics or journalists) and that the appellant could not bring himself within it.

The appeal therefore raised three linked questions of broader importance: (i) who can rely on the s.2(9) defence; (ii) what must be proved to make it out (especially limb (b)); and (iii) who bears the burden and to what standard. A further issue arose on sentence: whether the jury note suggested the jury may have convicted on recklessness rather than intent, potentially affecting offence categorisation.

2. Summary of the Judgment

  • Conviction appeal dismissed. The Court of Appeal held that the s.2(9) defence is not limited by occupation or status; however, it requires proof of both limbs, and limb (b) is objective. On the facts, there was no sufficient evidential basis to satisfy limb (b), so the trial judge was right (albeit for different reasons) not to leave the defence to the jury, and the convictions were not unsafe.
  • Sentence appeal dismissed. The jury note did not justify sentencing on a more favourable basis. Applying R v Cloud [2001] EWCA Crim 510, where the verdict basis is unclear it is for the judge to determine the facts for sentence. The judge was entitled to find intent proved and to sentence accordingly, including the extension period under s.278 Sentencing Act 2020.

3. Analysis

3.1 Precedents and materials cited

(a) Roger Sliney v London Borough of Havering [2002] EWCA Crim 2558 — “to show” as a legal burden

The court’s most consequential doctrinal move was to characterise the defendant’s burden under s.2(9) as persuasive. To reach that conclusion, it relied closely on Roger Sliney v London Borough of Havering [2002] EWCA Crim 2558, where similar statutory language (“It is a defence … to show…”) in the Trade Marks Act 1994 was treated as imposing a legal burden.

The court adopted Sliney’s approach that, in context, “show” is synonymous with “prove”, and that this drafting technique is “classic language” for a persuasive burden.

Although the parties had agreed only an evidential burden applied, the Court of Appeal stated (obiter in the sense that it said it was not strictly necessary, but expressed in firm terms) that the burden is legal, discharged on the balance of probabilities.

(b) Salabiaku v France [1988] 13 EHRR 379 — compatibility with Article 6(2)

Having found a legal burden, the court addressed proportionality under the presumption of innocence. It invoked Salabiaku v France [1988] 13 EHRR 379 for the proposition that presumptions (and, by extension, reverse burdens) are permissible if kept “within reasonable limits” and maintain defence rights. The court’s justification centred on:

  • the prosecution still proving all elements of the s.2 offence (including recklessness);
  • the strong public interest in controlling dissemination of terrorist material;
  • the first limb being within the defendant’s own knowledge, and limb (b) being a jury’s objective assessment;
  • the defence being designed to protect “objectively legitimate purposes”, making the burden not unduly onerous for its paradigmatic users.

(c) R v Cloud [2001] EWCA Crim 510 — sentencing where verdict basis is unknown

On sentence, the court treated R v Cloud [2001] EWCA Crim 510 as determinative of approach: uncertainty about what the jury thought does not mean the defendant automatically receives the “benefit of the doubt”; the sentencing judge must decide the factual basis.

“It was for the judge to decide.”

(d) Other authorities referenced in quotation: “Lambert” and “Johnstone”

Within the quoted passage from Sliney, there are references to “Lambert” and “Johnstone” as part of the established landscape on statutory interpretation of burdens in criminal defences and the caution against lightly construing reverse burdens. The present court did not conduct an independent analysis of those authorities, but adopted the settled reading of “to show” from Sliney.

(e) Parliamentary and explanatory materials

The court was shown:

  • the Explanatory Notes to s.1 (encouragement of terrorism), stating the similarly worded defence in s.1(6) is “intended, for example, to cover news broadcasters”;
  • a letter appended to the Government’s written response to the Joint Committee on Human Rights report: Counter-Terrorism Policy and Human Rights: Terrorism Bill and Related Matters (Third Report of Session 2005-06, HL 75-I, HC 561-I), indicating concern to protect legitimate librarians, academics, teachers, and others.

The court accepted that these materials illuminate legislative concern, but stressed limits on their role in construction—particularly that such letters are “even less admissible as aids to construction.”

3.2 Legal reasoning

(a) The defence is not status-based

The trial judge’s central error was to treat s.2(9) as confined to a “specific class” (academics/journalists). The Court of Appeal rejected that as inconsistent with statutory text:

  • No words of limitation appear in s.2(9)–(10).
  • The absence of an Explanatory Note for s.2(9) (contrasted with s.1(6)) did not justify importing a restriction.
  • Importing occupational categories would generate “arid and ultimately inconsequential arguments” about who qualifies.

The court’s construction therefore separates purpose (protecting legitimate dissemination) from eligibility (open to any defendant who can satisfy the test).

(b) The two-limb structure matters: limb (b) adds an objective “clarity” requirement

The core operational guidance is the court’s insistence that s.2(9) is not satisfied by the defendant merely saying, “it wasn’t my view” or “I didn’t endorse it.” Limb (b) requires that, in all the circumstances of the conduct, it was clear that the publication did not express the defendant’s views and did not have endorsement. The court held:

  • limb (b) is “separate and distinct” from limb (a);
  • “clear in all the circumstances” must mean objectively clear;
  • what satisfies limb (b) is fact-specific and for the tribunal of fact to evaluate.

