This application raises a large number of questions about the ability of claimants to have themselves added to a claim after a limitation period has expired and about amendments to the descriptions of claimants (to use a neutral term) after service and after the expiry of the limitation period.

Since 2005 the defendant (“G4S”) has provided security and other services to the government. Those services included the tagging of prisoners. On 17^th May 2013 the Secretary of State for Justice, Mr Chris Grayling, announced that G4S was to be investigated in respect of billing issues. An independent team was to audit the processes and information supplied by G4S. On 11^th July 2013 Mr Grayling announced that the Ministry of Justice had identified anomalies in the charges made by G4S in that charges were made in respect of prisoners for whom charges should never have been made (for example, because they had died). He had referred the matter to the Serious Fraud Office. More information emerged until in March 2014 it was announced that there had been a settlement of the government’s claim in the sum of over £108.9m.

On 10^th July 2019 various claimants (“the original claimants”) issued a claim against G4S under section 90A of the Financial Services and Markets Act 2000. Their claim was based on the fact that in the period since 2011 they had all bought and held shares in G4S and they had suffered loss in that their shares lost value as a result of the announcements about what was said to be a fraud. They claim to have relied on public statements by G4S in its annual reports and other documents in which it trumpeted the strength and honesty of its business practices (putting the matter shortly), and those statements turned out to be false. Based on that falsity, the reliance and the decline in share values, the claimants claimed losses under the section. It is unnecessary to set out the precise provisions of the section. The claim was not preceded by any form of letter before action or any of the other usual warnings or intimations of claims. This issue date is significant because it is the last day of the limitation period if the starting point was the announcement of 11^th July 2013 (and it is accepted by the claimants that that is at least arguable).

The claim form was not served until 30^th April 2020, there having been agreed extensions of time for service to cover the intervening period. In that intervening period, and unknown to G4S, the claim form underwent a series of amendments as to claimants. Some claimants were added on the basis that they had their own claims of the same nature as the claims of investors who were already parties. On the occasion of some amendments claimants were removed from the claim form. Some removed claimants were those who originally appeared, but most were of claimants subsequently added, so they came and went after the issue of the claim form.

There were no less than 6 sets of amendments made during that period. 3 of the amendments, adding 9 claimants, were separate amendments carried out on the same day (11^th July 2019, the day after issue and therefore the day after the arguable expiry of the limitation period applicable to the pleaded claim). 179 claimants were added on 28^th August 2019, and 3 more on 11^th November. On 30^th April 2020 (shortly before service) 93 claimants were removed. When one nets off the position there are 2 additional claimants (net) from the 11^th July amendments, and all the remainder of the surviving claimants (62 of them) were added in the August amendments. Mercifully the precise detail of this does not matter on the basis on which the arguments were run.

Thus the claim form has 5 “Re-“s before the word “Amended” and a virtually complete spectrum of colours seldom seen on pleadings, and all before the claim form was served. The original number of claimants numbered 43. At its high point the claim form contained 182 claimants (after deletions and additions), before settling down to 93. (I take all these figures from the unchallenged account of Mr Rabinowitz. If it turns out they are a bit wrong it does not matter for the purposes of this application).

Once the claim form, and then Particulars of Claim, had been served, G4S set about challenging the addition of claimants who were added after the issue of the claim form. I shall call those claimants “the additional claimants”. They do so on the basis that the CPR do not permit the addition of claimants before service (or at least not without permission of the court), either generally or where, as here, there is said to be an arguable limitation defence. The claimants oppose that, and mount their own further amendment application. This time the amendments are not to add (or subtract) more parties, but are to amend the words being or purporting to be the names of claimants as they appear on the claim form. I have tried to use as neutral a description as possible because the nature of the changes is a hotly contested matter on that application.

The applications before me were the subject of a woefully inadequate time estimate, with the result that extra days had to be found, with an interval of a week between two sets of days. That gave the defendants the opportunity to make a yet further application, being an application for relief from sanctions in case that was necessary to bring their attack on the addition of the added claimants under a different head from that specified in their application notice.

The dispute in the various applications centred around provisions of the CPR permitting amendments but limiting amendments which can be made after the expiry of a limitation period relevant to a claim. They were CPR 17.2, 17.4, 19.4 and 19.5. In what follows I shall not always prefix references to CPR provisions with the abbreviation “CPR”; it will be obvious when references to provisions are being made and it will, I hope, aid exposition if I omit unnecessary prefixes.

What is at stake in these applications is the ability of a large number of claimants, or would-be claimants, to bring claims under section 90A. The significance of the exercise can be judged by the amounts at stake. In broad terms the claims of all the would-be claimants are said to be worth £102m. If all the claimants joined from 11^th July 2019 onwards cannot be joined then £92m falls out of the picture. So in financial terms the present dispute, if decided entirely in favour of the defendant, could seal the fate of the bulk of the claims in financial terms.

In a little more detail, the applications made by the parties are as follows.

Declares that the addition of claimants to the claim form after 10 July 2019 was ineffective.

Pursuant to CPR 3.4(2)(b) or (c), references in the Particulars of Claim to claims by claimants who were added to the claim form after 10 July 2019 shall be struck out.

Pursuant to CPR 3.4(2)(a), (b), or (c), claims by claimants who have not been properly identified or are not legal persons shall be struck out.

Pursuant to CPR 3.4(2)(b) or (c), allegations in the Particulars of Claim in relation to statements, omissions or delays prior to September 2011 shall be struck out.”

There was a fifth claim, but that is no longer relevant. The following issues arise out of all that:

(3) embodies a complaint that a (large) number of claimants are not properly identified for a number of reasons - mainly that they are not legal entities with appropriate capacity to sue, as their names appear in the claim form. This affects a large number of claimants. I shall call this the “unidentified claimants point”, after the nomenclature adopted in argument, even though as a description it is not entirely apt.

(4) embodies a complaint to the effect that the claim form seems to limit claims to the period from 2011, whereas the Particulars of Claim seeks to extend the claims back to 2006 or earlier. This is the “2011 point”.

As foreshadowed above, during the hearing, in an application notice dated 9^th November 2020 G4S sought relief from sanctions so as to enable it to make an application under CPR 17.1 outside the time limit provided by 17.2. This is said to be a precautionary measure because G4S’s main point is that it is not necessary.

The claimants made an application in a notice dated 21^st October 2020 to amend their statements of case to do two things. First, they sought to amend the claim form so that the claims made went back to 2006 or so rather than 2011 and the following years. It is in effect the counterpart to head 4 of G4S’s application - the 2011 point. These two matching applications raise a limitation issue because the claims which are sought to be introduced fall outside the limitation period. Second, they sought to change the pleaded names (in using that word I do not intend to pre-judge one of the issues arising in these applications) of a large number of the claimants where it is said (putting it broadly) they have been misnamed. This raises questions of the extent to which alterations can be made to parties or names of parties outside the limitation period. There are 52 of these names, but at my insistence the matter was tested by reference to examples. This turned out to be a very substantial debate turning on the extent to which amendments to parties could take place outside the limitation period. This is all part of the unidentified claimants point.

The question of limitation and its interaction with the power to amend looms large in these applications and it will be useful to set out some generally applicable matters in this separate section of this judgment.

An amendment to a claim form dates back to the date of the issue of the claim form. Accordingly there is the potential for a claimant to amend to add to his/her claim a claim which was statute-barred at the date of the amendment but not at the date of the issue of the claim form, and avoid the consequences of the Limitation Act 1980 so far as that new claim is concerned. In order to meet or control that the Limitation Act contains section 35, and the CPR contain Rules 17.4 and 19.5, which limit the extent to which such amendments can be carried out. They all refer to claims sought to be brought after the period of limitation applicable to that claim has expired.

In many instances there will be a dispute as to whether the limitation period has expired at the date of the attempted amendment. It was said by G4S, and accepted by the claimants, that where there was such a dispute which affected an amendment (ie it is reasonably arguable that the claim is statute-barred), and the amendments would not otherwise fall within the provisions allowing a post-limitation amendment, then the usual course was for the court to refuse permission to amend and leave the claimant to issue a fresh claim form. This appears from Chandra v Brooke North [2013] EWCA Civ 1559, in an oft-relied on passage:

If, as is usually the case, the court adopts the first option, it will not descend into factual issues which are seriously in dispute. The court will limit itself to considering whether the defendant has a “reasonably arguable case on limitation”: see WDA at 1425 H. If so, the court will refuse the claimant’s application. If not, the court will have a discretion to allow the amendment if it sees fit in all the circumstances.

If the court refuses permission to amend, the claimant’s remedy will be to issue separate proceedings in respect of the new claim. The defendant can plead its limitation defence. The limitation issue will then be determined at trial and the defendant will not be prejudiced by the operation of relation back under section 35 (1) of the 1980 Act.

This leads on to a separate and important point. If a claimant applies for permission to amend and the amendment arguably adds a new claim which is statute barred, then the claimant should take steps to protect itself. The obvious step is to issue separate proceedings in respect of the new claim. This will have the advantage of stopping the limitation clock on the date of the new claim form. If permission to amend is granted, then the second action can be allowed to lapse. If permission to amend is refused, the claimant can pursue his new claim in the second action. The two actions will probably be consolidated and the question of limitation can be determined at trial.”

In the present case this would apply if it is arguable that, as at the date of any of the amendments, the limitation period applicable to the claim had expired. It would have the effect that if G4S was in a position to dispute the amendment and took the limitation point, the amendment ought to be disallowed unless it could be brought within the exceptions provided by the rules and the statute. I do not understand Mr Onslow to dispute that position.

It is therefore necessary to consider whether there is, as at any relevant date, an arguable limitation defence. G4S takes a limitation point in its Defence, saying that the limitation period started running in May 2013 when Mr Grayling made his first announcement. That is when loss accrued so far as diminution in value of the shares is concerned. It is not suggested that the claimants can have known about the wrongdoing before then, but that is the first date on which it can be said that the claimants can have sufficient knowledge that they had a claim. That, G4S would say, is at least arguable, with the effect that there was an arguable limitation defence as at 11^th July 2019, which is when the first amendments were made. Alternatively there is a good arguable case that the limitation period started running on 11^th July 2013 when the second announcement was made. G4S say that that, again, puts all the amendments outside the limitation period.

The claimants’ case is that they did not know enough to justify them bringing a claim until some date after the end of August 2013, by which time all their amendments to the claim form had been made. So far as necessary they would rely on section 32 of the Limitation Act 1980. The limitation period therefore did not start running until a date which would make all the amendments to add claimants in-time amendments for the purposes of limitation.

It is not for me actually to decide the limitation point in this case, or even to go into it in depth. The question which matters is whether it is arguable that the limitation period had expired by 11^th July 2019, or to put it another way whether it had started running by 11^th July 2013. It seems to me clear that that is arguable. It is arguable that by 10^th July 2013 the claimants would have known that the company in which they held shares was the subject of a serious investigation, and that loss had arisen by then. Mr Onslow actually accepted in argument, that what he called the primary limitation period (by which he meant one which was not subject to a particular extension) had expired on or before 10^th July 2019. However, he relied on section 32 of the Limitation Act, which provides that in cases of fraud or fraudulent concealment the period does not start running until the claimant could have discovered material relevant or necessary to the claimants’ case (a paraphrase suffices). He said his clients ran that argument, and that it was sufficiently clear now, on the material that the court has, that his clients could not with reasonable diligence have discovered the fraud before the end of August 2013. He submitted that that was clear enough to make the contrary case unarguable.

I do not accept Mr Onslow’s argument. As just appears, it depends on this court deciding, at this stage, that it is clear that on the facts Mr Onslow has a section 32 extension so there is no limitation point which can be run. He effectively accepted as much in argument, and it is reflected in his Reply. I find that that is simply not the case. It cannot be said that the section 32 clock clearly did not start running until the end of August 2013 at the earliest. It is arguable (it is unnecessary to go further) that it started running on 11^th July 2013, if not earlier. Mr Onslow’s argument may be right, and it may be wrong, but it is arguable. That much follows from the claimants’ Reply. In paragraph 5.4 they set out a series of dates, from latest to earliest, as being the dates before which the claimants could not have discovered, or with reasonable diligence could not have discovered, the fraud and concealment said to be involved. The fifth of those is “prior to 11^th July 2013 when Mr Grayling [made his July announcement]”. That is actually pleading a case which contributes to the arguability of the point.

