Blake & Ors v Fox: Court of Appeal reasserts Dingle for serious-harm causation, adopts “material contribution” for specific losses, and clarifies mitigation and media republication in libel

Introduction

This appeal refines key aspects of modern English defamation law at the intersection of the Defamation Act 2013 and long-standing common law doctrines. In a highly publicised social media dispute, actor and political figure Laurence Fox counterclaimed in libel after three public figures—Simon Blake, Colin Seymour (stage name “Crystal”), and Nicola Thorp—called him “racist” on Twitter in October 2020. Mr Fox’s replies accused each of them of being a “paedophile”, leading to their claims against him. The High Court (Collins Rice J) dismissed Mr Fox’s counterclaims for failing the “serious harm” threshold under section 1(1) of the 2013 Act, but upheld Mr Blake’s and Mr Seymour’s claims and awarded each £90,000 in damages.

The Court of Appeal (Warby LJ, with whom Laing LJ and Dingemans LJ agreed) has now:

• Allowed Mr Fox’s appeal on his counterclaims: holding that serious harm was established as a matter of inference and that the judge’s approach to causation breached core principles, including the rule in Dingle and the proper test of causation.
• Remitted Mr Fox’s counterclaims to the High Court for trial of the remaining defences (honest opinion for Mr Blake and Mr Seymour; truth for Ms Thorp) and, if necessary, damages.
• Upheld liability on Mr Blake’s and Mr Seymour’s claims against Mr Fox but halved their damages to £45,000 each, giving important guidance on mitigation and the treatment of mainstream media republications that “draw the sting”.

The judgment settles several contested points: (1) the continued, rigorous application of Dingle to the section 1(1) serious-harm inquiry; (2) the necessity of applying the “material contribution” test to causation of specific consequential losses; (3) the role of the single meaning in assessing harm; (4) the limited probative value of mainstream media “discussion” of a libel; and (5) how immediate mitigating steps reduce damages.

Summary of the Judgment

• Mr Fox’s counterclaims (racist tweets):
  • The Court of Appeal held that serious harm to Mr Fox’s reputation was an inevitable inference from the gravity of the allegation (being “a racist”), mass publication, and the presumption of pre-publication good reputation.
  • The High Court wrongly relied on prior third-party publications and specific incidents to suggest an existing bad reputation or to dilute causation of general reputational harm—contrary to Dingle v Associated Newspapers and reaffirmed in Lachaux and Mueen‑Uddin.
  • The judge conflated general reputational harm with specific career consequences and applied the wrong causation test for the latter; the correct test is whether the libel made a material contribution to the specific harm.
  • Result: Mr Fox’s counterclaims pass the serious harm threshold; issues of honest opinion (Blake, Seymour) and truth (Thorp), plus damages, are remitted for retrial.
• Claims by Mr Blake and Mr Seymour (paedophile tweets):
  • Liability upheld. The allegation of paedophilia is exceptionally grave; publication was “mass”; and the claimants’ circumstances heightened vulnerability. Serious harm was properly found.
  • However, damages were excessive. The Court of Appeal reduced each award from £90,000 to £45,000 because:
    • Mr Fox’s immediate mitigating steps (deletion, “baseless insults” clarification, and interviews) materially reduced harm.
    • Mainstream media reports “looked askance” and drew the sting; they should have mitigated, not multiplied, harm.
    • Reliance on Stonewall’s solidarity statement as proof of harm was a misstep.

