Judicial Proceedings Immunity and Whistleblowing Detriment: Rogerson v Erhard-Jensen Ontological/phenomenological Initiative Ltd [2025] EWCA Civ 1547

Judicial Proceedings Immunity and Whistleblowing Detriment: Commentary on Rogerson v Erhard‑Jensen Ontological/phenomenological Initiative Ltd [2025] EWCA Civ 1547


1. Introduction

This Court of Appeal decision re‑draws the boundary between whistleblower protection and judicial proceedings immunity (“JPI”). It establishes that an employer’s commencement of court or arbitral proceedings against a worker can itself constitute an actionable “detriment” under the Employment Rights Act 1996 (“ERA”) where motivated by a protected disclosure, and that JPI does not bar such a claim.

The case arises from post‑employment whistleblowing by Daniel Rogerson, a worker who alleged that Werner Erhard had regularly been verbally and physically abusive towards staff. His former employer is a Singapore‑registered charity (the Respondent) linked to Mr Erhard. After Rogerson’s disclosures, the Respondent commenced arbitration in Singapore under an ICC arbitration clause, claiming breach of a confidentiality and consulting agreement and seeking substantial liquidated damages and injunctive relief. Rogerson alleges that this was done to punish and stifle his protected disclosures and thereby constituted a “detriment” under s.47B ERA.

The central issue on appeal was narrow but important: whether this alleged detriment (“Detriment 3” – the commencement of the Singapore arbitration) is in principle actionable, or whether it falls within the “core ambit” of JPI, with the result that the claim must be struck out. The Employment Tribunal (“ET”) refused to strike it out. The Employment Appeal Tribunal (“EAT”) reversed, holding that JPI barred the claim. The Court of Appeal (Andrews, Elisabeth Laing and Males LJJ) has now restored the ET’s decision.

The judgment has significance in three intersecting areas:

  • whistleblowing protection under ERA, especially for post‑employment detriment;
  • the scope and limits of judicial proceedings immunity (including in arbitration); and
  • the relationship between domestic public policy on protected disclosures and foreign arbitration (and awards) under the New York Convention.

2. Summary of the Judgment

2.1 Core statutory framework

  • s.47B(1) ERA: a worker has the right not to be subjected to any detriment by any act (or deliberate failure to act) by his employer done on the ground that he has made a protected disclosure.
  • s.48(1A) ERA: a worker may complain to an Employment Tribunal of such a detriment.
  • s.43J ERA: any contractual term that purports to preclude a worker from making a protected disclosure is void to that extent.

2.2 Key factual and procedural background

  • Rogerson alleged protected disclosures concerning abuse of staff by Werner Erhard.
  • He had signed a 2016 Confidentiality and Independent Consulting Agreement containing a Singapore‑seated ICC arbitration clause.
  • In July 2021 the Respondent commenced ICC arbitration in Singapore, alleging that Rogerson’s disclosures breached the confidentiality agreement, and seeking:
    • liquidated damages of £50,000;
    • an injunction restraining him from contacting staff; and
    • costs.
    The arbitration claim accused him of an “extortion scheme” by making false allegations of abuse in order to extract a settlement.
  • In the ET claim, Detriment 3 was pleaded as the bringing of those arbitration proceedings.
  • The ET refused to strike out Detriment 3.
  • The EAT (Heather Williams J) allowed the Respondent’s appeal, holding that Detriment 3 fell within the core ambit of JPI and therefore was not actionable.
  • The Court of Appeal allows Rogerson’s appeal and reinstates Detriment 3.

2.3 Main holdings

  1. The commencement of Singapore arbitration proceedings can plainly amount to a “detriment” within s.47B ERA, on an ordinary meaning of “harm or damage”. It involves cost, stress, and exposure to financial and injunctive remedies (paras 9–11).
  2. There is no textual or policy basis in ERA for carving out an immunity for detriments constituted by the initiation of legal or arbitral proceedings (paras 10–15).
  3. s.43J ERA renders void, as a matter of UK public policy, contractual provisions that purport to preclude protected disclosures. Any Singapore arbitral award enforcing such a provision (by damages or injunction) would be unenforceable in the UK under s.103(3) Arbitration Act 1996 and the New York Convention (paras 16–17).
  4. Judicial proceedings immunity is a speech‑based immunity: it attaches to statements made in or closely connected with court/tribunal proceedings, not to all acts done in the litigation. Its core rationale is to encourage freedom of expression in judicial proceedings and to protect the administration of justice (paras 31–43).
  5. The immunity does not extend to the bare act of initiating court or arbitration proceedings, even though the proceedings are quasi‑judicial and even though they are commenced by a written document such as a request for arbitration (paras 47–48).
  6. Accordingly, JPI does not bar a s.48(1A) ERA claim complaining that an employer has brought proceedings (including foreign arbitration) as a reprisal for protected disclosures (paras 54–55).
  7. To hold otherwise would frustrate Parliament’s intent in ERA and leave an acknowledged statutory wrong without a remedy (paras 13, 55, 58).