Although it is not strictly relevant to the point, I would point out that the possibility of a limitation defence must have been apparent to the claimants’ solicitors (and by extension to the claimants) in the period leading up to the issue of the claim form. In the evidence adduced at the hearing Mr Warren-Smith, the claimants’ solicitor, explains the circumstances of the timing of the issuing of the claim form. In a witness statement dated 21^st October 2020 he says:

That is almost an acceptance in terms of an arguable limitation defence being available. Although that might be presented as just his view, which is irrelevant to the question of whether objectively there is an arguable limitation point, he is in effect acknowledging its arguability. In my view he was right in doing so.

It follows, therefore, that where it is relevant, the amendment debate must take place against the background that there would be an arguable limitation defence to any new claim akin to the issued claim were that new claim to be issued on or after 11^th July 2019. On the facts of this case that applies to all the amendments that purport to have been made or in respect of which applications to amend are now made.

As already indicated, the debate in this case turns around a handful of provisions of the CPR and the Limitation Act.

Rules 17 and 19 deal with the addition of new parties. CPR 17.1 provides:

CPR 17.4 provides for amendment after the expiry of the limitation period where the amendment is for purposes other than the addition or removal of a party, but it covers amending the name of a party. It is relevant to the 2011 point and the unidentified claimants point:

a party applies to amend his statement of case in one of the ways mentioned in this rule; and

a period of limitation has expired under—

The court may allow an amendment whose effect will be to add or substitute a new claim, but only if the new claim arises out of the same facts or substantially the same facts as a claim in respect of which the party applying for permission has already claimed a remedy in the proceedings.

The court may allow an amendment to correct a mistake as to the name of a party, but only where the mistake was genuine and not one which would cause reasonable doubt as to the identity of the party in question.

Rule 19.4 deals with the “Procedure for adding and substituting parties”. It provides:

an existing party; or

a person who wishes to become a party.

he has given his consent in writing; and

that consent has been filed with the court.

In the present case a point arises under 19.4 because it is said by G4S that there was no filed consent in writing to the addition of any of the added claimants.

CPR 19.5 deals with adding or substituting parties after the end of a limitation period. By now its relevance will have become apparent. It provides:

the relevant limitation period was current when the proceedings were started; and

the addition or substitution is necessary.

the new party is to be substituted for a party who was named in the claim form in mistake for the new party;

the claim cannot properly be carried on by or against the original party unless the new party is added or substituted as claimant or defendant; or

the original party has died or had a bankruptcy order made against him and his interest or liability has passed to the new party.

CPR 17.4 and 19.5 reflect the provisions of section 35 of the 1980 Act. Its limits need to be borne in mind when construing the CPR, because it contains limits as to what can be done outside the limitation period.

in the case of a new claim made in or by way of third party proceedings, on the date on which those proceedings were commenced; and

in the case of any other new claim, on the same date as the original action.

the addition or substitution of a new cause of action; or

the addition or substitution of a new party;

Except as provided by section 33 of this Act 17 or by rules of court, neither the High Court nor the county court shall allow a new claim within subsection (1)(b) above, other than an original set-off or counterclaim, to be made in the course of any action after the expiry of any time limit under this Act which would affect a new action to enforce that claim. For the purposes of this subsection, a claim is an original set-off or an original counterclaim if it is a claim made by way of set-off or (as the case may be) by way of counterclaim by a party who has not previously made any claim in the action.

Rules of court may provide for allowing a new claim to which subsection (3) above applies to be made as there mentioned, but only if the conditions specified in subsection (5) below are satisfied, and subject to any further restrictions the rules may impose.

The conditions referred to in subsection (4) above are the following—

in the case of a claim involving a new cause of action, if the new cause of action arises out of the same facts or substantially the same facts as are already in issue on any claim previously made in the original action; and

in the case of a claim involving a new party, if the addition or substitution of the new party is necessary for the determination of the original action.

The addition or substitution of a new party shall not be regarded for the purposes of subsection (5)(b) above as necessary for the determination of the original action unless either—

the new party is substituted for a party whose name was given in any claim made in the original action in mistake for the new party’s name; or

any claim already made in the original action cannot be maintained by or against an existing party unless the new party is joined or substituted as plaintiff or defendant in that action.”

I can now turn to the points that arise in this hearing, starting with the added parties point. Mr Rabinowitz’s case is that the addition is not permitted by the rule relied on (17.1) on its true construction and he applies to strike out all the added claimants. This point is based on the language of the rule, and the proposition that an existing “party” who seeks to amend “his statement of case” is not doing that if what is sought to be done is to introduce a different party with a distinct separate case.

Mr Onslow’s case is that the new claimants can be joined without permission under and by virtue of CPR 17.1. The wording permitted that. He sought to demonstrate that if that were not the case then there would seem to be no provision in the CPR which applied so as govern and permit the joinder of a claimant before service of the claim form. The way it worked was to allow joinder under that rule and, if the defendant did not like it, the defendant could apply to the court to disallow the amendment under CPR 17.2, provided that the application was made within the two weeks referred to in that provision. In the present case there was no such application, and in any event the present application had not been made within 14 days of service of the claim form. It was explicitly made under a provision other than CPR 17.2 (namely CPR 3.4) and no application had been made to extend the time for making an application under that provision (or at least not until the late application to be allowed to do so).

The point is a simply stated one on which I accept Mr Rabinowitz’s submissions. The provision allows “A party” to do something. After an initial attempt to submit that this word was capable of extending to a would-be but not-yet party, Mr Onslow retracted and accepted that this was a reference to an existing party who had already issued its claim form and not the person whom it is sought to introduce. I agree that that concession was correctly made. In their context the words can only refer to an existing party.

The provision then allows that party to amend “his statement of case”. In my view the natural meaning of those words is such that it refers to the statement of case embodying the claim that that claimant is making or seeks to make. It distorts the words to take it any wider than that. An amendment to plead another claimant’s entirely separate case is not so much an amendment of the existing claimant’s claim form by that claimant (though it would, I accept, change the document itself); it is bringing in a new person who is bringing in a separate and distinct claim. It does not seem to me to be a natural construction to treat that as an amendment by the existing party of “his statement of case” when one considers what a statement of case is.

What a statement of case is is set out in other parts of the CPR. The claim form of an existing claimant (which is the relevant statement of case in this instance) has to contain the matters referred to in CPR 16.2 - a concise statement of the nature of the claim (which must mean the claimant’s claim), the remedy which “the claimant” seeks, and certain material where “the claimant” is seeking certain remedies. All this points clearly to the fact that the claim form is, as one would expect, a document which is geared to the claim that the claimant is making.

That makes it harder to read CPR 17.1 as allowing a claimant to introduce another claimant with a different claim. In doing so the existing claimant is not doing anything to its existing claim. It is not even adding another of its own claims (which I accept would be possible under this rule). It is doing something rather different.

Mr Onslow sought to analyse the matter differently. He suggested that if that construction were correct it would mean that there could be no addition of a claimant before service of a claim form because there was no other provision of the CPR which provided for that. The remainder of CPR 17 and CPR 19 dealt with the addition of parties (and therefore a claimant) after service, but not before. The way the rules worked was that CPR 17 was the gateway to all amendments. If there was a change of parties before service, CPR 17.1 applied and CPR 17.2 provided for disallowance if the change was impermissible. The time limit in CPR 17.2 was necessary to make sure there was one; otherwise a very late challenge could be made to an amendment

I agree with Mr Onslow that one could produce a coherent whole out of the elements of the CPR by adopting his analysis of the provisions. If the defendant’s interpretation produced an incoherent whole then that would be a possible reason for construing 17.1 his way. However, the construction which I favour still produces a coherent regime. While 19.4 (change of parties) applies “unless the claim form has not been served”, and therefore assumes that there can be a change of parties prior to service of the claim form, that qualification can still have work to do even if adding a claimant with its own separate claim is not permitted under 17.1. It can, for example, cover the removal, addition or substitution of defendants. It could cover the position of a claimant which seeks to remove itself from a claim form - I do not see why that would not be within 17.1, because the removing claimant would be removing its own claim by amending its own statement of case. So it is not necessary to adopt Mr Onslow’s interpretation to make sense of 17.1 or to give it some useful effect.

The rest of the provisions make sense on Mr Rabinowitz’s construction as well. A would-be claimant is not deprived of an opportunity of suing. It can still commence its own proceedings and, in due course and if appropriate, apply for consolidation or for the cases to be heard together. The other provisions of CPR 17 and CPR 19.4 all work as they appear to operate on their face. The distinction between adding a party with a new claim on the one hand and other amendments which are properly viewed as amendments to a claim made by an existing claimant, whether by adding a cause of action or amending an existing one, or joining a new defendant, is one that can be justified rationally. They are all amendments which the existing claimant wishes to make to the claim that he/she originally brought, or (in the case of a new cause of action) to his position as a claimant set out in “his” statement of case. A separate claim made by a separate claimant can be said to be (and is) qualitatively different.

I therefore consider that the attempt to join separate claimants in this case fails for that reason. The amendment made was not one which was allowed under 17.1 and is liable to be struck out.

The next question is whether a challenge to such an amendment has to be made under CPR 17.2 or whether it can be made under CPR 3.4. The significance of that is that

In my view a challenge based on an averment that something has happened which is not actually an amendment within 17.1 can be made under CPR 3.4 and does not have to be, and indeed cannot be, made under 17.2. 17.2 applies where “a party has amended his statement of case where permission was not required”. In a case such as the present where a claimant has done something that it is not allowed to do under 17.1, it should not be treated as having amended its statement of case for these purposes. Accordingly CPR 17.2 does not apply. A challenge can be made under CPR 3.4, as it has been in this case. I do not consider that the challenge has to be made under 17.2 (which has to be made within 14 days of service) because the wording of that section pre-supposes an amendment which can otherwise be properly made under 17.1. On my conclusions about 17.1 that pre-supposition is not fulfilled.

Mr Onslow made submissions to the effect that such a conclusion left a claimant exposed to an application challenging the amendment long after the amendment had happened, and that certainty required that 17.2 should apply to prevent that happening. The defendant would have 14 days to challenge what has happened and there would be certainty as to the position if a challenge was not made within that time. I do not consider that that submission succeeds. Obviously the court would need to address the situation if, after proceeding on the footing that a joinder was apparently valid, a defendant, perhaps months later turned round and challenged it. Rules about waiver and estoppel, or analogous rules coupled with discretion, are amply sufficient to cope with that situation.

On this basis the addition of the claimants after the original issue falls to be disallowed.

That conclusion strictly makes it unnecessary to consider the other points which arise in relation to the added claimants. It also limits the number of claimants in relation to which the unidentified claimants point arises because a large proportion of those unidentified claimants were added after issue. However, in case this matter should go further, and since I heard extensive submissions on them, I will deal with them in the following sections.

This point arises if in principle a claimant can add another claimant without permission under CPR 17.1. Mr Rabinowitz’s next point is that it was impermissible (a “nullity”) to purport to add a party under 17.1 if the limitation period had expired or had arguably expired. His point was not just that amendments which arguably added claims outside the limitation period should be disallowed on an application to that effect pursuant to Chandra (though he made that point as well in his alternative challenge under CPR 17.2); it went further and averred that the attempt to amend was outside the scope of the rule and ineffective – it was a nullity.

