In Williamson v The Bishop of London & Ors [2023] EWCA Civ 379 the Court of Appeal held that a person subject to a Civil Proceedings Order must obtain permission from the High Court so they could issue valid proceedings.  The fact that permission was obtained after issue was irrelevant, the proceedings remained a nullity.

“… the filter is intended to ensure that neither respondents nor the courts and tribunals, are required to respond to, or otherwise deal with, claims sought to be brought by vexatious litigants unless and until the vexatious litigant has satisfied the High Court that the proceedings are not an abuse of process and there are reasonable grounds for instituting them … This can only be achieved if permission is sought before proceedings are instituted”


The appellant brought proceedings in the employment tribunal.  In 1997 the applicant had been made subject to an order under Section 42 of the Supreme Courts Act 1981 restricting the right to issue proceedings.  The order stated that the applicant was prohibited from:

“1. instituting any civil proceedings in any Court and

2. continuing any civil proceedings instituted by him in any Court before the making of this Order and

3. making any application other than an application for leave as required by section 42 of the [SCA] in any civil proceedings instituted in any Court by any person unless [the appellant] obtains the leave of the High Court having satisfied the High Court that the proceedings or application are not an abuse of the process of the Court in question and that there are reasonable grounds for the proceedings or application.”


The Employment Tribunal proceedings were started in April 2019, the appellant did not seek permission of the High Court prior to issue.


After issuing proceedings the appellant sought, and obtained, leave of the High Court, to continue the employment proceedings he had issued in April 2019.


The Employment Tribunal found that the proceedings had been commenced prior to permission being granted and was a nullity. A second set of proceedings, issued after permission had been granted, were dismissed because they were brought out of time.

The Employment Appeal Tribunal upheld the decision that the first set of proceedings were a nullity.


The applicant’s appeal to the Court of Appeal was unsuccessful.  It was held that the action was a nullity from the outset.  The statute was clear – permission had to be obtained prior to issue.

    1. The language of section 42 “no civil proceedings shall without the leave of the High Court be instituted…” clearly envisages that leave will be a condition precedent to the institution of proceedings (save in relation to existing proceedings at the time of the CPO). The “institution” of proceedings without permission is prohibited by section 42(1A)(a) SCA 1981 and permission can only be granted to continue proceedings where those proceedings were instituted before the making of the CPO: section 42(1A)(b) SCA 1981. There is no provision made for the grant by the High Court of retrospective permission to continue proceedings which were initiated without permission.

    1. Mr Wynne pointed out correctly that section 42 does not specify the consequences of a claim brought in breach of a CPO. Moreover, although on the face of it section 42 appears to envisage that leave will be a condition precedent to the institution of proceedings (save in relation to existing proceedings at the time of the CPO), he pointed out that this is achieved by saying that that will be the nature of the order made by way of a CPO, rather than by providing for it directly (unlike section 139(2) MHA). Accordingly, and since court orders with conditions precedent (like unless orders) have always been subject to relief from sanction, he submitted that the same must be true here. These are legitimate points to make, but it does not follow that that was the statutory intention. No doubt section 42 had to take this form because CPOs are to be granted on a litigant by litigant basis. In any event, I do not consider that this distinction can dictate the right answer. Nor is the statutory language on its own determinative, as Lord Bingham in Seal made clear.

    1. As Lord Bingham explained in R (Quintavalle) v Secretary of State for Health [2003] UKHL 13[2003] 2 AC 687 at paragraph 8:

“Every statute other than a pure consolidating statute is, after all, enacted to make some change, or address some problem, or remove some blemish, or effect some improvement in the national life. The court’s task, within the permissible bounds of interpretation, is to give effect to Parliament’s purpose. So the controversial provisions should be read in the context of the statute as a whole, and the statute as a whole should be read in the historical context of the situation which led to its enactment.”

    1. It is plain that the mischief at which section 42 is directed is to “avoid the unnecessary use of court time and resources on unjustified litigation and to protect prospective defendants from the expense which that involves”: see Ewing v News International (cited above, per Patten LJ at paragraph 18).

    1. Moreover, unlike the analogous provision made by the MHA, a CPO is directed only at vexatious litigants, who are as often as not motivated more by a desire to enjoy the oxygen of the legal process than any desire for or expectation of redress from it.

