Daniel Rogerson v Erhard-Jensen Ontological/Phenomenological Initiative Limited

The Employment Rights Act 1996 (“ERA”) confers certain rights upon workers who make protected disclosures (commonly referred to as “whistleblowers”). Section 47B(1) ERA provides that:

“A worker (“W”) has the right not to be subjected to any detriment by any act, or any deliberate failure to act, by his employer done on the ground that the worker has made a protected disclosure”.

Section 48(1A) ERA provides that:

“A worker may present a complaint to an employment tribunal that he has been subjected to a detriment in contravention of section 45A.”

On 15 November 2021 the claimant, Daniel Rogerson, began proceedings under section 48(1A) ERA in the Employment Tribunal (“the ET”) against two respondents, namely, the Respondent to this appeal (a Singapore-registered charity), and a Mr Werner Erhard, in whose London residence Mr Rogerson formerly worked, for post-employment whistleblowing detriment. After his resignation in 2019, Mr Rogerson made allegations that Mr Erhard had regularly been verbally and physically abusive towards staff, which he says were protected disclosures. Those allegations are denied. The claim against Mr Erhard was dismissed by the ET in a reserved judgment dated 19 April 2023, on the grounds that it was brought out of time. The Respondent’s case, in a nutshell, is that Mr Rogerson made the allegations not for altruistic reasons but in the hope of financial gain for himself and/or others.

It is common ground that Mr Rogerson falls within the statutory definition of a “worker” and that in due course it will be for the ET to determine whether the disclosures he made were protected disclosures. For the purposes of this appeal, which is against the striking out of part of the claim, the claimant’s case must be taken at its highest, and therefore it must be assumed that they were.

In paragraph 16 of his Claim Statement in the ET, Mr Rogerson set out a number of detriments to which he claims he was subjected by conduct on the part of the Respondent, on the ground that he had made the protected disclosures. The acts he complains of include formal communications from lawyers threatening legal proceedings against him. The Respondent accepts that this conduct can give rise to an actionable detriment.

At sub-paragraph 16 (g) Mr Rogerson identified as a detriment (although it might more accurately be described as an act giving rise to a detriment) the commencement by the Respondent of arbitration proceedings against him in Singapore in July 2021, in which it was claimed that he had breached a Confidentiality and Independent Consulting Agreement (“the Agreement”) which he had entered into with it on 26 April 2016. The Agreement contained an arbitration clause in favour of an arbitration seated in Singapore, to be conducted pursuant to the Rules of Arbitration of the International Chamber of Commerce (“ICC”).

Mr Rogerson set out in paragraph 16 (g) the relief that the Respondent was claiming in the arbitration, which included liquidated damages of £50,000 and an injunction to restrain him from contacting the Respondent’s and its related entities’ past and present staff. The Respondent also sought costs orders against him. Mr Rogerson stated that in the arbitration claim the Respondent had accused him of running an “extortion scheme” by “making false claims” which “included various allegations of physical and verbal abuse … by Mr Erhard… in efforts to extract a settlement.” That is clearly a reference to the protected disclosures.

The issue which arises on this appeal is whether that limb of Mr Rogerson’s claim (referred to in the proceedings below as “Detriment 3”) should be struck out on the basis that the claimed detriment is not actionable. Judge Fowell in the ET refused to strike out Detriment 3. On appeal by the Respondent to the Employment Appeal Tribunal, Heather Williams J held that the ET had made a material error of law. She accepted the Respondent’s submission that Detriment 3 “falls within the core ambit of Judicial Proceedings Immunity” (“JPI”). Mr Rogerson appeals to this Court with the permission of Elisabeth Laing LJ.

For the reasons set out in this judgment I would allow this appeal and restore the decision of Judge Fowell.

Public policy as manifested in the ERA is that workers should be protected in the workplace. This in turn requires that whistleblowers who suffer detriment at the hands of an employer on the grounds of making or seeking to make protected disclosures should have a remedy. “Detriment” is not defined in the ERA, and therefore must be given its ordinary meaning of harm or damage.