Analysis

Precedents Cited and Their Influence

• Lachaux v Independent Print Ltd [2019] UKSC 27; [2020] 2 AC 612:
  • Established that s 1(1) requires proof of fact of serious harm, not mere tendency; presumption of actual harm is displaced.
  • Confirmed inferential proof is permissible by reference to gravity, scale, situation, and inherent probabilities.
  • Critically, affirmed that Dingle applies to the serious harm inquiry: defendants cannot rely on similar third‑party publications to reduce or defeat the inference of harm.
• Dingle v Associated Newspapers Ltd [1964] AC 371:
  • Defendants cannot mitigate by pointing to similar libels by others. Here, the Court of Appeal rejected attempts to sidestep Dingle by re‑labelling it a “causation” point in the s 1(1) analysis.
• Mueen‑Uddin v Secretary of State for the Home Department [2024] UKSC 21; [2025] AC 945:
  • Reaffirmed Dingle’s “continued potency” post‑2013 Act, reinforcing the CA’s stance.
• Plato Films Ltd v Speidel [1961] AC 1090 and Scott v Sampson (1887) 8 QBD 491:
  • Bad reputation can be proved only by general reputation evidence from those who know the claimant; specific acts or rumours are inadmissible to prove general bad reputation.
  • The CA held the High Court breached these rules by relying on specific incidents and third‑party tweets to suggest Mr Fox already had a bad reputation for racism.
• Wright v McCormack [2023] EWCA Civ 892; [2024] KB 495:
  • Explains the Pamplin principle (damages only; whether claimant deserves vindication). Distinguishes mitigation tools (provocation, Pamplin) from the s 1 threshold.
  • Supports the CA’s insistence on separating general harm from specific consequences and rejecting “litigation misconduct” as relevant to serious harm.
• Morgan v Associated Newspapers Ltd [2018] EWHC 1725 (QB) and Singh v BCA [2010] EWCA Civ 350:
  • Opinion can be less harmful than fact, context‑sensitive. The CA endorsed considering the “opinion” character of the “racist” tweets as potentially mitigating—without erecting a rule.
• Economou v De Freitas [2018] EWCA Civ 2591 and [2016] EWHC 1852 (QB):
  • Single meaning frames remedies; and republication that “draws the sting” cannot be counted as additional harm—applied here to the mainstream media’s “looked askance” coverage.
• Banks v Cadwalladr [2023] EWCA Civ 219; [2023] KB 524:
  • Applied Dingle in the serious-harm context; supports CA’s approach to inferential harm in mass‑publication cases.
• Cooke v MGN [2014] EWHC 2831 (QB); [2015] 1 WLR 895:
  • “Paedophile” is an exemplar of an allegation where the likelihood of serious harm is plain, informing the gravity analysis for Mr Blake and Mr Seymour.
• Amersi v Leslie [2023] EWHC 1368 (KB) and [2023] EWCA Civ 1468:
  • Useful summary of s 1 jurisprudence; supports the evidential, not speculative, nature of inferential findings.
• Peros v Nationwide News Pty Ltd [2024] QSC 192:
  • Suggested (in Queensland) that defendants may tender third‑party publications to contest causation of (general) harm. The CA expressly disagreed, as inconsistent with Dingle and UK authority.
• Others: Barron v Vines (2016), Sicri v ANL (2021), Napag Trading (2020), Miller & Power v Turner (2023), and Bates v Rubython (2025) inform damages principles and the bad‑reputation rules relied upon but ultimately curtailed on this appeal.

Legal Reasoning

1) Dingle applies to the serious-harm threshold, not just quantum

The CA rejected a line of first‑instance reasoning that sought to distinguish between using third‑party publications to prove existing bad reputation (barred) and using them to contest causation of s 1(1) serious harm (permitted). Warby LJ held that permitting defendants to point to prior similar publications as alternative “causes” of general reputational harm simply repackages the very conduct Dingle forbids. Section 1(1) did not displace Dingle; Lachaux says it would be irrational to apply Dingle in damages but not in serious harm. Mueen‑Uddin cements that analysis at Supreme Court level.

Important nuance: third‑party publications can be used to contest causation when a claimant pleads specific consequences (e.g., “I lost this job because of the libel”). But they cannot be used to dilute or negate the inference of general reputational harm arising from the defendant’s own publication.