    1. A CPO can only be imposed where it is shown that the litigant in question has not only instituted civil proceedings or made applications in civil proceedings which can properly be stigmatised as “vexatious”, but also acted in one or other of these ways “habitually and persistently and without any reasonable ground”. In other words, the ordinary case management powers available to the court or tribunal will have proved insufficient to control the conduct of the litigant in question before a CPO can be made; and the vexatious litigant will be well aware of the situation. Thus Parliament clearly intended that an order under section 42(1A) would only be made where there is a proven need to protect the interests of the opposing party and the public against vexatious and abusive claims brought by the individual in question, and to protect the wider interests of justice by ensuring that the time and resources of courts and tribunals are not taken up by wholly unmeritorious litigation.

    1. The exercise of discretion to make a CPO is governed by clear safeguards: an application for an order may only be made on the authority of the Attorney General (or the Solicitor General acting on her behalf); and a CPO may only be made by the High Court. There is an oral hearing before the judge for this purpose unless the judge grants permission without a hearing or considers that the application is a substantial repetition of one which has already been refused.

    1. Moreover, a CPO operates as a filter and not a barrier. Once a CPO is made, it regulates a vexatious litigant’s access to the courts, rather than barring it. The vexatious litigant may not institute or continue or make an application in any civil proceedings unless a High Court judge is satisfied that the proceedings or application are not an abuse of the process of the court in question and that there are reasonable grounds for the proceedings or application. The vexatious litigant who is the subject of a CPO will know about the restriction that has been placed on their right of access, and the responsibility for making an application for leave must therefore lie on the subject of the CPO. Putative respondents or defendants (and the courts and tribunals themselves) may not have the same ready knowledge. While it is true that this process may act as a deterrent to further proceedings, it does not deny rights of access to justice.

    1. It seems to me that the filter is intended to ensure that neither respondents nor the courts and tribunals, are required to respond to, or otherwise deal with, claims sought to be brought by vexatious litigants unless and until the vexatious litigant has satisfied the High Court that the proceedings are not an abuse of process and there are reasonable grounds for instituting them (section 42(3) SCA 1981). This can only be achieved if permission is sought before proceedings are instituted. The requirement of permission will not operate as a filter or a safeguard against vexatious litigation if it can be given retrospectively (as this case demonstrates). Moreover, if proceedings can be instituted without permission, the onus will fall on others (instead of the vexatious litigant who is well aware of the restriction) to take the initiative and raise the question of permission. This reverses the deliberate onus provided for by section 42, and may undermine altogether the protection intended by section 42 SCA 1981.

    1. Thus the consequences of proceedings instituted in breach of a CPO made under section 42(1A) being a nullity do not involve any unfair prejudice or disproportionate breach of fair trial or access to justice rights. In particular, the vexatious litigant is afforded the safeguards just described. Further, the vexatious litigant will know that a CPO has been made and can apply timeously for leave. Even in the context of a relatively short limitation period of three months for proceedings to be commenced in the employment tribunal, I am in no doubt that this allows ample time to apply for and obtain leave to institute proceedings in the ordinary case. Moreover, an application can be made and heard urgently where necessary, for example in cases where urgent interim relief is sought. It is also the case that the fact that a vexatious litigant took reasonable steps to seek the permission of the High Court in good time before the expiry of the relevant limitation period, is likely to be a relevant consideration in deciding whether to extend time in relation to a fresh claim if that becomes necessary.

    1. Finally in this context, I reject Mr Wynne’s argument that respondents might suffer prejudice in fighting and winning a claim in ignorance of a CPO, and then not being able to recover costs on the grounds that the proceedings were a nullity. This should not arise. CPOs are gazetted and respondents always have the means available to discover whether a CPO is in force in respect of a litigant believed or known to be vexatious.

    1. On the other hand the adverse consequences of potential retrospective validation, where the claim is abusive, are considerable. Respondents and the court will have to incur the time and expense of being involved in that process. It may involve repeated applications as to when and how the stay application is to be heard. If the proceedings are not a nullity, the respondent will have to apply to strike out or stay the proceedings, and the vexatious litigant can invoke all sorts of procedures abusively, in relation to that application. In the experience of members of this court, lengthy written evidence, recusal applications, adjournment applications, disclosure and third party disclosure applications, applications to cross examine, consolidation with other applications are all weapons in the armoury used regularly by vexatious litigants intent on abusing the court’s processes. Respondents are also potentially exposed to having to engage with the substance of the claim because they can never be sure whether a discretionary retrospective validation will be sought and granted later. The deterrent of having to apply for leave first and pay a fee for it is altogether undermined.