Section 47B(1) ERA confers a right on a worker not to be subjected to any detriment by any act (or any deliberate failure to act) by his employer on the ground that he has made a protected disclosure (my emphasis). If Parliament had intended to carve out an exception, I would have expected it to say so in terms. On the contrary, the nature of the detriment is expressed in unqualified terms. So too is the nature of the act which causes it. The only qualification that appears on the face of the section is that an employer’s failure to act must be deliberate. On the face of it, “any act” is wide enough to cover the initiation of legal or arbitral proceedings.

On any ordinary understanding of the word “detriment,” the commencement of an arbitration in Singapore claiming that Mr Rogerson’s protected disclosures were a breach of confidence was (or, at least for the purposes of a strike-out application, can be taken to have been) an act which subjected him to a detriment. He was put to the trouble and expense, not to mention the stress, of defending arbitration proceedings in another jurisdiction which is a 12- or 13-hour flight away from the UK. Indeed, apart from dismissal, (Footnote: ¹) it is difficult to conceive of an act which is more likely to subject a whistleblower to detriment than the commencement of legal or arbitral proceedings against him or her which are designed to obtain financial recompense for previous protected disclosures and/or to prevent future protected disclosures. Although Mr Kemp KC (who appeared with Mr Mathur for the Respondent) wished to preserve his client’s position, and therefore made no concession when this point was put to him by the court, he was unable to suggest any argument to the contrary.

Moreover, it is difficult to see how the Respondent could resist the submission that the commencement of the arbitration was “on the ground of” the protected disclosures (another point that Mr Kemp reserved the right to dispute in due course, should we allow this appeal). Thus, at least on the face of it, Detriment 3 as pleaded satisfies all the requirements of section 47B(1) ERA.

Especially when viewed in the context of the aforementioned policy of the ERA, the proposition that a worker who is threatened with legal proceedings (e.g. for breach of confidence) because they have made a protected disclosure, can bring a claim in the ET, but if the threat is actually carried out they cannot, and therefore have no remedy, has no rational justification. A disclosure of the relevant type which leaves a whistleblower vulnerable to civil proceedings (whether judicial or arbitral, and whether at home or abroad) cannot be described as “protected” in any meaningful sense.

Therefore, on the face of it, Detriment 3 should be actionable. There is nothing in ERA itself to suggest that the commencement of such proceedings should attract immunity from suit. There are, moreover, powerful contra-indications to immunity besides the matters I have already mentioned at [9] to [11] and [13] above.

Judge Fowell made reference in his judgment to section 43J of ERA. This provides that:

“(1)

Any provision in an agreement to which this section applies is void in so far as it purports to preclude the worker from making a protected disclosure.

(2)

This section applies to any agreement between a worker and his employer (whether a worker’s contract or not), including an agreement to refrain from or continuing any proceedings under this Act or any proceedings for breach of contract.”

This would be apt to cover a confidentiality clause to the extent that it could be used to preclude a worker from making a protected disclosure (or to give the employer a remedy in damages in the event of a protected disclosure being made) although, as Judge Fowell rightly acknowledged, the statute does not say that such a clause would be void for all purposes. The employer might still be able to enforce the clause to prevent the disclosure of other confidential information such as trade secrets, for example.

I agree with Judge Fowell that because of section 43J ERA, if the Respondent’s claim for breach of confidence had been brought in the UK (that is, if there had been no arbitration clause in the Agreement) it would not have been effective to prevent any protected disclosure. However, I would go further than that. Any Singapore arbitral award which treated the confidentiality provision as valid and effective to preclude Mr Rogerson from making a protected disclosure would not be enforceable in this jurisdiction under the New York Convention (as given effect domestically by virtue of sections 100 to 104 of the Arbitration Act 1996). That would be the case irrespective of whether (as we are told it is) the Agreement is governed by the law of Singapore and not the law of England and Wales.

Under section 103(3) of the 1996 Act, recognition or enforcement of an award may be refused if it would be contrary to public policy to do so. Public policy, as clearly evidenced by section 43J of ERA, is that a contractual provision which has the effect of precluding a worker from making a protected disclosure is to be treated as void. It follows that an award of damages based upon such a provision (or which seeks to enforce it by granting injunctive relief) would not be recognised.