2) Separate the analyses: general reputational harm vs specific consequential losses

The CA held the judge erred by collapsing two distinct inquiries. General harm (the esteem in which the claimant is held) may be proved inferentially from gravity, scale, and inherent probabilities; specific losses (e.g., losing an agent) require a focused causation analysis based on evidence of that event.

3) The correct causation test for specific losses is “material contribution,” not dominant or sole cause

For the two specific harms Mr Fox alleged—his agent ending their relationship and the collapse of his acting career—the High Court’s language suggested it sought a dominant or exclusive cause. The CA confirmed the orthodox tort test: the libel need only make a material contribution to the harm. On that footing, the issues are remitted for fresh findings.

4) The opinion/fact distinction can mitigate harm on the facts, but is not a rule

The CA endorsed the possibility—reflected in Morgan and Singh—that the “opinion” character of a statement may reduce its reputational impact, depending on context and source. The judge’s finding that the “racist” tweets were opinions expressed in a contested public debate, and could therefore have somewhat less sting than a flat assertion of fact, was a permissible evaluative judgment; it did not undercut the ultimate conclusion that serious harm to Mr Fox should have been found.

5) Single meaning governs the harm analysis; belief in truth is not generally required

Mr Fox argued that because the court’s “single meaning” (that the claimants were paedophiles) is derived from a stripped‑down context, the serious-harm analysis should accommodate a fuller, real‑world Twitter context, including audience scepticism. The CA reiterated that the single meaning frames the imputations for liability and remedies. Claimants need not prove that readers believed the imputation unless the allegation is inherently outlandish or incredible. “Paedophile” is grave but not fanciful; therefore, the inference of serious harm was not displaced.

6) Mitigation: what counts and what does not

• Counts as mitigation:
  • Immediate steps by the defendant—deleting the tweets, publicly calling them “baseless insults,” and giving interviews reflecting that position—reduced the level and duration of harm.
  • Mainstream media coverage that “looked askance” at the allegation and reported the mitigating tweets “drew the sting” and therefore cannot be added to the quantum of harm; at most, it mitigates it.
• Does not count as mitigation:
  • Third‑party prior publications calling Mr Fox racist (Dingle).
  • Using specific incidents or the claimant’s own past statements to prove general bad reputation (Scott/Plato).
  • Claimants’ “litigation misconduct” (per Wright v McCormack) as a reason to reduce damages or defeat serious harm where it bears no logical relation to the defamatory imputation.

7) Republication and social media dynamics

• The claimants’ own republication of the defamatory material cannot be used to increase harm against Mr Fox in their claims; likewise, Mr Fox’s “quote‑tweet” responses republished the accusations of “racist,” but that did not meaningfully weaken his serious-harm case because the platform mechanics made republication in reply inevitable and his purpose was to rebut the charge.
• Each publisher’s liability must be assessed separately; absent evidence of overlapping audiences, there is no single indivisible injury warranting joint assessment for all publishers.

8) Damages calibration against personal injury benchmarks

Reaffirming the standard approach, the Court compared awards to Judicial College Guidelines to ensure libel damages are not out of kilter with PI awards. Given the mitigating factors, £90,000 per claimant was excessive; £45,000 each was “sufficient to show the world there was nothing in the allegations” and compensate reputational injury and distress (while excluding distress from having to litigate Mr Fox’s counterclaims).