    1. Nor do I accept Mr Wynne’s argument that the absence of provision in the ET Rules that a claim is a nullity means that employment tribunals are bound to exercise jurisdiction to hear a claim otherwise presented in accordance with the employment tribunal’s statutory jurisdiction and/or the procedural requirements for instituting such proceedings. If the statutory intention in section 42 is that CPOs should require prior leave as a jurisdictional hurdle, that must apply to all “courts”. The position must be the same in whichever “court” the vexatious litigant seeks to litigate. Notwithstanding that employment tribunals have statutory jurisdiction to hear and determine claims of unlawful discrimination under the Equality Act 2010 when presented in accordance with the ET Rules, their jurisdiction will be limited in precisely the same way as any court by a statutory requirement to obtain the leave of a High Court judge (effected through the mechanism of a CPO) before instituting such proceedings. The question must be a question of interpretation of section 42 and its effect.

    1. Equally it is by no means clear that every “court” to whose proceedings a CPO could apply, has the power to stay them on appropriate terms as Mr Wynne suggests, pending an application made subsequently to the High Court for retrospective leave under section 42. The only realistic terms on which such a stay might be ordered would be a time limited stay that is conditional on leave being sought and granted within a short time-frame, in default of which the claim should stand struck out without further order. Even if an employment tribunal has power to make such an order (as to which there might be some doubt given that there is no express rule providing for a stay order to be made, and although rule 29 empowers tribunals to make “a case management order”, the definition of “case management order” in the ET Rules might be said not to extend to such a limited and conditional stay that potentially results in the dismissal of the claim), Mr Wynne was not able to satisfy the court that the numerous different courts and tribunals to which this provision applies, all have the necessary power to stay proceedings on such terms. He frankly accepted that the power to stay or strike out proceedings, in criminal courts in particular, is limited, and may well not be available at all where proceedings are initiated in breach of a criminal proceedings order made under section 42. I consider this likely to be the case in relation to criminal proceedings in the Magistrates’ Courts in particular.

    1. The analogy Mr Wynne sought to draw with civil restraint orders does not advance his case either. First, while directed at a similar problem, CPOs and civil restraint orders are different powers. It is clear that CPOs are at the top of the hierarchy of orders available to restrain vexatious litigants in terms of their severity and the seriousness of the circumstances in which they apply. Unlike a civil restraint order, a CPO under section 42 covers all the litigation and all the applications a vexatious litigant may wish to bring, and if a High Court judge refuses permission in relation to any attempt the litigant may wish to make to bring a matter to the attention of a court that is the end of the matter. It involves the publication of the litigant’s name on a list which receives widespread circulation, and although CPOs may be made for a fixed period of time, they can be of unlimited duration, as was the CPO in the appellant’s case. Further, the statutory criteria for making an order under section 42 are different from, and more stringent than, the criteria for a civil restraint order which is made in exercise of the court’s inherent jurisdiction in accordance with CPR 3.11 and PD 3C. Accordingly, the fact that CPR 3.11 permits the relevant practice direction to set out the circumstances in which the civil restraint order power may be exercised, the procedure to be applied, and also “the consequences” of such an order, tells one nothing about the consequences of a more draconian CPO. There is no proper analogy between the two that can assist the appellant’s case.

    1. Mr Wynne placed significant emphasis on paragraph 4.3 of PD 3C – Civil Restraint Orders, which provides that:

“Where a party who is subject to a general civil restraint order – (1) issues a claim or makes an application in a court identified in the order without first obtaining the permission of a judge identified in the order, the claim or application will automatically be struck out or dismissed …”

He submitted that these consequences are noticeably different from the concept of a claim being a nullity because jurisdiction over a civil restraint order is maintained in the sense that the possibility of seeking relief from sanctions under the general provisions in CPR 3.9 continues to apply. I am far from satisfied that he is correct and that relief from sanctions is available in the case of an automatic strike out where a civil restraint order is breached (or indeed that Couper v Irwin Mitchell LLP [2017] EWHC 3231 (Ch)[2018] 4 WLR 23, at paragraphs 28 to 29, was correct in this respect, and note that the point was not argued or fully addressed).