The starting point for this argument is the Chandra case referred to above. It is argued that pursuant to that case, and pursuant to the provisions of the Limitation Act 1980 section 35 and CPR 19.5, 17.1 cannot be used to join a claimant where the limitation period has expired or has arguably expired. So far as the 1980 Act is concerned, section 35(5) permits a new claimant after the expiry of the limitation period only “if the addition or substitution of the new party is necessary for the determination of the original action”. It is said that to allow a new party to be joined after the limitation period has expired would be ultra vires unless the addition were necessary within subsection (5), and in the present case no case of necessity has been made out.

In making these submissions Mr Rabinowitz claimed support from two cases. The first was Best Friends Group v Barclays Bank plc [2018] EWCA Civ 601. This case involved an attempt to change the name of a claimant from Best Friends Group to “Andrew Bennett” after the expiry of the limitation period applicable to the claim. The claim form was issued 2 years after the expiry of one applicable limitation period, and 2 weeks before the expiry of another. There was an amendment, or purported amendment, to change it before service. However, it would seem that the claimant(s) regarded that as ineffective because he/they subsequently themselves applied to change the name of the claimant under CPR 17.4. In the course of his judgment Simon LJ observed:

Mr Rabinowitz relied on the reference to ineffectiveness in the second sentence of that citation.

He then went on to rely on Jalla v Royal Dutch Shell plc [2020] EWHC 459 (TCC). That case involved an attempt to amend the name of a corporate defendant by way of a name change (not adding a claimant, or at least that is how it was presented) under 17.1 but the matter proceeded in a procedurally odd way. It appears that the defendants did not take a point about the form or the propriety of joinder, but the claimants somehow and for some reason sought to regularise the position by an application to amend under CPR 17.4 and/or 19.5 - see paragraphs 15-17 of the judgment of Stuart-Smith J. At paragraph 21 the judge reflected on Chandra and set out the position, namely that if on an amendment application it appeared that there was an arguable limitation point, then the appropriate course was not to decide it but to refuse permission and leave it to the claimant to issue fresh proceedings in which the limitation point could be tried without prejudicing the position of a claimant by virtue of the relation back effect of an amendment.

The significance of Jalla to the present case lies in remarks made by Stuart-Smith J to the effect that an attempt at joinder of a defendant after the expiry, or arguable expiry, of the limitation period applicable to that party is a “nullity”. In paragraph 88 he said:

His statement about that is foreshadowed in paragraph 15:

Mr Rabinowitz fastened upon the paragraph 88 statements as supporting his case that joinder under CPR 17.1 outside the limitation period was a nullity. He was entitled to have the purportedly joined parties struck out as of right, and did not have to engage with CPR 17.2 and its time limit because there had been no valid amendment under

I do not accept Mr Rabinowitz’s case on this point. I agree with Mr Onslow that this would produce an unworkable scheme. It might just be workable if the bar were confined to cases where it could be plainly determined that a limitation period had expired, but as soon as one introduces the concept of a case which, at the time, is only arguably statute barred (which is what Mr Rabinowitz needs in this case) then the workability breaks down. The concept of nullity is an absolute one. It can only be sensibly applied in clearly specified circumstances on the basis of clearly applicable criteria. The more indeterminate concept of an arguable limitation bar is not one that can sensibly applied to a concept of nullity as proposed in the present context. It can (and indeed is - see Chandra) applied to disallow an amendment or refuse an application for permission to amend, but that is different. A regime under which an amendment (if prima facie allowed) is permitted with scope to disallow it on a subsequent application is more sensible. If an amendment falls outside the scope of post-limitation amendments permitted by section 35 and the rules, then the discretion in CPR 17.2 could only be exercised one way (Mr Rabinowitz abandoned an earlier submission that he made that if Mr Onslow were right then the discretion would bypass the limitation safeguards introduced by those provisions). I do not consider that nullity based on the concept of an arguable limitation defence makes sense. It also does not sit with the well-established concept that limitation is a procedural matter and a point which does not have to be taken by a defendant - that would not be possible if joinder was strictly a “nullity”. A procedural regime which allows for a subsequent challenge if limitation is shown to be arguable (or indeed conclusively established) makes more sense.

I do not consider that the two authorities relied on by Mr Rabinowitz compel me to the contrary conclusion. They are neither of them determinations by a court of the point which is now in issue. Best Friends records a concession, albeit one approved by Simon LJ. The concession is expressed in terms of the attempt being “ineffective or must be set aside”. The word “ineffective” might be equated with the word “nullity” for these purposes, but it is twinned with the concept of setting it aside, which rather reduces its force and suggests that it might not be the correct concept, or if it is then it is not as strong as “nullity”. So far as Jalla is concerned, again the point would seem to be rooted in a concession and is not the object of a full determination by Stuart-Smith J on the basis of argument. He was clearly reflecting the position of both sides, whose respective positions obviously puzzled the judge (see paragraphs 15 to 17). The claimants were not pressing 17.1, which they had originally relied on, and the defendants did not themselves take the point (as a knockout blow to the original amendment) that the amendments were not effective and ought to be struck out (paragraph 15). In those circumstances the effect to be given to paragraph 88 is very limited.

In the course of the break in the hearings forced by the over-running of this case Saini J delivered a judgment in Qatar Airways Group v Middle East News [2020] EWHC 2975 (QB). That was the case of a joinder of a defendant, not a claimant, under CPR 17.1 and outside what was said to be the applicable limitation period. The defendants took the same point about nullity based on the same authorities. Saini rejected it for much the same reasons as I do, and I respectfully agree with him.

It follows that this particular attack on the joinder would fail if it were relevant. However, Mr Rabinowitz still has another, simpler, limitation point available to him. If he needs to he succeeds on this point, as identified at the end of the next section of this judgment.

This point arises if I am wrong in my first conclusion about CPR 17.1 and it is possible to add a claimant under that rule before service and without permission. In that event a defendant would have to challenge the joinder and invite the court to disallow it. Mr Onslow’s case is that the proper and only route is under 17.2, which requires an application under that provision and an application made within the 14 days specified. G4S had not made such an application. Were it to seek to do so now it would, in effect, have to fulfil the test for relief from sanctions because of the lateness of the application but, as at the date of opening of the hearing, it had not done so.

Mr Rabinowitz disputed the need to apply under 17.2, and the absolute need to do so within 14 days if there were such a need. If necessary his application could stand as an application under 17.2, notwithstanding that it claimed to be under CPR 3.4, and during the hiatus in the hearing he actually made an application for relief from sanctions so he could be seen to be applying under 17.2 out of time should that be necessary.

A number of questions arise from these submissions and activities. The background is that G4S plainly did not apply within 14 days of service of the claim form (which clearly indicated on its face that the amendments had been carried out pursuant to CPR 17.1) - G4S’s original application notice (invoking CPR 3.4) came some 8 weeks afterwards, as did the Defence in which the point was also taken.

Logically the first question is whether an application under 17.2 is the only route for a challenge to pre-service joinder, or whether CPR 3.4(2)(b) (the power to strike out for abuse of process or obstructing the just disposal of the proceedings) is available as well. If the latter is available then no time limit is imposed. Professor Zuckerman in his work on the Principles of Practice para 7.44 seems to think that both are available, with the time consequences just referred to:

The question of whether there are two routes or just one, and the obligatory/voluntary nature of the 14 day time limit in 17.2(2), are intertwined. There are two possibilities.

First, the provisions and requirements of 17.2 are the proper line of attack, having been apparently provided for just that purpose. One starts with 17.1 - the power to amend a statement of case before service. That is a facility afforded to the amending party (who will almost invariably be a claimant in respect of his/her claim form - most other statements of case can be changed before service simply because they have no status at all until service). 17.2 gives a protection to the amended-against party. Without it it would not be so clear what that party could do. 3.4(2) would be the only route, for which purpose one would have to establish an abuse or obstruction of justice, which are quite stringent descriptions. However, it becomes unnecessary to worry about that because 17.2 provides that an application can be made. There is, however, a limit on that. The party applying has to apply within 14 days of service. That is to make sure that the matter is addressed reasonably promptly, and it is obligatory - “may” means “must”. If CPR 3.4(2) were also available, then 17.2 would be redundant and the time limit completely irrelevant. 17.1 and 17.2 go together as the complete code in terms of the timing and challenging of pre-service amendments.

The second is that proposed by Mr Rabinowitz. CPR 3.4(2) and 17.2 are parallel routes, as Prof. Zuckerman suggests. There is no reason why they should not be - an

I do not consider Mr Rabinowitz’s argument to be correct and prefer the first line of argument. I do indeed consider that 17.1 and 17.2 fall to be read together, with the first provision providing the power and the second imposing a constraint on the power. It may well be the case that without it the amended-against party could invoke CPR 3.4(2), but 17.2 makes that debate unnecessary. It makes it clear that there can be a challenge. However, it also makes clear that the challenge has to be made within 14 days. The “may” is permissive as to the making of the application, but the 14 days has to be complied with if the permission is to be used. Nothing else makes sense. Were it otherwise the effect of the rules would be to say: “The other party may apply to challenge the amendment, and can do so within 14 days, or after 14 days, either under this rule or under CPR 3.4(2), as it suits that challenger”. That is hardly rational. Nor does it make sense to allow CPR 3.4(2) to sidestep the 14 day provision, which must be taken to be there for a purpose. It is rational to allow 3.4(2) to have some part to play in abuse cases, but one should apply the 14 day period to it by analogy where appropriate. Of course, if it subsequently becomes apparent that a joinder is an abuse, then again 3.4(2) can be invoked, and it would be inappropriate to apply the analogy, but that does not mean that in the normal course of events the application need not, under the rules, be made within the 14 days.

It is not a bar to this conclusion that no sanction is specified under CPR 17.2. Mr Rabinowitz said that that was a pointer against the 14 day period being mandatory and in favour of its being some sort of vaguer target. I disagree. One does not have to have a sanction to make a time period significant. Time limits under the CPR will invariably have some significance, and require something to be done if they are not complied with, even though sanctions are not specified. They are seldom, if ever, optional. Mr Rabinowitz pointed out that in Qatar Saini J seems to have contemplated that there could be a 17.2 challenge in due course despite the fact that it would have been long past the 14 days (see paragraph 240), but I do not think that that is significant for the purposes of my decision. The question of whether the 14 days was optional or mandatory was not before him. The question of a 17.2 application was something for the future in that case.

That means that the application attacking the amendments made by G4S on a limitation basis was made out of time. It also purported to be made under arguably the wrong rule because of the terms of the application notice. However, I do not think that that latter point, of itself, would be a bar to entertaining the application. If it had been the same application and made within 14 days, apparently applying for the same relief as would be available if it had been couched as an application under 17.2, then I do not think that specifying the wrong rule would of itself matter. That would be at worst a procedural error which could be corrected under CPR 3.10. Mr Onslow accepted as much. The real problem for G4S is the delay.

I have held that the 14 days matters. That means that if Mr Rabinowitz is to be allowed to make his application he needs to get the equivalent of an extension of that time period. Mr Onslow submitted that what Mr Rabinowitz should have done is apply for an extension as if applying for relief from sanctions, on the basis of an implied sanction. He relied on the doctrine of implied sanction as it is described in the notes to CPR 3.9 in the White Book (para 3.9.15):

The doctrine was recently confirmed in R (Hysai) v Sec of State for the Home Department [2015] 1 WLR 2472.

Mr Rabinowitz did not challenge this as an exposition of the law. He said it did not apply in this case because the 14 day period in CPR 17.2 was not mandatory. I have already held that that view is wrong. It is a time limit which falls within the sort of provision to which the principles just expounded applies. Accordingly, in order not to be found out of time, G4S has to make, and succeed in, an application for relief from sanctions.

Having heard Mr Onslow’s first submissions on the point, Mr Rabinowitz has done just that. In case he is wrong, he has made an application for relief from sanctions. His evidence in support is short. Mr Bushell of Herbert Smith Freehills (solicitors for G4S) explains that his firm did not appreciate the ability to challenge the addition of the added claimants until after the 14 day period had expired, though he does not say when. When counsel was instructed in June Best Friends was identified as giving a basis on which the addition could be challenged, and that it seemed that a 4 month delay in that case was not a bar. It was decided to take that approach. Nothing is said as to whether CPR 17.2 was spotted and rejected as a route of challenge, and if so why, or whether it was not spotted. He then briefly urges the disproportionality of disallowing relief because the consequence might well be to allow in statute barred claims which should not have been allowed in under the Chandra principle, and he suggests that the claimants have suffered no prejudice from the delay.