3. Detailed Analysis

3.1 Statutory policy: broad protection for whistleblowers

3.1.1 The meaning of “detriment” and “any act” (paras 9–12)

The ERA does not define “detriment”. The Court proceeds on the basis that it bears its ordinary meaning: “harm or damage”. Lady Justice Andrews emphasises the breadth of the statutory language:

  • “any detriment”: not limited to workplace sanctions; includes economic, reputational, psychological, and procedural disadvantages;
  • “by any act … by his employer”: equally broad; the only qualification is that omissions must be “deliberate”.

On that plain wording, the “act” of commencing legal or arbitration proceedings undoubtedly falls within s.47B(1). The Court rejects any idea of an implied exception for litigation steps: if Parliament had intended to exempt such acts, it would have said so expressly (para 10).

On the facts, Rogerson was exposed to:

  • the costs, trouble and stress of defending foreign arbitration proceedings (12–13 hour flight away);
  • liquidated damages of £50,000; and
  • an injunction designed to cut him off from staff, thereby potentially inhibiting future disclosures or evidence gathering.

These are quintessential forms of “detriment”, especially in a whistleblowing context. Indeed, the Court remarks that, apart from dismissal, it is hard to conceive of a more severe detriment than subjecting a whistleblower to proceedings designed to obtain financial recompense for past disclosures or to block future ones (para 11).

3.1.2 The “on the ground that” requirement (para 12)

The detriment must be “on the ground that” the worker made a protected disclosure, i.e. causally linked to that disclosure. Here, the Respondent’s arbitration claim:

  • expressly characterised the disclosures as false allegations and an “extortion scheme”;
  • treated them as breaches of confidentiality.

The Court considers it “difficult to see” how the Respondent could deny that the arbitration was commenced “on the ground” of the protected disclosures, although that issue will be determined by the ET at trial (para 12). For the present appeal, the claimant’s case is taken at its highest.

3.2 Section 43J ERA and public policy against “gagging” clauses (paras 15–18)

Section 43J ERA is central:

“Any provision in an agreement … is void in so far as it purports to preclude the worker from making a protected disclosure.”

Two important points flow from this:

  1. The clause is void only to the extent it would bar protected disclosures; it may still validly protect other confidential information (trade secrets, commercial data, etc.) (para 15).
  2. As to the present facts, the Court agrees (and goes beyond) Judge Fowell’s conclusion:
    • If the Respondent had sued in the UK to enforce the confidentiality agreement so as to prevent protected disclosures, that action would fail because s.43J deprives the clause of legal effect for that purpose (para 16).
    • Any foreign arbitral award treating that clause as valid to preclude protected disclosures would be denied recognition or enforcement in the UK on public policy grounds under s.103(3) Arbitration Act 1996 (para 16–17).

The Court is explicit: UK public policy, clearly expressed in s.43J ERA, is that contractual terms suppressing protected disclosures are void. Therefore, a foreign award that:

  • awards damages for making protected disclosures, or
  • grants injunctions restraining such disclosures,

would not be enforceable in the UK, even if the contract is governed by foreign law and the arbitration is seated abroad (para 16–17).

This has two systemic implications:

  • Employers cannot use foreign arbitration and foreign governing law to circumvent UK whistleblowing policy.
  • Whistleblowers have a measure of security that, at least within the UK, such awards cannot be used against them.

Andrews LJ also endorses Judge Fowell’s criticism of attempts to draw a fine distinction between:

  • arguing that the confidentiality agreement bars the whistleblowing claim, and
  • arguing that the employer is immune from suit because it commenced proceedings to enforce that very agreement (para 18).