    1. In any event, a CPO is consciously made in more serious circumstances and with more draconian effect: see Bhamjee v Forsdick [2003] EWCA Civ 1113[2004] 1 WLR 88 at paragraphs 20 to 24 and 39 to 47. Accordingly, whatever the consequences of a civil restraint order, that does not mean that the same consequences must or do apply where proceedings are commenced in breach of a CPO.

    1. Nor, contrary to the arguments developed by Mr Wynne, is there any real analogy with the provisions of section 285(3) Insolvency Act 1985 and section 115 Charities Act 2011, both of which have been interpreted as providing that the failure to obtain the necessary permission does not render proceedings a nullity. The interpretation of different statutory provisions in a different context, does not advance the appellant’s case for the following shortly stated reasons.

    1. Dealing with the insolvency context first, the purpose of section 285 of the Insolvency Act 1985 is to prevent the liquidator or administrator’s task being made difficult “by a scramble among creditors to raise actions, obtain decrees or attach assets” (see Lord Coulsfield in Carr v British International Helicopters Ltd [1993] 8 WLUK 52, cited by Lindsay J in In Re Saunders [1997] Ch 60; [1997] 3 All ER 992 at paragraph 79). This purpose is not undermined if proceedings instituted without the necessary permission are not a nullity. Further, an obvious injustice would arise if proceedings brought by a creditor in that position were subsequently treated as a nullity. By contrast, the subject of a CPO is necessarily someone who has been held to have “habitually and persistently and without any reasonable ground… instituted vexatious civil proceedings… or made vexatious applications… or instituted vexatious prosecutions …” and no obvious injustice arises in a case where the vexatious litigant has knowingly chosen to institute proceedings without obtaining leave first. It is also significant that Lindsay J expressly recognised that his interpretation (that proceedings were not a nullity) went against the literal meaning of the words of section 285 in any event: see paragraphs 72 and 82 to 83.

    1. The short answer to the comparison drawn by Mr Wynne in the charities context is that section 115 Charities Act 2011 uses significantly different language to that used in section 42 SCA 1981. It provides that no charity proceedings are to be “entertained or proceeded with” unless “the taking of the proceedings” is authorised by an order of the Charities Commission. It does not restrain the institution of charity proceedings without permission. Those words do not contain the same clear prohibition against the institution of proceedings as section 42(1A) SCA 1981. Indeed, in Park v Cho [2014] EWHC 55 (Ch)[2014] PTSR 769 at paragraphs 39 to 40 the “taking of” proceedings was held not to be limited to the commencement of proceedings and extended to the continuation of proceedings already commenced. Charity proceedings are therefore expressly capable of being continued despite their commencement without the permission of the Charities Commissioner. This does not undermine the purpose of section 115(2) of the Charities Act 2011 (or its predecessor provision in section 33(2) of the Charities Act 1993) because the court can place a stay on the proceedings pending authorisation by the Charities Commission and thus ensure that charitable funds are not frittered away in the pursuit of litigation relating to internal disputes.

    1. There are good reasons why as a matter of general principle procedural failures should not lead to proceedings being a nullity. But that does depend on the purpose and importance of the provision in its statutory context. It seems to me to be evident for all the reasons I have given, that the express terms of section 42 SCA 1981, read in context and in light of the object and purpose of the section, impose a jurisdictional (and not merely a procedural) barrier on a litigant subject to a CPO wishing to institute proceedings. In my judgment Parliament intended to make leave under section 42 SCA 1981 a jurisdictional bar to the institution of effective proceedings where a CPO has been made. Neither the prospective respondent nor the court is required to take action where a proposed claim is made by a vexatious litigant unless and until the proceedings have the required leave of a High Court judge. As with section 139(2) MHA and as Lord Brown observed in Seal, the very inflexibility of the provision is an integral part of the protection it affords.

  1. It follows that the tribunals below were correct to conclude that the employment tribunal proceedings commenced by the appellant without first obtaining the necessary leave of the High Court were and remain a nullity. For these reasons, which are essentially the same as those given by Eady P, the grounds of appeal cannot succeed. I would dismiss this appeal accordingly.