As Judge Fowell pointed out at [73] of his decision:

“it would be very difficult to distinguish between an argument that Mr Rogerson could not pursue his claim because of the confidentiality agreement, and an argument that he could not pursue his claim because the Respondent had commenced proceedings to enforce that confidentiality agreement (the basis of the claimed immunity).”

The effectiveness of Parliament’s intended protection for whistleblowers should not depend on such nice distinctions.

That brings me to the scope and operation of the principle of JPI. Judge Fowell found, rightly (and there has been no challenge to that finding) that the Singapore arbitration proceedings were quasi-judicial in nature. Thus JPI is applicable in principle. Two questions arise, namely: (1) to what conduct does the immunity apply? and (2) from what is the person concerned immune? In order to answer those questions it is necessary to identify the rationale behind the grant of the immunity.

We were shown a plethora of authorities but it is unnecessary to go through them all, not least because Mr Kemp did not seek to rely upon most of them and, indeed, largely sought to distinguish them. He submitted that the vast majority of the cases related to the immunity of witnesses from suit for what they said in evidence (mainly, though not exclusively, immunity from actions for defamation). Some cases addressed the extent to which that immunity extended to actions which occurred outside the court proceedings themselves, usually prior to the initiation of such proceedings. On his analysis, none of these cases was directly in point.

Mr Kemp did place considerable reliance, however, on Lincoln v Daniels [1958] 1 QB 237. In that case, the issue was whether, in a libel action, absolute privilege attached to certain communications sent by the defendant to the Secretary of the Bar Council alleging professional misconduct by the plaintiff, a Queen’s Counsel. The defendant contended that the documents constituted a step in a (disciplinary) inquiry before the plaintiff’s Inn of Court, to which absolute privilege attached. The Court of Appeal agreed that a step taken in such an inquiry would be subject to absolute privilege because it was a judicial process recognised by law, but held that the communications in question did not constitute such a step.

Devlin LJ, at page 257, said this:

“The absolute privilege which covers proceedings in or before a court of justice can be divided into three categories. The first category covers all matters that are done coram judice. This extends to everything that is said in the course of proceedings by judges, parties, counsel and witnesses, and includes the contents of documents put in as evidence. The second covers everything that is done from the inception of the proceedings onwards and extends to all pleadings and other documents brought into existence for the purpose of the proceedings and starting with the writ or other document which initiates the proceedings. The third category is the most difficult of the three to define. It is based on the authority of Watson v M’Ewan, in which the House of Lords held that the privilege attaching to evidence which a witness gave coram judice extended to the precognition or proof of that evidence taken by a solicitor. It is immaterial whether the proof is or is not taken in the course of proceedings. In Beresford v White the privilege was held to attach to what was said in the course of an interview by a solicitor with a person who might or might not be in a position to be a witness on behalf of his client in contemplated proceedings.”

Mr Kemp submitted that nearly all the authorities about JPI were concerned with the third of Devlin LJ’s categories, which is concerned with extensions to the core immunity, whereas this case was concerned with the second, namely: “everything that is done from the inception of the proceedings onwards…” (emphasis added). Devlin LJ recognised that JPI attaches not only to judges, counsel and witnesses but to the parties to the litigation themselves. His second category covers the writ or other document which initiates the proceedings. In this case that would be the request for arbitration. The commencement of the arbitration by sending the request for arbitration to the ICC therefore fell squarely within the “core immunity” and there was no question of seeking to extend the ambit of the protection it afforded. Cases concerning extensions to the core immunity were irrelevant.

Mr Kemp further submitted that the rationale for the immunity afforded to parties to the litigation (or arbitration) is different from the rationale for the immunity afforded to witnesses that has been articulated in many cases, namely, that they should be encouraged to give their evidence freely, without fear of being sued for what they say. He argued that parties should be free to initiate proceedings without the fear of being sued for doing so. That is a novel proposition, which derives no support from the authorities, and indeed appears to me to be contradicted by them. I also consider that it is incorrect as a matter of principle.