To allow the latter would subvert Parliament’s decision in s.43J.

3.3 The key question: does judicial proceedings immunity apply?

With detriment and causation apparently satisfied, the only remaining barrier is the Respondent’s invocation of JPI. The analytical structure is:

  1. Is the Singapore arbitration quasi‑judicial so that JPI can in principle arise? (Yes, and not disputed: para 19.)
  2. What is the scope of JPI – to which conduct does it apply, and against which claims?
  3. Does it cover the act of commencing arbitral proceedings alleged here?

3.3.1 The Respondent’s argument: “core immunity” for initiating proceedings (paras 21–24)

The Respondent’s core submission (advanced by Mr Kemp KC) is derived from Devlin LJ’s analysis in Lincoln v Daniels [1958] 1 QB 237. Devlin LJ described three categories of absolute privilege for proceedings in or before a court of justice (para 22):

  1. Matters done coram judice (in court) – statements by judges, parties, counsel, witnesses, and evidence documents.
  2. “Everything that is done from the inception of the proceedings onwards” – pleadings and documents, starting from the writ or initiating document.
  3. Pre‑trial witness preparation, following Watson v M’Ewan and Beresford v White.

Mr Kemp argued that most modern cases about JPI concern the more problematic third category (extensions to core immunity). By contrast, the present case falls squarely in category (2): the request for arbitration is the initiating document; JPI attaches to such a step as “core” and bars any claim arising from its existence. He also advanced a more policy‑based contention: parties should be free to initiate proceedings without fear of being sued for doing so (para 24).

3.3.2 The Court’s three reasons for rejecting a blanket immunity for initiating proceedings (paras 25–29)

Andrews LJ firmly rejects the proposed “freedom to sue without fear of suit” principle as both novel and inconsistent with authority (para 24). Three main reasons:

  1. Historically there was no civil tort of bringing civil proceedings maliciously.
    Until Willers v Joyce [2016] UKSC 43; [2018] AC 779, malicious prosecution was thought to apply only to criminal proceedings. The recognition that it can encompass malicious civil proceedings is recent (para 25). If there is now civil liability for maliciously bringing proceedings, that itself shows that initiating proceedings is not sacrosanct or immune in principle.
  2. Existing torts already restrict abusive litigation: malicious prosecution and abuse of process (paras 26–28).
    • Malicious prosecution: where proceedings are brought with malice and without reasonable or probable cause.
    • Abuse of process: using legal proceedings as a device to achieve an extraneous objective, rather than genuinely seeking the relief claimed.
    In both these spheres, the House of Lords has accepted that JPI does not shield the initiation of proceedings where malice or an ulterior motive can be plausibly pleaded (see Roy v Prior [1971] AC 470; Lord Hoffmann in Taylor v SFO [1999] 2 AC 177 at 215E–G). Thus, there is no absolute JPI for the act of bringing proceedings.
  3. “Anything done” was never literal.
    Older statements that JPI covers “anything said or done” in proceedings (Dawkins v Lord Rokeby (1873) LR 8 QB 255) must be read in context. In Darker v Chief Constable of the West Midlands [2001] 1 AC 435, Lord Hutton construed Kelly CB’s phrase as directed to submission of written witness statements, not all forms of conduct during litigation (para 30).

The cornerstone of the Court’s reasoning is that JPI is not a general immunity for all litigation‑related conduct. It has a narrow, speech‑centred core, to be defined by rationale, not literalism or over‑broad paraphrases.

3.4 The modern doctrine of JPI: Singh and XGY

The Court relies heavily on two modern Court of Appeal decisions:

3.4.1 Singh v Reading BC (paras 31–33)

In Singh, the claimant alleged that the employer had pressured a witness to give untruthful or inaccurate evidence in an ET witness statement. The question was whether JPI barred such a claim. Lewison LJ held it did not.

Two strands of policy were identified (para 32):

  1. Freedom of speech in litigation: participants must be able to speak freely in proceedings, without fearing civil claims.
  2. Avoidance of multiplicity of actions: the courts should avoid collateral litigation over whether evidence given in earlier proceedings was true.

However, Lewison LJ concluded that immunity could “no longer be said” to attach to anything said or done “whatever the nature of the claim” (para 33; see Singh at [61]). The Court must look to the policy underlying the immunity and ask if its extension is truly necessary.