Practically, the court indicated that occupational context (teacher/librarian/journalist acting in role) is often a strong evidential route to making the objective clarity obvious. But the logic is broader: any context that objectively signals non-endorsement (e.g., accompanying disavowal, critical framing, educational setting, research context, explicit warning) may be capable of meeting limb (b) if evidenced.

(c) Burden: a legal burden on the balance of probabilities

The court’s articulation is explicit: s.2(9) places a legal (persuasive) burden on the defendant for both limbs, to be proved on the balance of probabilities. Its reasoning proceeds in two steps:

  1. Textual: “it is a defence … to show” is classic reverse-burden language.
  2. Authority: Roger Sliney v London Borough of Havering [2002] EWCA Crim 2558 treats “show” as “prove” in this context.

The court also sought to pre-empt Article 6(2) objections by grounding proportionality in public protection and the idea that the defence is an “exemption” operating after the offence elements are proved.

(d) Why the defence failed on the facts: no sufficient basis for limb (b)

Importantly, the appellant lost not because the defence was unavailable in principle, but because it was unavailable on the evidence. The court held that the defence was raised late, not foreshadowed, and—most decisively—counsel could not identify evidence capable of satisfying the objective clarity limb. The court therefore concluded:

  • the trial judge was right to refuse to direct the jury on s.2(9) when the jury note prompted the point;
  • there was “no or no sufficient evidence” to justify leaving the defence.

(e) Sentence: jury note did not undermine a finding of intent for sentencing

The appellant argued the jury note was “perhaps indicative” the jury had rejected intent. The court rejected that as speculative: the note might reflect one juror, might relate to one count, might be exploratory, and might have changed by verdict. Applying R v Cloud [2001] EWCA Crim 510, the judge was entitled—indeed required—to find the factual basis. The sentencing judge’s conclusion of intent (based on messages, “mindset material”, and “progressive” introduction of extremist material) was within the proper ambit of trial-judge evaluation.

3.3 Impact and significance

  • Wider accessibility of the defence, but stricter proof demands: defendants are not barred by their lack of professional status, yet the defence becomes harder to run unless the circumstances of dissemination objectively communicate non-endorsement.
  • Trial management and routes to verdict: where a s.2 case is left on recklessness, courts and parties should consider early whether s.2(9)–(10) is realistically in play, and if so craft a verdict route that sequences: (i) intention; (ii) recklessness; (iii) statutory defence.
  • Reverse-burden confirmation: the court’s treatment of “to show” as a persuasive burden is likely to be cited in future Terrorism Act prosecutions when judges must direct on burdens and standards for statutory defences.
  • Objective clarity as the central battleground: the decision signals that the defence is less about subjective disavowal and more about objectively legible context (what a reasonable observer of the dissemination circumstances would conclude).
  • Sentencing resilience despite jury notes: speculative readings of jury communications will rarely displace the Cloud principle; judges must decide the basis for sentence from the evidence they heard.

4. Complex concepts simplified

  • “Terrorist publication” (s.2(3)): material whose content is likely to be understood (by a reasonable person) as encouraging terrorism (directly or indirectly), or as useful for committing/preparing terrorism and presented as such.
  • “Indirect encouragement”: includes “glorification” from which a person could reasonably infer that the glorified conduct should be emulated (s.2(4)).
  • Recklessness (as directed): foreseeing a risk of the prohibited effect (encouragement/assistance) and unreasonably taking that risk anyway.
  • The s.2(9) defence in plain terms: if the prosecution has not proved intent to encourage (s.2(10)(b)), a defendant accused on recklessness can avoid conviction by proving:
    • (a) subjective non-endorsement: it wasn’t his view and he didn’t endorse it; and
    • (b) objective clarity: the circumstances made that non-endorsement clear.
  • “Legal (persuasive) burden” vs “evidential burden”: an evidential burden requires only raising an issue fit for consideration; a legal burden requires persuading the fact-finder of the defence on the balance of probabilities.
  • Extension period / offender of special concern: the court upheld the extension period imposed under s.278 Sentencing Act 2020, which can add a further licence/extension component to the custodial term for specified offenders.

5. Conclusion

R v Hossen clarifies three important points about s.2 Terrorism Act 2006. First, the statutory defence in s.2(9)–(10) is not an occupation-specific carve-out: it is available to any defendant in principle. Second, it is substantively demanding because limb (b) requires an objective showing that non-endorsement was clear in all the circumstances. Third, the defendant must prove both limbs on the balance of probabilities. On the facts, the appellant could not point to evidence satisfying the objective clarity requirement, so the convictions and sentence stood. The decision is likely to shape how terrorism-publication cases are defended and directed, focusing future disputes on whether dissemination circumstances objectively communicated non-endorsement, and confirming that this exemption is one the defendant must establish.