First, a principle of immunity from suit self-evidently only comes into play in a scenario where the behaviour complained of would otherwise be potentially actionable. Until relatively recently, there was hardly any question of anyone fearing that they might be sued for bringing civil proceedings, even if it could be shown that they had brought them maliciously. The tort of malicious prosecution was believed to apply only when the complainant (and putative claimant) had been prosecuted in a criminal case. It was only in 2016, in Willers v Joyce [2016] UKSC 43; [2018] AC 779, that the Supreme Court established that the tort of malicious prosecution does include the prosecution of civil proceedings.

Although a different tort, abuse of process, can in principle apply to civil proceedings, in order to get such a claim off the ground the claimant would have to be able to establish a sufficiently arguable case that the defendant had used those proceedings as a device to secure an entirely extraneous objective rather than genuinely seeking the relief claimed.

In both those torts, where malice or some ulterior motivation is a key ingredient of the cause of action, JPI does not attach to the initiation of the legal proceedings by the alleged tortfeasor (provided there is sufficient evidence that that ingredient is present to enable the claimant to plead it): see Roy v Prior [1971] AC 470 and the observations of Lord Hoffmann in Taylor v Serious Fraud Office [1999] 2 AC 177 at 215E-G. On the other hand, if the necessary malice or ulterior motive is absent, the person who brought the civil proceedings does not need the protection of the principle of JPI, for the simple reason that the common law would not recognise their behaviour in bringing the proceedings as an actionable wrong.

Secondly, there is no blanket rule that the initiation of civil proceedings can never be a civil wrong. The torts of malicious prosecution and abuse of process illustrate this, but so too does the availability of anti-suit injunctions to restrain the commencement of proceedings in breach of agreements to arbitrate or to litigate disputes in a particular forum.

Thirdly, the ambit of JPI is not as wide as the language of “anything said or done”, which is used in the older authorities, suggests. This emerges from the modern authorities which explain the ambit of the core immunity and the rationale underpinning it. In Lincoln v Daniels, Devlin LJ was not addressing the question whether immunity should attach to the act of bringing proceedings. His reference to category 2 encompassing “everything that is done from the inception of the proceedings” is not to be taken literally.

For example, in Darker v Chief Constable of the West Midlands [2001] 1 AC 435, Lord Hutton considered a similar statement by Kelly CB in Dawkins v Lord Rokeby (1873) LR 8 QB 255 at 264, namely:

“no action lies against parties or witnesses for anything said or done, although falsely and maliciously and without any reasonable or probable cause, in the ordinary course of any proceeding in a court of justice.”

Having pointed out that in that case the claim related not only to the defendant’s oral evidence before a military court of inquiry, but to a written statement containing in substance a repetition of that evidence, which was handed to the court after the evidence was given, Lord Hutton said that the reference to “anything… done” was probably intended to cover the submission of a written statement to a court.

In order to ascertain what conduct falls within the “core immunity” and the rationale underlying the principle, I found two decisions of this Court particularly helpful: Singh v Reading Borough Council [2013] EWCA Civ 909; [2013] 1 WLR 3052, (“Singh”) and Chief Constable of Sussex Police and another v XGY [2025] EWCA Civ 1230 (“XGY”). Neither was a defamation case, and the target of the claim said to be subject to JPI was not a witness in the proceedings. In one case the claim for JPI failed; in the other, it succeeded.

In Singh the issue was whether an allegation that pressure had been put on a witness to give untruthful or inaccurate evidence in a witness statement prepared for use in proceedings before an employment tribunal fell within the scope of JPI.Lewison LJ, with whom Maurice Kay and Gloster LJJ agreed, held that it did not. At [1] he defined the principle of JPI as “the principle that those involved in the judicial process should be immune from civil suit for what they do or say in the course of the litigation.” At [23] Lewison LJ identified two strands of policy as underlying the principle; the first being that those engaged in litigation should be able to speak freely without fear of civil liability, the second being a wish to avoid a multiplicity of actions where one court would have to examine whether evidence given before another court was true or not. He cited several authorities for those propositions, culminating in Darker.

At [61] Lewison LJ said that it could no longer be said that immunity from civil suit “attaches to anything said or done by anybody in the course of judicial proceedings whatever the nature of the claim made.”