3.4.2 XGY and the “core immunity” (paras 34–42)

XGY involved a vulnerable witness whose address, given confidentially to the police, was revealed in court by a CPS advocate during a bail application. She sued the police and the CPS for breach of confidence, breach of the Human Rights Act and Data Protection Act. The issue was whether the core immunity barred these claims.

The Court of Appeal (Lady Carr LCJ, Sharp P and Coulson LJ) held that:

  • there is a core immunity for advocates, parties, witnesses, judges and jurors for statements made in court, and for certain closely connected statements outside court;
  • this immunity applies regardless of the cause of action (defamation or otherwise) and even where the statement is malicious;
  • investigators and those transmitting information (e.g. police to CPS) can fall within an extension of the immunity where their acts are intimately tied to preparing evidence for court.

In synthesising the authorities, XGY set out key principles quoted by Andrews LJ (para 42):

  1. The core immunity:
    • is necessary for the proper administration of justice;
    • covers statements made in court and certain allied communications “from the inception of the proceedings onwards”; and
    • is founded on public policy encouraging free communication in court proceedings.
  2. Extensions are strictly limited:
    • witnesses and potential witnesses are immune for statements made outside court with a view to giving evidence;
    • investigators are immune for statements that are part of the process of investigating crime or possible crime.

Crucially, XGY emphasised that:

  • JPI is “limited to actions in which the alleged statement constitutes the cause of action” (Taylor, cited at para 36–37); and
  • the touchstone is whether an extension is necessary for the proper administration of justice (para 41).

3.4.3 The Court’s synthesis

From these authorities, Andrews LJ concludes (paras 43–46):

  • The core immunity attaches to statements – spoken or written – made in the course of proceedings; it is, in substance, a speech‑based immunity.
  • The phrase “anything said or done” must be read as referring to acts that are intrinsically connected to the making, transmission or use of such statements.
  • It does not extend to all possible acts within litigation, such as:
    • negligent case conduct (cf. Arthur JS Hall v Simons);
    • the malicious initiation of proceedings (cf. Willers, Roy v Prior); or
    • non‑speech misconduct like flawed disclosure exercises (cf. Daniels).

JPI thus has a narrow, principled core, anchored in the protection of candid participation in judicial decision‑making.

3.5 Applying JPI to the initiation of the arbitration (paras 47–55)

3.5.1 The nature of the alleged wrong

The critical analytic move is to identify what constitutes the alleged wrong in Detriment 3:

  • It is not the content of statements in the request for arbitration or pleadings as such.
  • It is the act of initiating the arbitration proceedings themselves.
  • That act is alleged to have been done “on the ground of” Rogerson’s protected disclosures, thereby subjecting him to detriment.

As Andrews LJ puts it (para 47):

“the cause of action is not founded on statements made in the request for arbitration or in the arbitration itself. The act complained of … is the act of initiating the arbitral proceedings. That is what causes the detriment.”

That act does not fall within core JPI, “any more than the act of initiating proceedings in breach of an agreement to arbitrate, or in breach of an exclusive jurisdiction clause” would (para 47).

3.5.2 Relevance of how the case is pleaded (paras 48–50, 53)

The EAT had relied on the way Detriment 3 was pleaded – emphasising phrases such as “groundless arbitration” and “false allegations” – to conclude that the complaint inherently attacked the content and nature of the proceedings, and hence fell within JPI.

The Court of Appeal rejects that approach, in line with XGY:

  • Whether JPI applies cannot be determined on a fact‑sensitive or pleading‑sensitive basis (para 48, 35).
  • The claimant’s characterisation of the proceedings as malicious or groundless does not define the ambit of JPI; malice is not an ingredient of a s.48(1A) ERA claim (para 53).
  • Any examination of the content of the arbitration documents would be limited to establishing:
    • that the proceedings were causally grounded in the protected disclosures; and
    • the factual matrix of detriment;
    rather than to re‑litigate the substantive merits of the breach of confidence claim (para 50, 54).

The Court therefore isolates the constitutive act of detriment (commencing proceedings) from the evidential use of statements made in those proceedings.