Mr Onslow’s response is that the application is made late, especially since it was pointed out in evidence in September that an application for relief from sanctions would have been necessary for the purposes of 17.2 and it was not made then; in giving evidence of the reasons for not doing it earlier Mr Bushell has not been candid; prejudice to the claimants was not necessary and the application could fail on other grounds; and in any event significant prejudice had been caused to the claimants. He also complained that it was unattractive for G4S to be taking the technical points that it was taking in this litigation while at the same point trying to get itself off the hook for its own technical failure. I did not think that this last point was Mr Onslow’s best. He made further points under an “All the circumstances” heading to which I will come so far as I consider them to be significant.

I shall consider this matter under the now familiar three stage test in Denton v TH While Ltd [2014] 1 WLR 3926 - whether the default was serious and significant; why it occurred; and then considering where justice lies, considering the circumstances as a whole.

In considering the first point there are two elements - what was the default, and for how long did it subsist?

The first question arises because of the unusual development of the facts in this case.

So the nature of a 17.2 challenge was always apparent in the July application. That is confirmed by Mr Onslow’s concession (which was correct, in my view) that an early application to amend the application notice to add a reference to 17.2 could not have been resisted. The concession is right not because it would have been made early enough to be dealt with in the application, but because it would have made no real difference to the nature of the application, or required any different evidence.

That means that the relief from sanctions point should be treated as one geared to justifying the 8 weeks delay, because that is the real material delay in this matter. It is not as though Mr Rabinowitz was suddenly making a new application 6 months late; he is seeking justification for an application made 8 weeks late. I acknowledge that delay in making the relief from sanctions application is something that can also be taken into account, but that is more appropriately dealt with under the third head of Denton. In considering the first stage of seriousness of breach, the reality is more important, and I have just identified it.

With that in mind I turn to the first question of the seriousness and significance of the breach. The failure to make an application under 17.2 until 8 weeks after it ought to have been made was certainly significant. It cannot be dismissed as slight. The rule is there for a purpose, which is presumably to require prompt challenges to preservice amendments so as not to hold up the proceedings and achieve clarity at an early stage. 8 weeks is a significant, and indeed fairly serious (if that label matters) delay.

Turning to why the delay occurred, it would seem that G4S did not consider that an application had to be made within the time limit. It thought it had another way of dealing with the matter, which was thought to be rooted in Court of Appeal authority (Best Friends). Mr Onslow invited me to find that the failure to apply within the time limit was deliberate. I do not go that far. It is not credible that G4S’s advisers looked at 17.2, considered that they could use it but deliberately decided to do something else and then do it deliberately late. Whatever lack of candour there may be in Mr Bushell’s witness statement I do not consider it is covering up something like that. It would seem that initially, and in the 14 day period, G4S did not realise it had a right of challenge at all. It started to consider the point on about 8^th June and identified a case (Best Friend) where a challenge was made after 4 months and the amendment was said to be ineffective or had inevitably to be set aside. Attention was probably diverted away from 17.2 and/or the time limit within it. I do not consider that the failure to invoke 17.2, and to apply under 17.2, was in any relevant sense deliberate.

As to the second limb of Denton (why the default occurred), I have already substantially dealt with that. It seems that once the possibility of a challenge was appreciated and put in train, G4S proceeded down a different route, prompted by authority, while taking the same points (limitation) as it would have taken under an express 17.2 challenge. It was believed that no time limit was imposed by that route. On the supposition that an application under CPR 17.2 was required, this amounted to a mistake, not a deliberate act in defiance of the provisions of the rules. It is perhaps a slightly surprising mistake for an experienced legal team, but it is not a particularly weighty factor against relief.

Under the third limb Mr Rabinowitz stressed the disproportionate effect of not allowing his application to be made. On the assumption (for these purposes) that his client has a good point then a failure to allow the application would mean that his clients would be deprived of a limitation point that they would otherwise be entitled to run. The practical significance of that appears above - limitation could effectively bar the bulk of the claims made. That would be an excessive effect to attribute to a failure to make an application for 8 weeks or so. No significant prejudice had been suffered by the claimants, and any disruption to the proceedings from the late application, and the manner in which it was taken (as averred by the claimants) was over-stated. He also relied on what he said was the “obscurity” of the rules and the absence of authority on the effect of 17.2.

Mr Onslow started by setting out what he said was important background to this part of the case. He took me to correspondence which took place between the issue of the claim form in 2019 and its eventual service, and showed me how the claimants were seeking to engage with the G4S with a view to settling the claim, and G4S seemed to be interested in that process and sought information and clarification. I must say I fail to see how that went to the point in issue on the relief from sanctions application. Then he pressed the prejudice said to arise from the late application and the late taking of the 17.2 point. His clients had been proceeding first on the assumption that there would be no challenge, once the 14 days was passed, and then on the footing of the original challenge which was not mounted under CPR 17.2. Work was done to advance the case - for example, work to identify personnel at Invesco (one of the major claimant groups), work done to identify documents which needed to be preserved for disclosure purposes and work done to identify the status of various claimants once a status point was taken. Work was also done in this application on points that would not have arisen had the 17.2 point been taken in time. The litigation process has been disrupted by the addition of a considerable period of time devoted to the relief from sanctions application, and court resources have been unnecessarily extended. His own side had devoted resources to dealing with the question of whether CPR 17 applied at all on the application as mounted and the extra time taken in court has meant that junior counsel was not available on one day. It was important that the rules be observed and it was unattractive for the defendant to seek to avoid the technical and substantive requirements of 17.2 while taking a number of technical points itself and indulging in “inflammatory” statements about the conduct of the claimants’ solicitors, querying whether they had instructions or not and questioning the veracity of statements of truth. The original application, when made, was unheralded by any form of contact indicating or investigating the points that were to be taken and there was no attempt to see if matters could be shortened by discourse.

Much of Mr Onslow’s case under this third head seems to me to be overstated. As far as I could see, the absence of his junior did not seem to have any effect on his presentation of his case. So far as extra work is concerned, unless it were to be suggested that the claimants would have more or less immediately caved in under a timeous challenge under 17.2 it is not easy to see, in reality, what work has been done now that would not have been done anyway. It is impossible to conclude that the claimants would have given way promptly or at all, both because that is implausible in this sort of litigation, and because they do not say they would. In those deemed circumstances the defendant would have made its application earlier, and it would have been based on, and met with, the same points as now. The same work would have been done on the application, by and large, and it is not easy to imagine why any of the other work would not have gone on behind the scenes as well. If extra work has been done, then I do not regard it as very serious, and it can be allowed for in costs if relevant. I do not regard the other points relied on by Mr Onslow as particularly significant individually (especially the point about competing reliance on technicalities - it is invidious for the court to engage in a comparison of which party is being more desirably technical than the other), but I do take on board the need to comply with the rules. The case has been disrupted by the application for relief from sanctions, especially its being injected where it was, but overall I do not think that the situation is very different from what it would have been had the application been

I consider the most significant factor to be the proportionality point. G4S has a strong point when it says that the effect of not allowing it to make its 17.2 challenge late would be to deprive it of the benefits of the Limitation Act which it clearly seeks to invoke and has clearly sought to invoke since the application was made. I consider that a weighty matter. Of course, the claimants would be entitled to say that they would be deprived of the benefit of a rule which, if it works as they say it should, would rescue them from the clutches of the Limitation Act on the facts as they have turned out to be in this case, but I do not consider that factor to be as strong. In that context I bear in mind that the whole problem has arisen because of the obvious last minute rush to issue proceedings and gather in claimants, in the face of an obvious potential limitation date, and an apparent failure to get all the claimant’s ducks in pen, let alone in a row, when that could have been done some time before. That does not attract a lot of sympathy.

Taking all the above matters into account, and the other matters urged on me in the evidence and in submissions, I have come to the conclusion that the defendant should have the benefit of relief from sanctions and be allowed to make its application to challenge the 17.1 additions late, so far as necessary. I consider the application to have been made not too late, the reasons for not making it not too serious (albeit hardly commendable) and the other factors (particularly the effect on the defendant of failing to grant relief) combine to lead me to that conclusion.

Mr Rabinowitz also had a technical point. Under CPR 19.4(4) (set out above) a party cannot be added as a claimant unless it consents in writing and the consent is filed with the court. Mr Rabinowitz said that in this case there was no such consent in writing, whether filed with the court or not. The new claimants sought to rely on the statement of truth signed by their solicitor on the amended Claim form as being a filed consent, but that is said by the defendant to be not good enough because such a document is inadequate and in any event the signature is not that of the claimants themselves. He also disputed the submission of the claimants that CPR 19.4(4) did not apply to a joinder in respect of which permission was not required under CPR

What happened in the present matter is as follows. The original parties appeared in the claim form with a statement of truth signed by Mr Warren-Smith of the claimants’ solicitors. His signature was attached digitally, as is permitted by CPR 5.3:

No point is taken on that particular process, thus described, in this case in relation to the signature on the original claim form.

The evidence on what happened thereafter emerged in bits during the course of the hearing rather than before the hearing. On the various occasions when the claim form was subsequently amended, Mr Warren-Smith’s assistants were instructed to prepare the amendments, apply his electronic signature and submit it to the court for filing. The first attempt at providing that information suggested that the process of applying an electronic signature was the same as the original application - the computer was instructed to apply it at the end of the newly amended claim form. However, when it was pointed out that the last page of the claim form, bearing the signature, was absolutely identical in layout and positioning to the original claim form, with the signature at precisely the same point in the signature box on each page, it transpired that there was no fresh application of a digital signature (as some sort of digital process) each time. Instead Mr Warren-Smith’s staff just took the last page from the first claim form, appended it in toto to the new claim forms and submitted those for filing. That is not quite the same process as was first described, and it is a little unfortunate that the first explanation was not wholly correct.

G4S says that there is no written consent in this case. All there is is the claim form apparently bearing a signature of the claimants’ solicitor. That is said to be inadequate. First, it has been demonstrated not to be a signature for these purposes. All that has happened is that a page has been annexed, and that page is the equivalent of a photocopy of a digital signature of a solicitor applied to an original document. Second, in any event such a document cannot be treated as a demonstration of the consent of the would-be parties when one looks at the authorities, and principally Kay

The claimants do not accept that 19.4(4) has anything to do with pre-service joinder at all. Mr Onslow submitted that its position in CPR 19.4, and Practice Direction 19A, demonstrated that it applied to applications for joinder (which would be post-service) only. Pre-service joinders did not require this consent. If that was wrong then on the authorities the claim form, appropriately signed, could be a consent, and this one was appropriately signed by the solicitor and agent of the new claimants. More modern authority demonstrated a flexible approach to the concepts.

I will first dispose of the argument that 19.4(4) does not apply to pre-service joinder. In my view it plainly does. Its wording is general and on its natural construction applies to joinder whenever it takes place. Its position in the CPR does not affect that meaning. The heading to the rule shows it is general in its application, and the reference to a non-service situation in 19.4(1) does not mean that the rest of the rule does not apply to that situation; it tends to indicate that that situation is within the scope of the rule even if most of the subsequent provisions deal with a permission that is not required pre-service. Furthermore, there is no good reason in practice for excluding pre-service additions from the consent requirement. If a clear expression of consent is required to support an application post-service, it is impossible to think of a reason why that should not apply pre-service. The contextual argument advanced by Mr Onslow cannot overcome this, and there is nothing in Practice Direction 19A, also relied on as context by Mr Onslow, which assists him either. Accordingly, a filed consent in writing is required and it is necessary to consider whether the amended claim forms were meaningfully signed at all (so as to signify consent) and if so whether the solicitor’s signature on that type of document is sufficient.