XGY concerned a claim for damages for, inter alia, breach of confidence, breaches of the Human Rights Act 1998 and breaches of the Data Protection Act 2018, which was brought against the Crown Prosecution Service and the Chief Constable of Sussex Police following the revelation of a vulnerable witness’s current address by a CPS advocate in the course of the hearing of a bail application. The address had been provided by the witness to the police in confidence and passed by them to the CPS as part of the police’s file for preparation for the bail hearing, without any indication that it was confidential. The issue before the Court of Appeal was whether the core immunity for things said and done by advocates in court applied, and in particular whether it extended to the actions of the police in supplying information to the CPS. The Court of Appeal answered both these questions in the affirmative.

The constitution in that case comprised Lady Carr LCJ, Dame Victoria Sharp P. and Coulson LJ. The appellant submitted that the approach of the judge in the lower court, which had been to take a “fact-specific” approach to whether the core immunity applied, on a case by case basis, cut directly across the policy rationale underlying the immunity. The Court of Appeal accepted that submission [65]. That meant that the rationale for the immunity lay at the heart of the decision.

The court first identified what earlier case law had defined as the core immunity at paragraphs [8] and [9]:

“[8] Since at least the sixteenth century it has been recognised by the common law that it is necessary for the administration of justice for certain participants in the administration of justice – advocates, parties, witnesses, judges and jurors – to be immune from suit, i.e. that legal claims against them arising from almost anything done or omitted to be done in the course of conducting a case in court are barred (see Darker v Chief Constable of the West Midlands [2001] 1 AC 435 per Lord Hutton at 463). This has been described as the core immunity.

[9] The core immunity is “limited to actions in which the alleged statement constitutes the cause of action”, see Taylor v Director of the Serious Fraud Office [1999] 2 AC 177 per Lord Hoffmann at p.215C.”

In a footnote to paragraph [8] they added:

“Nothing turns in this appeal on the various exceptions to this for claims of perjury, malicious prosecution and analogous claims involving malicious initiation of criminal proceedings and proceedings for contempt of court, see for example The Chief Constable of South Wales Police v Daniels [2015] EWCA Civ 680 per Lloyd Jones LJ at [32].”

In paragraph [10] the court explained that whilst the early cases concerned actions for defamation brought against advocates and witnesses,

“In order to prevent the immunity being circumvented or outflanked, however, for many years the immunity has been held to apply to whatever claim was being made against the advocate or witness for things done or spoken in court and however ingeniously such a claim was framed.”

They went on to give examples of cases in which the immunity had been held to bar claims for conspiracy to give false evidence, conspiracy to mislead the court or conspiracy to present false information, and a claim under the Protection from Harassment Act 1997.

At paragraphs [12] and [13] the court went on to consider the rationale for the core immunity. They referred to passages in Munster v Lamb (1883) 11 QBD 588; (1882) 5 LJ QB 46 in which Brett CJ identified the rationale as being that advocates must be able to speak freely in order to properly perform their duties to the court and their client, and Fry LJ explained that:

“…it is not the desire to prevent actions being brought in cases where they ought to be maintained that has led to the adoption of the present rule of law; but it is the fear that if the rule were otherwise, numerous actions would be brought against persons who were merely discharging their duty.”

At [14] the court went on to consider more recent authorities at the highest level:

“More recently, the House of Lords and the Supreme Court have emphasised that the scope of the immunity involves a balance between different public interests. The first is that every wrong should have a remedy. The second is the encouragement of freedom of speech and communication in court proceedings, which is necessary to protect the proper administration of justice and the interests of justice, see Taylor at 208; Arthur JS Hall & Co v Simons [2002] 1 AC 615 at 679 and Jones v Kaney [2011] UKSC 13; [2011] 2 AC 398 at [55] –[57], [67], [85]- [86] [87] and [108]–[114].”

Having pointed out at [16] that the effectiveness of the immunity would be undermined if the protection afforded had to be tested after the event in every case, quoting from Lord Hoffmann in Taylor at 214 E-G, the court then went on to consider cases in which the immunity applied “beyond the narrow limits of what is said or done in court” to any matter “so intimately connected with the conduct of the cause in court that it can fairly be said to be a preliminary decision affecting the way that cause is to be conducted when it comes to a hearing.”