3.5.3 No policy justification for immunity in this context (para 54–55)

Andrews LJ then measures the Respondent’s claimed immunity against the recognised policy rationales of JPI:

  • Freedom of expression in judicial proceedings?
    The s.48(1A) claim does not chill what advocates, witnesses or parties say in court or arbitration. It targets the decision to deploy proceedings as a retaliatory measure. The ET will not be “re‑trying” the truth of evidence given in the arbitration; it will assess:
    • whether the disclosures were protected;
    • whether the arbitration was brought on the ground of those disclosures; and
    • whether that caused detriment.
    JPI’s speech‑protecting rationale is not engaged.
  • Avoidance of multiplicity of actions?
    The ET will not sit in appeal on any arbitral determination. Rather, it will apply a wholly different statutory test (protected disclosure and detriment) based on its own evidence. There is no impermissible collateral attack on judicial findings (para 54).
  • Public policy in ERA positively favours allowing the claim.
    Parliament has:
    • created a right not to suffer detriment for protected disclosures (s.47B);
    • provided a tribunal remedy (s.48(1A)); and
    • voided contractual gagging of disclosures (s.43J).
    To superimpose a judge‑made immunity blocking claims where an employer uses legal proceedings themselves as the instrument of retaliation would, in Andrews LJ’s words, “seriously undermine the protection that Parliament intended to apply” and leave a statutory wrong without a remedy (para 55).

Both Elisabeth Laing LJ and Males LJ expressly endorse this statutory‑policy‑driven conclusion (paras 57–58).


4. Precedents and Authorities: Influence on the Court’s Reasoning

4.1 Lincoln v Daniels [1958] 1 QB 237

Lincoln provided the conceptual starting point for the Respondent’s argument, via Devlin LJ’s three categories of absolute privilege. However:

  • Devlin LJ was concerned with whether communications to the Bar Council were steps in a disciplinary process, not with whether bringing proceedings could itself be a tort.
  • His reference to immunity for “everything that is done from the inception of the proceedings onwards” was not intended as a literal, exhaustive enumeration of all litigation acts (para 29).

Andrews LJ thus treats Lincoln as contextually limited and subordinated to later authorities (esp. Darker, Taylor, Singh, XGY) that refine the scope of JPI.

4.2 Willers v Joyce [2016] UKSC 43; [2018] AC 779

Willers is cited (para 25) as the landmark that confirmed malicious prosecution can extend to civil proceedings. Its relevance is twofold:

  • It dispels any notion that instituting civil proceedings is a constitutionally immune act.
  • It illustrates the judiciary’s willingness to recognise civil liability for abusive litigation conduct.

In tandem with Roy v Prior [1971] AC 470 and Lord Hoffmann’s analysis in Taylor v SFO, it supports the proposition that JPI does not cloak the initiation of proceedings where a recognised cause of action (malicious prosecution/abuse of process) can be pleaded.

4.3 Taylor v Serious Fraud Office [1999] 2 AC 177

Taylor is important primarily through Lord Hoffmann’s dicta, emphasising that:

  • the core immunity is “limited to actions in which the alleged statement constitutes the cause of action” (quoted in XGY and by the present Court: para 36–37);
  • immunity cannot be tested case‑by‑case ex post without undermining its protective function (reaffirmed in XGY at [16]).

Those points underpin Andrews LJ’s insistence that:

  • the cause of action here is not a statement‑based tort; and
  • JPI should not be retro‑fitted to insulate the Respondent from a statutory whistleblowing claim.

4.4 Darker v Chief Constable of the West Midlands [2001] 1 AC 435

Darker refined the ambit of JPI by:

  • construing historical statements about “anything said or done” as referring to acts of communication (e.g. submission of written statements), not every act connected with litigation (para 30); and
  • demonstrating that fabricating evidence may stand outside immunity, depending on its stage and purpose.

This authority supports the Court’s narrow, communication‑focused reading of “done” in the older cases and thus its refusal to extend JPI to the act of initiating proceedings.

4.5 Singh v Reading Borough Council [2013] EWCA Civ 909

As discussed above, Singh is a key stepping stone:

  • It articulates the twin rationales for JPI: free speech in proceedings and avoidance of multiplicity of litigation (para 32).
  • It holds that pressuring a witness to give inaccurate evidence in a statement prepared for proceedings was not protected by JPI.
  • It explicitly disavows the idea that immunity attaches to “anything said or done … whatever the nature of the claim” (para 33).