So far as the first of those points is concerned (was there a meaningful signature), it has to be borne in mind that under CPR 19.4(4) the consent which has to be filed does not actually have to be signed, as Mr Rabinowitz conceded, though that will normally be the way of indicating consent in a document. However, in this case what is relied on as the indication of consent is a “signed” document. If it had not been “signed” then the additional claimants would not be able to make their case. So it is relevant to consider whether the reproduction of a signature relied on was actually a signature for these purposes.

No issue at all was taken about this when it was thought that what was applied was a digital signature applied in accordance with CPR 5.3. When it became apparent that the process used was not that allowed by that provision Mr Rabinowitz did indeed rely on the apparent shortcomings in the process, though he expressly disclaimed any allegation of impropriety. However, he did not really pursue the question of whether

I can therefore turn to the more substantial question of whether the signed amended claim forms can also stand as filed consents within CPR 19.4(4). The defendant’s point on this is that there is no written consent of the new claimants to their joinder, and certainly none filed. It relies heavily on relatively modern Court of Appeal authority (Kay v v Dowzall [1993] WL 13726011, applying the rather older case of Fricker v van Grutten [1896] 2 Ch 649). Those cases are said to make it clear that the written consent, which must be filed, must be in a document which can be said to be the new claimant’s (and not his solicitors). A claim form signed by the solicitor does not qualify even if (which is said to be unclear) a claim form signed by the claimant himself would qualify. The new claimants dispute that analysis. The case of Fricker involved an older rule which had the word “own” which made all the difference - “without his own consent in writing thereto”. More modern authority suggested a different approach - see TRW Pensions Trust Ltd v Indesit Cmpany Polska [2020] EWHC 1414 (TCC) - pursuant to which a solicitor’s signature would be allowable. Kay v Dowzall was a “seemingly obscure, unreported decision” which does not really deal with the absence of the word “own” from the more modern rule. When the rules required signature by the party himself/herself, they said so – see eg CPR 22.1(6). Other provisions require the written consent of parties in a context which cannot possibly mean that the party itself, as opposed to its solicitor, has to sign - see eg CPR 17.1(2).

In Fricker v Van Grutten the existence of a bankruptcy was thought to make it desirable to join the trustee in bankruptcy of an already existing plaintiff. The trustee apparently gave a form of consent, and the solicitor for the existing plaintiffs, who was to become the trustee’s solicitor as well, endorsed the summons seeking permission to amend to add the trustee with a statement of consent on behalf of the trustee, but signed by the solicitor. It was held by the Court of Appeal that that was not a consent within the then existing rules. The existing rule was similar to the present rule, save that it had the word “own” in it as appears above. When things took a turn which led to the trustee being liable for costs he challenged his addition on the basis that there was no written consent by him filed with the court. The Court of Appeal acceded to his submissions.

Lindley LJ said the point was important “on the question of practice” (p654) and held that the signed endorsement of the solicitor was not sufficient. At pages 655-6 he said:

He held that the joinder ought not to have been made because there was no consent in writing. Lopes LJ and Rigby LJ agreed. Rigby LJ said:

That reasoning was applied in Kay v Dowzall. By then the rule had been amended to read:

The word “own” had thus been omitted. Nonetheless Sir Thomas Bingham MR held that that made no difference and gave the new words the same effect as the old. He (and the other members of the court) held that a deed of appointment of new trustees, which anticipated their participation in the proceedings and which was signed by them (and indeed sealed by them) did not, in its terms amount to a consent within the rules. In arriving at that conclusion the Master of the Rolls considered the rule and said:

Then he considered Fricker and went on to consider various possible distinctions between that case and his, and dismissed them all as making no difference. Among the distinctions that he considered was the absence of the word “own” from the rule then under consideration (as in the rule which I have to consider):

Then he found that the deed relied on did not contain the relevant consent, on its true interpretation.

That would seem to cover the present situation, and Mr Rabinowitz says it does. Mr Onslow seeks to distinguish the decision first on the basis that it is unreported and obscure. That objection fails. The English law of precedent does not allow for cases to be distinguished on either basis. For what it decides it binds me.

It is therefore necessary to determine what the ratio of the case is. The first thing to note is that the wording of the rule before the court, like the present wording, differed from the wording in Fricker because of the absence of the word “own”, whose presence was regarded as significant in Fricker. Mr Onslow regarded that as a significant difference. I agree that it is potentially significant, but the Court of Appeal clearly considered that it made no difference. In the extracts just cited, the court rejected the submission of counsel that it was a relevant distinction. So there is a decision of the Court of Appeal which seems to support the proposition that a solicitor cannot sign for his/her client in these circumstances.

The next point, technically, is whether that is part of the binding ratio of the decision. Looking at it carefully, I do not think that it was. The document relied on as being a consent in that case was one which was signed by the client himself. It was not necessary to decide whether a consent could be signed by someone other than the new “party”.

However, that does not mean to say I should treat myself as completely free to decide the question on the basis of the modern rule, from which the word “own” has been dropped. Left to my own devices, I am not wholly convinced that a signature by a solicitor should be insufficient. To say that it is not would be tantamount to saying that a document signed and verified by a duly authorised agent would not be sufficient. If that is right then one has to ask how a corporate body is supposed to mark its consent. It can only do so by a duly authorised person, who for those purposes must be an agent. If a corporate body can use an agent, why cannot an individual? And in the present case most of the would-be claimants are corporate bodies, so refusing to allow them to sign by one agent (a solicitor) but allowing them by another (an officer or director) might be thought to be inconsistent.

Having said all that, I do not consider myself as being free to decide the point in favour of allowing a solicitor to consent. The decision of the Court of Appeal is strong and firm, and it approves a historic practice. That practice is said to be rooted in principle - the need to be quite sure that a party wishes to be joined. I do not consider it to be appropriate for me to depart from what was said by the Master of the Rolls (and thus Lindley LJ).

It does not appear that Fraser J proceeded on the same basis in the TRW case relied on by Mr Onslow. His case did not really turn on this point, but he did deal with it in two short paragraphs of his judgment (paragraphs 53 and 54). He decided that apparent consents from first the Financial Controller of the group of which the claimant formed part, and then the group’s General Counsel, were sufficient consents. However, he also observed that the claimants’ solicitor gave what amounted to a consent:

With all due respect to Fraser J, I do not consider that that is a sentence which I should follow. The content of the two paragraphs in which he deals with the point

In the circumstances I do not consider that the consent said to be impliedly expressed by the solicitor who signed the claim form can count as a consent under 19.4(4).

However, there is a further reason why, on any footing, the claim form should not stand as a consent even if a solicitor could sign one for the client. In my view the wording of CPR 19.4(4) requires a separate document from the sort of pleading that a new claimant would inevitably have to sign anyway when he/she is added (someone would have to sign an amended claim form for them). In my view, what the rule, and the reasoning behind it as expressed in the Court of Appeal cases, requires is a separate document which is filed for the purpose of expressing the consent. The filing has to take place before the addition as a party. That can logically only be done in a separate document before the addition which takes effect via an amendment. The amending document itself (here, the claim form) cannot achieve that function. It may be that a prior document which achieves the purpose of expressing consent, but is filed for a different primary purpose (such as the witness statement referred to in TRW) could accidentally (or incidentally) have the same effect, but I do not need to decide that. What seems to me to be clear enough is that a separate consent document has to (a) exist and (b) be filed, and the claim form introducing the new claimants does not qualify.

There is no other consent expression which is relied on by the would-be claimants in this case, so the provisions of CPR 19.4(4) are not fulfilled. There was reference at the hearing to some sort of behind the scenes documents available to the claimants which indicated consent, but they have not been produced and they have certainly not been filed. They therefore do not save the additional claimants in this matter.

Drawing all those elements together I therefore conclude:

Joinder of the added claimants could not take place under 17.1 on its true construction.

If that is wrong, then the defendant should be allowed to mount a challenge under 17.2.

That challenge succeeds because there is an arguable limitation point which operates in relation to all the added claimants.

In any event, the joinder fails for want of the filing of a prior signed consent.

Accordingly, the added claimants all fall to be struck out.

As before, that means that practically issues about name changes do not arise because those whose names the claimants would wish to change are struck out as claimants. However, the point arises in relation to a very small number of the original claimants, and in case this matter goes further I shall consider the matters advanced to me on the unidentified claimants point.

I use the description “unidentified claimants” because it was used in argument before me, and in using it I do not intend to prejudge one of the points that arises under it, which is whether certain claimants are indeed unidentified. It is not a wholly apt description of the problem, but it will do.

The point arises in this way. It turns out that a large number of the claimants (including all but 16 of the added claimants) are not entirely accurately described in the amended claim form. The defendant challenged the status of a large number of the claimants and sought to strike their claims out on the basis that the proper claimant (if any) was not properly identified. The claimants did not necessarily accept all the criticisms made but met the defendant’s application with their own application to amend the names of the disputed claimants (or most of them). This was done outside the limitation period, and the debate before me therefore centred around whether claimants should be allowed to amend to carry out the name changes which they seek outside the limitation period, in the context of the claimants’ applications rather than the defendant’s.

The numbers involved are very extensive. The defendant says there are 56 claimants who are “unidentified” in the sense that the current descriptions of them are said not to refer to identifiable entities as things stand. 36 of those are also additional claimants. I have disallowed their joinder on other grounds, but since I have heard extensive debate on this point I shall decide the point in relation to them even though it is academic if I am right in my earlier decisions. The assumption for these purposes is that they were properly joined in the first place.

I did not hear argument on all 56 unidentified claimants. It was said that the cases can be divided into various different categories. At my insistence the argument took place by reference to examples (or exemplars) which were said to be illustrative of each category of mistake, rather than arguing all 56 or so cases. It is hoped that a decision on those examples can be applied to other instances in their respective categories.

It will be useful to give some examples of the proposed amendments taken from the examples at this stage in order to provide some context for the discussion of the principles involved:

CPR 17.4 and CPR 19.5 are set out above, but I set out the material parts again here:

the relevant limitation period was current when the proceedings were started; and

the addition or substitution is necessary.

the new party is to be substituted for a party who was named in the claim form in mistake for the new party; or

the claim cannot properly be carried on by or against the original party unless the new party is added or substituted as claimant or defendant; …”

The editors of the White Book (2020 Edn) have suggested what the difference is between 17.4 and 19.5(3)(a):

Stuart-Smith J accepted that that rationalisation was indeed correct in Jalla at para 112.

The issues to which reliance on all these provisions gives rise are the following:

The nature of the mistake.

Whether there is reasonable doubt as to the identity of the party in question

The extent to which that is relevant to a claim under 19.5(3)

The extent to which the operation of 19.5(3)(b) can be applied to the instances to which the claimants seek to apply it.

The date on which the test has to be applied - whether it is the date of issue or some other date.

Some of those issues have been the subject of consideration on the authorities, and I shall set out some of the relevant law before considering other issues in the particular factual context in which the provisions are sought to be applied.

There has been a fairly large number of cases on this topic, no doubt reflecting the difficulty of implementing some of the concepts involved. I can confine myself at this stage to a handful of them. Adelson v Associated Newspapers Ltd [2008] 1 WLR 585 reveals that the legislative history behind these provisions means that one is entitled to look to authorities under the old RSC, and in particular RSC Ord 20 r5 in understanding those provisions of the CPR (see para 26). This justifies a reference to a pre-CPR case as the starting point. It draws an important distinction betwen a mistake of nomenclature and a mistake of identification.