The test in such cases was whether an extension to the immunity was necessary for the administration of justice; this could be gauged by reference to established exceptions. Examples of where it had been applied were to witness statements made outside court with a view to the maker actually (or potentially) giving evidence in court, information transmitted by the police to the CPS for the purposes of enabling a witness to give evidence in court, and statements by investigators which can fairly be said to be part of the process of investigating crime or a possible crime. On the other hand, the immunity was held not to attach to claims for conspiracy to fabricate evidence (Darker), a claim that a witness had been pressured to include inaccurate information in her witness statement (Singh) and a claim for misfeasance in public office based not on a statement, but upon the manner in which a disclosure exercise had been performed (Daniels).

The court summarised the principles to be extrapolated from the cases at paragraph [30]. The first two paragraphs of that summary are directly pertinent to this appeal. They are worth quoting in full:

“i)

It is a general principle that every wrong should have a remedy. Nevertheless, it is necessary for the proper administration of justice that advocates, parties, witnesses, judges and jurors are immune from suit for statements made in court whatever the cause of action, regardless of whether the statement was made maliciously or was irrelevant to the court proceedings. This is known as the core immunity. It is founded on public policy and is intended to encourage freedom of expression and communication in court proceedings in order to protect the proper administration of justice and the interests of justice.

ii)

The core immunity can be extended if the extension is necessary for the proper administration of justice, which is a strict test. There are two established extensions. witnesses and potential witnesses are immune from suit for statements made outside of court with a view to giving evidence. This extends to the preparation for evidence they are likely to give in court proceedings including their preliminary examination to ascertain what they could prove. And investigators are immune from suit for statements made as part of the process of investigation.”

It is clear from XGY, in which the Court of Appeal followed and applied Lord Hoffmann’s analysis in Taylor, that the core immunity attaches to the making of statements, and that the key rationale underlying it is to encourage freedom of speech and communication by all those involved in the litigation process which is necessary for the proper interests of justice. The court made it clear at [65] (ii) that the core immunity is not limited to evidential matters but attaches to all statements, said or written, made in court or in the course of the litigation from the inception of the proceedings onwards.

The court in XGY said nothing about the second rationale for the principle that was identified by Lewison LJ in Singh, namely the avoidance of multiplicity of litigation, possibly because XGY was not concerned with a cause of action which would have involved examining the truth or falsity of evidence given in court.

However, it is clear from XGY and Singh that the core immunity does not cover all things done (or omitted to be done) by the parties, advocates, or witnesses in the course of litigation, as demonstrated by the analysis at [26] to [29] of XGY of how the tort of negligence is impacted by the principle of JPI. That passage includes a reminder that the justification unsuccessfully put forward in Hall v Simons for maintaining an advocate’s immunity for suit for negligence in respect of conduct of a case in court was not that this conduct was covered by the core immunity, but rather that an analogy could be drawn with it.

It follows that where the immunity attaches to something that is “done” in the course of legal proceedings (whether prior to or at the hearing) the act in question must relate in some way to the statement that is the foundation of the cause of action, whatever that cause of action may be. It would encompass, for example, the promulgation of a defamatory statement in pleadings or witness statements; or making reference to confidential information in written or oral submissions, or the act of providing information for use in proceedings in court.

In the present case, the cause of action is not founded on statements made in the request for arbitration or in the arbitration itself. The act complained of by Mr Rogerson is the act of initiating the arbitral proceedings. That is what causes the detriment. That act does not fall within the core immunity, any more than the act of initiating proceedings in breach of an agreement to arbitrate, or in breach of an exclusive jurisdiction clause, would fall within it. Whilst it is true that the arbitral proceedings are commenced by the sending of a document, namely, a request for arbitration, statements made in that document are not the acts which give rise to the detriment complained of.