Singh thus lays the groundwork for a disciplined, policy‑sensitive rather than literal approach to immunity.

4.6 Chief Constable of Sussex Police v XGY [2025] EWCA Civ 1230

XGY is arguably the direct doctrinal precursor to Rogerson:

  • It affirms a robust core immunity for statements in proceedings, applying to all causes of action, including non‑defamation claims like breach of confidence and data protection.
  • It adopts a non‑fact‑specific approach to JPI – either the category of conduct is within or outside immunity; it cannot depend on case‑by‑case factual nuance (para 35, 48).
  • It stresses that extensions beyond the core must be justified as “necessary for the proper administration of justice” and are strictly limited (para 41–42).

Rogerson faithfully applies these principles but reaches the opposite conclusion on immunity because the category of conduct is fundamentally different: initiating proceedings (a non‑speech act) rather than making statements in or for proceedings.

4.7 Other authorities

The Court also references:

  • Munster v Lamb (1883) 11 QBD 588: classic statement of the rationale for advocate immunity – to enable fearless advocacy and to avoid chilling litigation by fear of suit (para 38).
  • Arthur JS Hall & Co v Simons [2002] 1 AC 615: abolishing barristers’ immunity from negligence suits in civil and criminal advocacy, while preserving the separate speech‑based JPI – illustrating the distinction between negligence liability and JPI (para 39, 45).
  • Jones v Kaney [2011] UKSC 13; [2011] 2 AC 398: abolishing expert witness immunity from negligence claims; again, it emphasises balancing “every wrong should have a remedy” against the needs of justice (para 39).
  • The Chief Constable of South Wales Police v Daniels [2015] EWCA Civ 680: showing that some public‑law torts (e.g. misfeasance relating to disclosure) can fall outside JPI (referenced in XGY and noted in para 36, 41).

5. Complex Concepts Simplified

5.1 “Protected disclosure” and “whistleblower”

  • A “protected disclosure” is a specific kind of disclosure of information, usually about wrongdoing (e.g. criminal offences, breaches of legal obligations, health and safety dangers), made by a worker in accordance with the detailed rules in ss.43A–43H ERA.
  • A “whistleblower” is someone who makes such a disclosure.
  • If the statutory conditions are met, the disclosure is “protected”, meaning the worker has enhanced protection against detriment and (for employees) unfair dismissal.

5.2 “Worker” vs “employee”

  • “Worker” is a broader category than “employee”. It includes individuals who personally perform work, even if they are not employees in the strict contractual sense.
  • The judgment notes that:
    • an employee dismissed for whistleblowing claims under s.103A ERA (automatic unfair dismissal);
    • a “worker” who is not an employee, but whose engagement is ended because of whistleblowing, must use the detriment route under s.48(1A) ERA (note 1).

5.3 “Detriment”

  • Simply means “harm” or “damage”.
  • Examples include:
    • disciplinary action, demotion, or denial of promotion;
    • financial penalties or adverse changes to terms;
    • harassment, ostracism, or reputational harm;
    • being sued or threatened with proceedings designed to punish disclosures.

5.4 Judicial proceedings immunity (JPI)

  • A common‑law rule that certain participants in court (judges, lawyers, parties, witnesses, jurors) are immune from being sued over what they say in the course of proceedings.
  • Originally developed in defamation law (absolute privilege) but now applied to a broader range of claims.
  • Core rationale:
    • allow participants to speak freely without fear of later litigation; and
    • avoid endless collateral actions about the truth of evidence.
  • Not a blanket shield for all litigation conduct; focused on statements that form the foundation of the cause of action.

5.5 Malicious prosecution and abuse of process

  • Malicious prosecution:
    • a tort where someone maliciously brings proceedings without reasonable and probable cause, causing damage; historically criminal, now includes civil proceedings (Willers);
    • JPI does not protect the malicious initiation of proceedings in this context.
  • Abuse of process:
    • a tort where proceedings are used for a collateral or improper purpose, rather than to obtain genuine relief;
    • shows that the law recognises wrongful use of litigation as actionable in its own right.

5.6 New York Convention and Arbitration Act 1996, s.103(3)

  • The New York Convention requires courts of contracting states (including the UK) to recognise and enforce foreign arbitral awards, subject to limited defences.
  • Section 103(3) Arbitration Act 1996 allows UK courts to refuse enforcement if it would be contrary to UK public policy.
  • In this case, UK public policy (s.43J ERA) treats clauses preventing protected disclosures as void; therefore, awards enforcing such clauses will not be enforced in the UK.