The Sardinia Sulcis case referred to in Adelson is a shipping case reported at [1991] 1

He went on to observe that the concept of “the identity of the person intending to sue” was not always an easy one to grasp but elucidation was available from the authorities:

The last three sentences of that citation are the most significant. Stocker LJ provided the following elaboration:

The approach in this case was approved and explained in Adelson, in which Lord Phillips CJ delivered the lead judgment of the court:

Having considered the authorities, Lord Phillips summed up the position as follows:

In paragraph 55 Lord Phillips identified the person who has to make the relevant mistake and summarised the authorities that he had considered:

The nature of the mistake required by the rule is not spelt out. This court has held that the mistake must be as to the name of the party rather than as to the identity of the party, applying the generous test of this type of mistake laid down in The

Almost all the cases involve circumstances in which (i) there was a connection between the party whose name was used in the claim form and the party intending to sue, or intended to be sued and (ii) where the party intended to be sued, or his agent, was aware of the proceedings and of the mistake so that no injustice was caused by the amendment. In the SmithKline case [2002] 1 WLR 1662, however, Keene LJ accepted that the Sardinia Sulcis test could be satisfied where the correct defendant was unaware of the claim until the limitation period had expired. We agree with Keene LJ’s comment that, in such a case, the court will be likely to exercise its discretion against giving permission to make the amendment.”

It is plain from the analysis of the facts at paragraphs 67 to 74 that the mistake, and the person making it, must be demonstrated by evidence, and cannot be left to surmise and inference.

In Best Friends Simon LJ summarised the position under CPR 17.4(3):

In Jalla at paragraph 114 Stuart-Smith J observed that under CPR 19.5(3)(a) there was no express requirement that the mistake would not have caused reasonable doubt as to the identity of the party intending to sue, but that point may be relevant to the court’s discretion.

In Insight Group Ltd v Kingston Smith [2014] 1 WLR 1448 Leggatt J reviewed the authorities in this matter and pointed out a number of difficulties in the concepts involved, and in the difficulties in deciding whether a mistake should be classified as a mistake of identity or a mistake of name, and the consequential difficulties in applying The Sardinia Sulcis - see paragraphs 28 to 42. I completely and respectfully agree with what he says there. However, like Leggatt J, I should apply the Sardinia Sulcis test - see his paragraph 43 - even though it still has the difficulties which he describes. He identifies at paragraph 46 that one has to try to identify the limits to the “generosity” of the Sardina Sulcis test, and came to a conclusion at paragraph 52:

From these authorities I derive the following conclusions for the purposes of the points I have to decide:

Under both 17.4 and 19.5, the mistake must be as to name and not identity.

19.5 refers in terms to a substitution. However, in reality 17.4(3) has also been interpreted so as to allow what is, in fact (and law) a substitution.

That is because the concept of a mistake as to name is interpreted generously.

Generosity is achieved by looking to the description of the legal requirements for qualification as the claimant or defendant (as the case may be) - Insight at paragraph 52 - usually as described in the claim form (and perhaps Particulars of Claim if served with it).

If a description is to be relied on as saving a misdescribed party it must be sufficiently specific to allow identification in the circumstances - “more or less specific to the particular case”, in the words of Sardinia Sulcis. A successful amendment will very often be a case where there is an intention to sue in a certain capacity (landlord, tenant, shipowner).

The true identity must be apparent to the litigation counterparty, at least under 17.4(3) (Adelson para 43). It is not clear to me why this would be a requirement under CPR 19.5(3)(a) when it seems to omit the reasonable doubt criterion.

Under CPR 17.4(3) it is a requirement that the mistake would not have caused reasonable doubt as to the identity of the party intending to sue. That is not a requirement under CPR 19.5, but the point may be relevant to the court’s discretion, and may be a significant factor. Mr Onslow accepted that it was capable of being relevant to discretion. I confess that it is not wholly clear to me how it is likely to play into discretion, but I suppose it is relevant to consider it as a test for whether the counterparty in reality knew in substance who the proper claimant/defendant was supposed to be. If they did then there might be more of a case for allowing the amendment, though I confess I do not find this wholly logical.

The level of generosity is demonstrated by a large number of the reported cases, but I can just confine myself to just one. In the TRW case (above) there was a claim by what was intended to be the tenant under a lease to one of a group of companies. After proceedings were issued, and after the limitation period had expired, the claimants discovered that the actual leaseholder was a different company from that which had originally sued. The wrong entity had been listed on the claim form because of a mistaken belief by the claimants’ solicitors as to the identity of the lessee, and the mistaken belief came from a description given by loss adjustors. Fraser J allowed substitution under CPR 19.5 on the basis that there had been a relevant mistake. The company which was joined as claimant was the company that the mistake maker (the solicitor) intended to be joined. There was no mistake as to the identity of that company. There was no mistake as to the name of that company. His mistake was in thinking that it was the tenant company. Nonetheless this was treated as a mistake as to name within the generous test.

That decision, and other more historical ones which follow the Sardinia Sulcus test, are a helpful background in my considering the mistakes in this case.

According to Adelson, the person whose mistake is relevant has to be:

One might have thought that that would mean the individual within the client who authorised the proceedings. However, that would be too narrow a view. In TRW it was, on the facts, the solicitor who was responsible for the litigation. There is no indication in that case that anyone in the client companies made a mistake at all. The source of the mistake was in the casual use of names by the loss adjuster, and a misinterpretation by the solicitor. That was sufficient for Frasier J. I would respectfully agree with that approach of allowing that mistake to count. It does not seem to me that too nice an inquiry into who made the mistake is going to matter much, because at the end of the day if there is a mistakenly joined party that will be down to the solicitor who will be mistaken as to the party who should be joined. He may have made the mistake himself without a contribution from anyone else; or he may have made it because of something he was told (as in TRW, and as in BDW Trading Ltd v AECO Infrastructure etc Ltd [2020] 10 WLUK 206). I do not see why the rule requires a particularly strict approach to this inquiry. In my view it is more important to identify the nature of the mistake, though that will of course involve identifying who made it, and obviously the mistake must have been causative of the error in question.

This provision was advanced by Mr Onslow in a manner which suggested it was a sort of afterthought or alternative, but it was advanced so I have so deal with it. It makes substitution necessary (and therefore justifiable) if “the claim cannot properly be carried on by or against the original party unless the new party is added or substituted as claimant or defendant”. There must, in my view, be some limit on how this operates, because if taken too literally it can be invoked by any claimant who demonstrates no more than that a substitution is necessary to save the action, which will be every case where there is a problem.

In Irwin v Lynch [2011] 1 WLR 1364 an administrator brought a claim against directors of a company, making the claim in his own name when it ought to have been brought in the name of the company. He applied, after the limitation period, to substitute the company as claimant. On appeal he was allowed to do so. Lloyd LJ, giving the lead judgment, did not feel he needed to decide whether the provision

He went on to find that the substitution should be allowed because what stood in the way of a properly formulated claim was the substitution of someone who had locus standi (see paragraph 26). The cause of action was identical, though with respect I find it hard to treat this as a special or different qualification for 19.5(3)(b) because it will be true of most substitution cases.

In AIG Europe Ltd v McCormick Roofing Ltd [2020] EWHC 943 Mr Roger ter Haar QC, sitting as a deputy judge of the High Court, held that the provision operated so as to allow the substitution of an assignee of a cause of action in circumstances in which the action had been started in the name of an original corporate owner of the cause of action which has since ceased to exist and whose assets had been transferred to the would-be claimant. The deputy judge decided that it was immaterial to the application of the provision that the original company had ceased to exist at the date of the commencement of proceedings, and that the case fell within the wording of 19.5(3)(b). This seems to have been a literal application of the rule. He went on to find, in the alternative, that the amendment also fell to be allowed under 19.5(3)(a). He does not seem to have considered what, if any, limits there are to the apparent literal application of sub-paragraph (b).

In TRW Fraser J considered that 19.5(3)(b) did not apply in the case of a mistake. He held that 19.5(3)(a) and 19.5(3)(b) were mutually exclusive or alternatives (see paragraphs 49-50).

In Insight Leggatt J considered the provision. Having considered the authorities of Irwin v Lynch and Parkinson Engineering Services plc v Swan [2009] EWCA Civ 1366 (where a liquidator was substituted to a claim originally brought by his company), he held:

It strikes me that there is a curiosity about all this. If it operates as Mr Onslow would have it operate, and indeed in line with the rationalisation suggested by Leggatt J, it would make the other provisions, and the need to analyse the type of mistake, redundant. The test of whether the same cause of action is advanced would provide a much easier test for a would-be claimant than a quasi-philosophical debate about the nature of the mistake. It would seem to me that the claimant in TRW could fulfil the test - the cause of action was the same, they just found the wrong claimant. With all due respect to Fraser J, I do not see why the presence or absence of a mistake should be a touchstone for the operation of 19.5(3)(b). It would also seem to provide a solution to the cases where an assignment was overlooked, or where the effect of a corporate merger on the whereabouts of a claim was not appreciated, resulting in the joinder of the wrong (sometimes non-existent) claimant. In all those cases it seems to me that it can be said that the same cause of action is being advanced, but it cannot be properly advanced without the joinder of the newly discovered “correct” claimant. It seems to me to be unlikely that this was the intention of this apparently more focused provision. Nor do I think, with respect, that knocking mistake cases out on the basis of discretion, as suggested by Lloyd LJ, is a satisfactory way of trying to reconcile the provisions.

It seems to me that insufficient attention has been paid to the word “properly” in the provision. While it is not possible to define its precise effect, it seems to me that it is intended to correct errors of the kind in Irwin and Parkinson which are in the nature of locus standi errors. The word would be unnecessary if the provision were to have the broad effect which Legatt J’s analysis would give it. An interpretation along these lines is the interpretation which I would prefer, and it is not inconsistent with the Court of Appeal authorities which bind me.

This is a gateway requirement for CPR 17.4, but not for 19.5(3)(a), though, as observed above, it may be a discretionary factor in relation to the latter.

In the present case a question arises as to the point of time at which this falls to be applied. G4S submits that the relevant time is the date of the mistake. This will be the date of the claim form or, in the case of a claimant added later (assuming, contrary to my determination above, that this can be done) the date of the amendment adding them. The claimants say that the date is the date of service or, if earlier, the date on which a defendant inspected the claim form on the court file. It matters in this case because after issue, but before service, there was a lot of correspondence to which the claimants point as making it apparent who the party was intended to be in each case. This correspondence is referred to in the next section of this judgment.

Mr Rabinowitz argued from construction and principle, pointing out there was no authority dealing with the point. Mr Onslow very fairly pointed out that arguably the decision in Best Friends was against him because in that case the claim form was amended before service, so by the time it was served the mistake would no longer have confused a defendant, yet the judge’s finding of reasonable doubt was upheld (see paragraph 40). However, as Mr Onslow pointed out, the point was not argued, and by then the court had found that the amendment was a nullity anyway. Accordingly, the point has to be considered without the benefit of authority.

Mr Rabinowitz started with the language of the provision. He pointed out that the description of the mistake as being (i) a genuine one and (ii) not one which would cause reasonable doubt, suggested that those two factors were to be tested at the same time. I agree with him about that. I do not, however, agree with his next point, which is that the Limitation Act 1980 generally focuses on the date on which a claimant takes the crucial act of issuing proceedings, not a later date at which a defendant becomes aware of the action, so the date of issue has to be the correct date for present purposes. I do not consider that this factor assists him; it is not necessarily relevant to section 35, which is the section which empowered the enactment of CPR 17.4 and CPR 19.5. Third, it was submitted that this interpretation better reflected the apparent objective nature of the test, and fourth that the alternative interpretation would enable a claimant (if one is looking at a mistake by a claimant) to adopt the unattractive position of delaying service and informing the defendant of the error before service, thereby unjustifiably and irrationally avoiding the adverse effects of the requirement.

Mr Onslow disputed the linguistic point and said that the Limitation Act point did not work (on the latter of which I agree). He then further submitted that the adoption of an objective test for the point, removed from reality, was artificial. As to the possibility of correction, he submitted that that did not assist because it was not a compelling point - the claimant still had to satisfy the Sardinia Sulcis test.