Impact

• Serious harm and Dingle: Defendants cannot rely on a pre‑existing “pile‑on” of similar accusations to defeat or dilute serious harm for general reputation. Dingle is alive, well, and applies at the s 1 threshold, not just damages.
• Pleading and proof of bad reputation: If defendants wish to run “general bad reputation,” they must call general reputation witnesses (“those who know him”), not third‑party publications or specific acts. Rumours and prior articles are out; general evidence from the relevant community is in.
• Case management: Courts and parties must keep separate:
  • (a) inferential general reputational harm; and
  • (b) specific consequential losses.
  Different evidential rules and causation tests apply to each.
• Causation for specific losses: The correct test is material contribution. Claimants need not show the libel was the sole or dominant cause of a lost opportunity or relationship.
• Media republication: Neutral or sceptical coverage that reports both allegation and immediate mitigation draws the sting and should be treated as mitigatory, not aggravating.
• Mitigating steps matter: Prompt deletion, clarifying statements (“baseless insults”), and early interviews meaningfully reduce damages, especially on high‑volume platforms.
• Opinion vs fact: In public‑interest debates, clearly opinion‑framed accusations may carry somewhat less reputational force. But grave opinions (e.g., “racist”) remain capable of causing serious harm at scale.
• Social media realities: Quote‑tweet mechanics, deletion effects, and analytics (impressions) are central to proof and mitigation analyses. Evidence of scale (analytics) will often drive inferential findings.
• Damages trend: For an exceptionally grave, short‑lived social media libel with prompt mitigation and sceptical media reception, awards around £45,000 (per claimant) are within range; £90,000 was excessive.
• Comparative note: The Court expressly distances English law from the Queensland approach in Peros on using third‑party publications to contest causation of general harm.

Complex Concepts Simplified

• Single meaning rule: The court determines the one natural and ordinary meaning an average reader would take. That meaning frames liability, defences, and remedies.
• Serious harm (s 1(1) Defamation Act 2013): Claimant must prove as a fact that publication caused (or is likely to cause) serious reputational harm. This can be inferred from gravity, scale, and context.
• Dingle rule: A defendant cannot rely on similar defamatory statements by others to reduce damages or defeat serious harm. Prior pile‑ons don’t save a defendant.
• General vs specific harm: General harm is damage to overall reputation; specific harm is a concrete consequence (e.g., losing a job). Different evidence and rules apply.
• Material contribution: For specific losses, the libel need only be a real, not trivial, cause; it need not be the sole or main cause.
• Scott/Plato rule on bad reputation: General bad reputation may be proved only by witnesses speaking to reputation, not by specific past acts or third‑party publications.
• Burstein material: Directly relevant background context can be used to explain, not to prove, bad reputation.
• Pamplin rule: Facts properly in evidence for other issues can be used to show the claimant does not deserve full vindication—but this goes to damages, not whether serious harm was caused.
• “Draws the sting”: If republication is in a context that questions or neutralises the defamatory sting, you cannot count that republication as additional harm.
• Reverse innuendo: A defence argument that extra context means reasonable readers took a less defamatory meaning than the single meaning. It may reduce harm, but you need evidence.

Conclusion

Blake & Ors v Fox is a major restatement of defamation causation and mitigation principles in the social media age. The Court of Appeal has:

• Re‑anchored the serious harm analysis to the core Lachaux framework and confirmed that Dingle bites at the threshold as well as at damages, preventing defendants from weaponising prior pile‑ons to defeat causation of general harm.
• Clarified that specific consequential losses turn on the orthodox “material contribution” test, not dominant or sole causation.
• Provided practical guidance on mitigation: immediate deletions and clarifications count, and neutral/sceptical mainstream coverage reduces rather than inflates harm.
• Confirmed that opinion may carry less sting than fact in context, without erecting a categorical rule, and that claimants generally need not prove that readers believed the allegation to establish serious harm.
• Reset damages calibration for short‑lived but very grave social media libels where meaningful mitigation occurs, aligning awards with broader personal injury benchmarks.

Going forward, claimants can meet the s 1 threshold inferentially in mass‑publication cases involving grave imputations; defendants will need to focus on admissible mitigation rather than third‑party pile‑ons or specific acts to show pre‑existing bad reputation. Trial management should keep general harm analytically distinct from specific losses, and causation language must reflect the material contribution standard. The decision will become a central citation in serious-harm litigation, social media defamation, and damages assessments across England and Wales.