6. Impact and Significance

6.1 Strengthening whistleblower protection against retaliatory litigation

The judgment makes clear that employers cannot use litigation or arbitration as a de facto instrument of retaliation against whistleblowers and then shield themselves behind JPI. Key practical effects:

  • Threats vs actions: There is no irrational gap whereby threatening proceedings is actionable detriment, but actually issuing proceedings is not (para 13).
  • Foreign proceedings: Workers remain protected under ERA even if the employer chooses a foreign arbitral forum and foreign governing law, at least so far as UK‑based enforcement and UK tribunal remedies are concerned (paras 16–17).
  • Chilling effect reduced: Knowing that an employer cannot immunise itself from ERA liability by suing in arbitration may embolden workers to report wrongdoing.

6.2 Limits on the strategic use of confidentiality clauses and arbitration clauses

The decision reinforces and develops the law’s scepticism about “gagging” provisions:

  • Confidentiality clauses remain enforceable for legitimate purposes (trade secrets, commercial information), but are void insofar as they restrain protected disclosures (s.43J; para 15).
  • Arbitration clauses cannot be used as a shield against UK public policy:
    • the seat of arbitration (here, Singapore) and governing law are irrelevant to UK enforcement if the award violates whistleblowing policy;
    • ETs may treat the act of commencing such proceedings itself as a whistleblowing detriment.

For employers and their lawyers, this decision signals that:

  • drafting of NDAs and arbitration clauses must factor in the non‑excludable regime of ERA;
  • using proceedings to punish or deter protected disclosures carries real statutory risk in the UK.

6.3 Clarifying and containing judicial proceedings immunity

Doctrinally, Rogerson continues a modern trend of:

  • Protecting the core of JPI (statements in and for proceedings) but
  • Resisting its expansion into a more general shield for litigation conduct.

The case confirms:

  • JPI is not an immunity for the mere act of bringing civil or arbitral proceedings;
  • attempts to analogise initiation of proceedings with speech‑based privilege will fail absent a clear necessity rooted in the administration of justice;
  • statutory rights, especially those with strong public‑policy underpinnings (like whistleblowing protection), will not lightly be subordinated to judge‑made immunities.

6.4 Interplay between domestic employment law and international arbitration

The judgment is also important for the interface between:

  • international arbitration (ICC, foreign seats, foreign governing law); and
  • domestic employment and public policy regimes.

Core implications:

  • UK tribunals are competent to treat foreign arbitration as a potential “detriment” and to award remedies under ERA.
  • UK courts may refuse to enforce arbitral awards that penalise protected disclosures, under the public policy exception in s.103(3) Arbitration Act 1996.
  • Employers with cross‑border structures or charities based abroad remain subject to UK whistleblowing norms in relation to UK workers.

7. Conclusion

Rogerson v Erhard‑Jensen Ontological/phenomenological Initiative Ltd makes a clear and significant contribution to UK employment and procedural law.

Substantively, it holds that:

  • the initiation of court or arbitral proceedings can constitute an actionable detriment under s.47B ERA where motivated by a protected disclosure; and
  • judicial proceedings immunity does not apply to that act of initiation, even where the proceedings are quasi‑judicial and even though they are commenced by a written document.

Conceptually, the case:

  • reaffirms that JPI is a targeted, speech‑centred immunity, not a catch‑all shield for all litigation conduct;
  • insists that extensions to JPI must be tightly linked to the administration of justice; and
  • refuses to deploy JPI as a device that would negate Parliament’s deliberate design of whistleblower protections in ERA.

Practically, the decision:

  • bolsters whistleblower protection by recognising retaliatory litigation and arbitration as a potential detriment;
  • curtails the strategic use of confidentiality and arbitration clauses to suppress protected disclosures; and
  • clarifies that UK public policy on whistleblowing can override foreign arbitral awards when they conflict with s.43J ERA.

In combination, these elements give Rogerson lasting importance as an authority on the limits of judicial proceedings immunity and the robustness of UK whistleblowing protections, especially in an increasingly international and arbitration‑heavy employment landscape.

Case Details

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

Comments