In my view Mr Rabinowitz is correct about this. The reasonable doubt element is a measure of the quality of the mistake and closely associated with it. The linguistic connection is close, and is reinforced by the likely purpose of the test. It is intended to be a test of the mistake, in its context. It does not make much sense to me that its inherent quality as a mistake can be altered by an explanation given to the defendant (or other counterparty) after the event of the mistake. Mr Onslow’s submissions can be tested by taking two scenarios in relation to a mistaken claimant. First, a scenario in which the test is not fulfilled at the time of the issue of the claim form and is still not immediately before service. Immediately before service the claimant explains the mistake and says it is going to serve and make an application to amend to correct it. Second, a scenario in which the requirement is not fulfilled at the date of issue, the claimant then serves, and immediately afterwards it points out the mistake and says it is going to amend. If Mr Onslow is correct then the requirement is fulfilled in the first case but not the second. But in my view the difference between those scenarios is immaterial. There is no good reason why the clarification given by the claimant (which makes the mistake apparent and removes confusion from that point) should assist the claimant in the first case but not the second. It is more rational to test the matter as at the date of the mistake and its operation (the issue of proceedings in the wrong name), after which point the clarification by the claimant can make no difference.

I shall therefore approach any 17.4 arguments on that basis, and insofar as the factor might be relevant to the discretion under 19.5(3)(a), I shall apply the same reasoning.

I can now deal with some of the general evidence in this case which goes to the points of law considered in the two preceding sections.

In order to make good its claim that the need to amend arises out of a relevant mistake on the part of the claimants they have filed evidence from their solicitor Mr WarrenSmith. It will be necessary to give more details about the source and nature of the relevant mistakes when I come to consider the actual examples, but Mr WarrenSmith also provides some background which is applicable generally.

In his second witness statement (paragraph 25) he says that the claimant group companies (most of them in the Invesco group) operate complex structures and have had to interrogate their records going back many years, and it has been difficult for them to reconstruct their data in some cases. In each case where there was an error, the relevant claimant had always intended the correct shareholding party to be named as a party to the claim, as had he and his firm. In the case of some claimants, he had been given to understand by the claimants in question that they were separate legal entities, but he came to understand that they were not - they were sub-funds without separate corporate identity. He does not identify the sort of mistakes further, or identify who made them.

His third witness statement gives a little more information about the source of the information as to claimants. In it he explains that the majority of the claimants in the action are represented by “intermediaries (both lawyers from other jurisdictions, and securities and governance professionals)”, and where that is the case his firm takes instructions from them. His firm relied on those intermediaries to provide the correct name of the shareholding entity which should be named as the party to the action. If there were doubt, his team would raise a query. His over-arching point was that at all times “my firm intended to include the correct Claimant and the name of the entities which owned the shares”. If intermediaries were not acting, then his firm took instructions from the claimants themselves. The proceedings were issued under the pressure of limitation.

His fourth witness statement elaborates further, to an extent. So far as naming subfunds as claimants as opposed to the actual owning entity is concerned (a common feature in this case), he says that the names of the sub-funds were provided by the Invesco group asset management teams. The names of the sub-funds were provided to his firm in response to a request for the names of the shareholding entities. He says:

The information from Invesco on which his firm acted was provided to him by the legal team at Invesco Asset Management Ltd to whom management of all sub-funds was delegated. There was:

At paragraph 18 he sought to justify the mistake:

For present purposes it can therefore be noted that the source of this sort of mistake was the relevant claimant (or its group representatives) itself. No further particulars than that are provided (by and large). Their solicitors, not surprisingly, acted in accordance with what they were told. It can also be observed that the mistake arose because the claimants did not understand their own holding structure and what their corporate entities were. That is less than impressive.

Further details of the mistake emerge in the consideration of the individual examples appearing below.

There was also evidence of information provided to the defendant before service which is said to go to the resolution of the unidentified claimants point. There was no letter before action in this case. The claim form was issued on 10^th July 2019. It was first notified to G4S by a letter from the claimants’ solicitors (Morgan Lewis & Bockius - “MLB”) dated 7^th October 2019, at which point of time only a small number of amendments (to add parties) had been carried out. It was sent purportedly pursuant to the relevant pre-action protocol so that G4S “could understand the basis of [the] claim”. It sought an extension of time for service of the claim form, which it enclosed but not by way of service, and it sought a lot of information germane to the general cause of action in order to remedy what it said was an evidential imbalance.

On 7^th November Herbert Smith for G4S declined to provide information but in turn sought information about “each claimant’s shareholdings” and how the loss was calculated, and said G4S would consider whether to join a settlement process when draft particulars had been served. Time for service of the claim form was extended to 17^th January 2020.

On 15^th November MLB sent draft Particulars of Claim which were said not to be final because work was being done on them. The draft Particulars had a list of claimants (I am told all the current claimants) under their unamended names, and without showing additions that had taken place over time. In response to the request for information about shareholdings and the calculation of losses, the solicitors also enclosed a table showing the number of shares held by each of the claimants (again described by reference to the names which can now be seen to have appeared in the claim form) on various dates and the estimated losses per client (all subject to further revision). Herbert Smith responded on 11^th December by complaining that the information provided did not enable them to assess the loss and requested that the share trading activity of each client be particularised. They indicated that they might be amenable to settlement if the information was provided. They pressed for the same information in a letter of 24^th December 2019. Mr Onslow drew attention to the fact that Herbert Smith did not express themselves as being confused about the identity of the claimants.

On 14^th February 2020 MLB provided share transaction data for 31 (out of 151) claimants, with more detail to follow. The relevant claimants were listed in an Appendix. So far as G4S complains about errors in naming now, the same errors appear in that Appendix. On 19^th February MLB further said there were difficulties in getting some of the details together but they were “close to finalising the data for a large number of Claimants and will provide that as soon as we can on a rolling basis”. The letter pointed out that G4S would have its own data about historic shareholdings and could undertake its own analysis of loss from that. A further extension of the date for service of the Claim form was proposed (18^th May 2020). On 21^st February MLB provided revised data for some shareholdings. Further share transaction data was provided on 26^th February for a further group of claimants (again using the same names as appear on the claim form). On 6^th March Herbert Smith again complained about the quality of the data, saying it did not have any meaningful information about reliance, and it agreed a pro tem extension of time for service to 31^st March 2020. Over the course of the ensuing correspondence that time was extended to 30^th April. When the settlement suggestions ran into the sand the claim was finally served on that date.

The significance of this evidence is said to be that it provided the defendant with information about shareholdings and losses which would have enabled it better to identify who the claimants were where there might otherwise have been a problem with that. During the whole of this exercise G4S never suggested any confusion, difficulty or doubt about that. I was invited to infer that G4S used the information for its stated purpose (verification of losses) which would have involved the identification of shareholders. This is said to go to an argument as to the identifiability of the mis-named claimants.

On the state of the evidence in this case, and bearing in mind the manner in which the case has developed, I consider that great caution is required before drawing the sort of inference the claimants invite for the purposes for which they invite it. The evidence demonstrates this sequence. The defendants were told a claim had been made by shareholders identified in the schedule to the letter after action. So far as actual claimants were concerned, they will have been identified by names many of which will have been erroneous. The schedule cannot have contained all the current names because the claim form schedule was amended at least twice after that. The draft particulars of claim were provided and at that point all the names will have been apparent to the defendant in their correct (where correct) or erroneous form, as the case may be. Then over time details of some of the trading activities were provided in order to enable some sort of assessment of loss. They were provided in stages. The erroneous names were no more accurate then than they are now. Errors in the data were corrected on 21^st February 2020. The claimants themselves apparently had difficulty in piecing together their own data, according to the correspondence (see eg an MLB letter of 5^th March 2020). The details provided came in two forms. First, there was a table which averred loss for each of the claimants as at 3 particular dates. Second there were developing (as to the number of claimants dealt with) spreadsheets in which detail was provided as to holdings of shares and (according to the tab name on the spreadsheet) transactions in the shares.

If it is suggested that G4S could have identified the actual claimants from that information then I find against such a suggestion. The claimants themselves were still propounding erroneous descriptions, and it is not apparent how this information would have enabled the defendant to do better than the claimant. There is no evidence that some register would have contained the correct name, and the evidence was that for the years 2011 and 2012 registers were not available to the defendant anyway. I can see no basis for saying that the defendant could have made a better fist of identifying the correct claimant than the claimants themselves.

It is true that at a lower level of specificity the information provided might have enabled the defendant to find out, if it wished to do so, whether an entity with something like the name referred to held shares as alleged in the claimants’ data. I agree that the information must have been provided so that the defendant could see whether there was at least a plausible claim, though there is no evidence as to what the defendant actually did. But for present purposes I do not consider that the information was any more than a particularisation of the general allegation that the claimants (as identified) were said to be shareholders who bought and/or held shares in the relevant period. It does no more than that.

The issue of the status of certain funds and sub-funds, and the corporate personality involved or not involved, is a factor which is said to be relevant to a number of the original namings and attempts to amend. It is therefore useful to deal with the underlying position before considering the exemplars to which it is relevant.

A number of the investment vehicles in this case were structured as Open Ended Investment Companies - OEICs - under the Open-ended Investment Companies Regulations 2001. They are FCA authorised collective schemes. For present purposes they can be treated as incorporated.

An OEIC can be (but does not necessarily have to be) structured so as to have its assets held in segregated sub-funds. Where such sub-funds exist and the assets are pooled separately, that pooling operates (under the Regulations) so that the liabilities of one sub-fund cannot be used to discharge the liabilities of another fund, and the property of one fund can be subject to the orders of the court as though it were a separate legal person even though in law it is not such a person. This is achieved by Regulation 11A(1)-(2) and 11A(5), the latter of which reads:

A sub-fund may be wound up by the Court as if it were an OEIC and for that purpose shall be “treated as though it were a separate legal person” (Regulation 33C(1)-(4). This is reinforced by Regulation 33C(4):

Thus for the purposes of asset and liability allocation a sub-fund is treated as if it were a separate entity. However, it is not in law a separate entity, and those provisions do not make it such. Mr Onslow submitted that sub-funds had a certain measure of legal personality. I am not at all sure that that is an accurate summary, but in any event their status as being not entitled to sue in their own name (leaving aside what might happen in a winding up) is apparent from the absence of anything giving them that right, and the presence of Regulation 11(A)(6):

Thus the Regulations make even clearer what would otherwise be clear enough, that in proceedings it is the umbrella company (OEIC) that has to sue or be sued.

The exemplars in this instance are items numbered 34 and 39 in the numbering in this case. The unamended names were Invesco Income and Growth Trust and Keystone Investment Trust. In each case the abbreviation “plc” ought to have been added in order to express their legal personality properly, and an amendment is sought to add it.

This is said to be a 17.4 case - Mr Onslow seeks to say this is a mistake as to name. As with all the other entities sought to be amended, there is no evidence specifically directed as to how this mistake in particular came to be made. Mr Warren-Smith’s evidence is that his firm relied on the claimants and their intermediaries to supply the names of claimants (he gives the name of two individuals as the client). It is said that

That is a slightly surprising reason, but it has not been challenged, and the evidence describes a mistake. It was made by someone responsible for the claim form in this case, in that it was the mistake of the person providing the information to the solicitors, and of the solicitors who followed through and acted on what the client told them. In my view it is a mistake which qualifies, unimpressive though the conduct is - one would have thought that the question of legal personality would inevitably have come to mind when one sees some “bodies” without any trailing “plc” or “Ltd”.

The main attack by Mr Rabinowitz was on the basis that there was and would be reasonable doubt as to the identity of the party claiming. Starting with the Invesco entity, he points to the large number of Invesco entities, under their descriptions in the various versions of the claim form, and says that there would be reasonable doubt as to which entity was intended.

Mr Onslow counters that by saying there could be no reasonable doubt because the status of all entities, and therefore their identities, are clarified by the description of all of them in the claim form as shareholders who acquired or continued to hold shares in G4S at the relevant times. It is significant that until these applications were mounted neither G4S nor its solicitors ever expressed any difficulty in understanding who was suing them, and that included the entire period when the claimants were providing details of shareholdings for the purposes of settling the matter before actual service of the claim form. G4S had access to its shareholder register, at least for the period from 2013 and it is to be inferred that it consulted that register and was not confused; if it did not consult the register it could have done so and resolved any doubt.