Mr Kemp sought to argue that the particulars of Detriment 3 pleaded by Mr Rogerson demonstrated that on proper analysis his case went wider than a complaint about the commencement of the arbitration. But as XGY establishes, whether JPI applies cannot depend on the facts and circumstances of the individual case. Therefore it cannot depend on the way in which the case has been pleaded. The way in which a claimant chooses to characterise the proceedings (malicious, groundless, etc.) cannot make a difference to the answer to the question whether JPI does or does not apply to the bringing of those proceedings.

Heather Williams J attached considerable importance to the way in which the case had been pleaded; she said the pleaded detriment was not simply the commencement of the arbitration but that the Respondent initiated a groundless arbitration process that was based on false allegations. She also said that it was completely artificial to separate out the bringing of the arbitration proceedings from the basis on which they were brought. The nature and content of the proceedings was what Mr Rogerson relied upon as constituting the detriment.

Whilst it is true that it would have to be established that the proceedings were brought “on the ground of” the protected disclosures, and that inquiry might involve the ET looking at the content of the request for arbitration or statements of case, the complaint is about the proceedings themselves, and their impact upon Mr Rogerson. Statements made in the proceedings, whether in the request for arbitration or in the statements of case, do no more than evidence the link between the act complained of (commencing the arbitration) and the protected disclosures.

A request for arbitration generally contains basic information about the identities and contact details of the parties and their legal representatives, and sufficient information to satisfy the person or body to whom the request is made (here, the ICC) that there is a binding written and signed agreement to arbitrate the dispute – generally, a quotation from the arbitration clause, a copy of which may be annexed to the request. It will also identify the number of arbitrators, how they are to be appointed and by whom, and where the seat of arbitration is to be. The information about the dispute will often be no more detailed than a broad indication of its general nature.

The court asked the parties if they were willing to let us see the request for arbitration in this case. Mr Polak, who represented Mr Rogerson on this appeal told us that his client was content that we should see it, but Mr Kemp’s instructions were that the Respondent did not consent. In those circumstances the court could not assume that there was anything in the content of that document that could be regarded as essential to, or even advancing Mr Rogerson’s claim under s.48(1A) ERA.

Likewise, the answer to the question whether JPI covers this situation cannot depend on whether it is part of Mr Rogerson’s case that the arbitration was commenced maliciously or in bad faith; malice is not an ingredient of the cause of action under s.48(1A) ERA, as Judge Fowell rightly pointed out.

In my judgment none of the justifications for the principle of JPI articulated in the authorities would apply to the act of commencement of an arbitration aimed at stifling protected disclosures or penalising someone for making them. The ET would not be involved in re-litigating the rights or wrongs of the action for breach of confidence. The ET judge hearing Mr Rogerson’s claim would have to make up his or her own mind, on the evidence adduced by the parties in the proceedings before the ET, as to whether the allegations were true and whether the disclosures were protected. If the judge was satisfied on both those matters, they would then go on to ask whether the act of commencing the arbitration had been carried out “on the ground of” the protected disclosures and whether the claimant had suffered a detriment in consequence.

I can see no reason why it could possibly be regarded as essential to the administration of justice that an employer should be immune from suit under s.48(1A) ERA for commencing litigation or arbitral proceedings against a whistleblower, irrespective of whether the employer considers himself fully justified in doing so. On the contrary, to apply JPI in this context would leave a wrong, recognised by Parliament in s.47(1B) ERA, without the very remedy to which Parliament itself has stated the whistleblower is entitled under s.48(1A). The public policy underlying the protection afforded to whistleblowers strongly indicates that JPI should not attach to the commencement of such proceedings. If it did, it would seriously undermine the protection that Parliament intended to apply.

For those reasons, I consider that the ET judge did not make an error of law in refusing to strike out Detriment 3. I would allow this appeal and reinstate that aspect of the claim.

I agree.

I also agree that the appeal must be allowed for the reasons given by Lady Justice Andrews. The provisions of the Employment Rights Act 1996 demonstrate a clear intention on the part of Parliament to provide a remedy, available in the Employment Tribunal, for a worker who suffers a detriment as a result of making a protected disclosure. For that remedy to be defeated by a claim for judicial proceedings immunity in circumstances such as exist here would be contrary to that clear parliamentary intention.