It can be seen that Invesco companies started to emerge in the evolving claim form in the first amendment of 28^th August 2019. Five of them, all bearing the word “Limited” appear there, but not the Invesco-named entity under present consideration. More Invesco-named “claimants” emerged in another version of the schedule to the claim form which listed claimants on 28^th August 2019, including Invesco Income Growth Trust. They are numbered 62 to 106. None of their names give any clue as to their legal personalities. All start with the word Invesco.

In line with my ruling above that the question of reasonable doubt has to be decided at the date of the amendment (the equivalent of the date of issue for this claimant added by amendment) then that context is significant. As at that date there had been no communication about the case between the claimants and G4S. There was therefore

So far there would be considerable doubt as to the identity of the proper claimant. The only additional factor which might tilt the scales away from doubt is the possible fact that the proper identity of the claimant might be apparent from the share register of G4S. It is not known whether the share register reflected the proper name. The claimant would not know. The defendant has the knowledge within its power and has not provided the information, despite Mr Bushell for the defendant making a general reference to the number of similar names on the register. It would be fair to assume for these purposes that the full name of this entity might well appear in the register. However, even allowing for that I do not think that this removes the reasonable doubt. This is one name in a sea of not dissimilar names (which could be seen to have grown over time) and in that particular context it is not possible to say that there is no reasonable doubt as to the intended identity. That conclusion is even stronger when one looks to see the other potentially confusing names, and names which in some cases can be seen to be nothing like an appropriate name.

I therefore find that this particular claimant does not surmount that hurdle.

If my conclusion on the date at which the point is tested is wrong, and the relevant date is the date of service, then my conclusion would be the same. By this time G4S had received a certain amount of listed transactional information in relation to all the claimants, or at least in relation to this one (under its erroneous name). However, there would still be nothing obvious to link the given name to the corporate entity whose name contains “plc”. The name would still be one of a number of names many of which could not be linked obviously to a given real entity. What has troubled me at this point in the reasoning is whether the share register had the full name of the entity which could link it to its shareholding, details of which had now been given, and thus remove any doubt as to who the claimant was. As I have pointed out, it is not known what the share register showed, and G4S has not said whether it consulted the share register for any or all of the large number of claimants. However, again I do not think that that matters. The resolving of any doubt which might exist should not depend on the defendant having to put together a jigsaw out of material provided for a different purpose. It is true that the defendant does not say it was confused as to the identity of any particular claimant, but that is not the point. The test is an objective one (though of course the objective assessment might be affected

I therefore consider that it has not been demonstrated that there is no reasonable doubt as to the proper name of this claimant (the double negative is needed to reflect the correct test). I consider that there would have been reasonable doubt whenever the matter is tested.

The other entity in relation to which the point arises is Keystone Investment Trust plc. There were no other Keystone entities identified in the list of claimants to give an equivalent of that part of the context for an Invesco entity, but looking at the name it is not apparent what sort of entity, or who, was being referred to. There is no evidence as to what the state of knowledge in the industry might be. This might have been some sort of misdescribed trust (there were some of those in the list), and its identity as a plc would not have been apparent to a point beyond reasonable doubt. The same result applies to this entity as applies to the Invesco company, and for effectively the same reasons.

The exemplar for this is the entity numbered 23. It is described as Invesco Corporate Bond Fund (UK). The amendment seeks to add words to that description as follows (shown underlined): Invesco Fixed Interest Investment Series for and on behalf of Invesco Corporate Bund Fund (UK). Mr Onslow seeks to amend under all three heads - 17.4, 19.5(3)(a) and 19.5(3)(b). There are another 12 claimants in a similar position, seeking similar amendments based on the same mistake.

The mistake is said to be as to the corporate nature of the funds in question. The originally named claimant is a sub-fund within an OEIC of the nature described in the previous section of this judgment dealing with such entities. These were not subfunds with separate legal personality. They were ring-fenced funds under an umbrella company, with the ring-fencing consequences referred to in that section. The evidence is that “the Claimants” (no particular individual identified) thought that it was unnecessary and inappropriate for the umbrella companies to be parties to the claim because each of the Investment Manager and the Investment Advisor to each of the sub-funds (unidentified to me) were party to the claim, the former body “having

It appears that this mistake as to legal personality persisted into this application because in his first witness statement Mr Warren-Smith for the claimants averred that this fund was an ICVC (Investment Company with Variable Capital) which had separate legal personality. However, by the time of his second witness statement he seems to have appreciated the error and proposed the amendment which is currently under consideration. It thus seems that the error extended from the claimant group to the solicitors. No individuals within the group are named. Mr Warren-Smith contents himself with saying that his information came from individuals in the legal team at Invesco Asset Management Ltd, to whom all management of sub-funds is delegated.

The mistake in this case was genuine within the meaning of CPR 17.4(3). Although it is not attributed to any given individual, it is clear enough from the evidence that it was a failure of the internal organs of Invesco to understand its own structure. The internal legal team at the Invesco group apparently misunderstood the status of these sub-funds. One might say that that ought not to have happened, but it did. That mistake carried into MLB when the names of claimants was provided. Mr Rabinowitz sought to say that the decision was a deliberate one, of the nature which disqualified the claimants in Best Friends and Adelson. I disagree. The decision was deliberate in the sense that it was decided to use the name that was used, but it arose as a result of a mistaken belief as to the status of the funds in question. There was no decision to use the particular name in order to keep options open, or anything like that.

The next question is therefore whether it was a mistake as to name within the meaning of 17.4(3). I do not consider that there was. This rule applies where a party wishes “to correct a mistake as to the name of a party”. It is not apparent from the authorities that the same generosity should apply to this concept as applies to 19.5(3)(b). This was not a case where the correct entity was identified and the wrong name applied. It was a case where the correct entity was not identified at all. A nonexistent entity was identified and named. The correction sought is not to correct the name of the intended party. It is to substitute a completely different name, the name of something that actually exists. It may be that the concept of correcting the name can become a matter of fact and degree, but if that is so then the present falls well on the wrong side of the line. This instance involves no more of a correction of a name than was the case in TRW, in which Fraser J held his case was not a 17.4 case either. In that case there was a mistake as to which entity owned the lease in question. One existing entity was named, when it should have been the other. Fraser J held:

Although the present instance does not involve the joinder of one actual entity in place of another, the position is analogous. There was a conscious attempt to join something which did not exist when there should have been a joinder of a real entity which did exist. Like Fraser J, I do not consider this falls within 17.4(3).

If I am wrong about that then I consider that the 17.4 case falls at the hurdle of reasonable doubt. The best evidence of total uncertainty (not just reasonable doubt) is that the claimants themselves continued to labour under misapprehensions as to which entity owned the shares in question even after doubts were raised as to other entities - see Mr Warren-Smith’s first witness statement. It is not apparent why an objective observer should have no reasonable doubt as to the correct identity of the party in question when the Invesco organisation, and its solicitors, themselves remained confused.

The position under 19.5(3)(a) seems to me to be different. I think it correct to regard this as a substitution even though the existing claimant does not exist. The nonexistence of the originally pleaded claimant is not a bar to the notion of substitution (see the Sardinia Sulcis and Rosgosstrakh Ltd v Yapi Kredi [2017] EWHC 3377 (Comm)), a conclusion to which I would have come even in the absence of authority. The more difficult question is whether the error made was one as to name (appropriately expanded, as the authorities require) or a mistake as to identity. Mr Rabinowitz submitted that where a claimant thinks about which claimant to join, and makes a wrong decision, that is not an error of nomenclature but is an error of identity. Again, I would agree with him on a literal meaning of the words, and I suspect a philosopher would too, but the law has adopted an expanded meaning of the concept of an error as to naming - see the discussion of the authorities above. According to the claim form, the claim was being brought by a fund which claimed a shareholding in G4S. In my view that is a description which is sufficiently specific to this case (per Stocker LJ in Sardinia Sulcis).

That sort of consideration presents itself in a situation which differs as between this case on the one hand and all the other authorities on the point on the other. In those cases the specificity could be established (where it was established) by the singular description of which there was likely to be only one relevant to the particular case. Stocker LJ cited various examples - landlord, employer, owners or shipowners.

The essential description “the owner of shares held by the [Invesco fund]”, which is an acceptable paraphrase which does not beg the questions which arise under this head, is a description which is sufficient to fall within Sardinia Sulcis as elaborated by Adelson.

I have not overlooked the fact that one of the apparent requirements of the old Ord 20 r5, set out in paragraph 43 of Adelson, is:

That would probably not be fulfilled in the present case. For the reasons appearing above in relation to the reasonable doubt element of the 17.4(3) test, the true identity of the legal owner of the Fund’s shares would not have been apparent to G4S. However, I do not regard that requirement as necessarily carried over into CPR 19.5(3)(a). The Court of Appeal did not say that it was, and I do not detect it in the later authorities. If it had been carried over it would have involved the introduction into 19.5(3)(a) of a requirement which would be stricter than the reasonable doubt test, which has been held not to be applicable to this provision as a matter of rule. Accordingly, I do not consider that absence of this factor stands in the way of the conclusion that I have reached.

I therefore consider that the mistake made in this instance was a mistake as to name and not as to identity, being appropriately generous to the claimants for these purposes. For what it is worth, I consider that the position in relation to this particular exemplar falls naturally within the wording of CPR 19.5(3)(a), though I accept that there is little that is natural in the approach which the authorities require to be taken to this provision.

That conclusion makes it unnecessary for me to consider whether 19.5(3)(b) is also available to Mr Onslow, and I shall not lengthen this already lengthy judgment with a consideration of that point in detail. I shall content myself by saying that, in the light of my previous section on this point, I do not consider that it applies, and if it might otherwise do so I would adopt Lloyd LJ’s suggestion that it is appropriate to disallow it on discretion grounds if (contrary to my views) the claimant has already failed on both the mistake grounds.

I add one further point about the form of the amendment, though it is not one taken by Mr Rabinowitz (though he does make a similar point in relation to the next category). I find the new formulation curious. The expression “for and on behalf of” suggests one entity doing something for another entity. One cannot do an act “on behalf of” a non-entity. I have wondered whether that formulation was adopted as a presentational one so as to suggest that the proper claimant and the wrongly named claimant were more or less the same thing, in order to improve the argument. Whether or not that is right, it does not seem to me appropriate. If it is desired to indicate that Invesco Fixed Interest Investment Series (the legal owner of the shares) was bringing its claim in respect of the particular shares held in the Fund, then that could be more aptly phrased, though in my view it is strictly unnecessary in the heading to a claim form (though obviously it would have to be properly pleaded). However, this point is not taken as a bar to the amendment so I shall not take any action on it, other than to draw the attention of the parties (and particularly the claimant) to it in this paragraph.

2 differing exemplars are picked out of a total of 7 similar cases:

“Retirement Annuity Plan for Employees of the Army and Air Force Exchange Service” is sought to be amended to “Trustees of the Retirement Annuity Plan for Employees of the Army and Air Force Exchange Service”.

“APIF - Europe Fund” was sought to be amended to “Bank Consortiun Trust Company Ltd for and on behalf of APIF - Europe Fund”, but in his submissions Mr Onslow said he wished the amendment to reflect trusteeship - “Bank Consortium Trust Company Ltd as trustee of APIF-Europe Fund”.

I will have to take these instances separately, taking the Annuity Plan claimant first.

Mr Warren-Smith has not explained how this particular mistake came to be made. He has explained that this is one of a group of pension and insurance plans to whom any losses accrue. His second witness statement acknowledges that:

In this instance little is known about how the mistake came to be made, or who made it behind the scenes. In his third witness statement Mr Warren Smith says:

It would seem that the originally pleaded name was supplied to the solicitors without sufficient thought being given as to what the appropriate entity was. There is no indication that any consideration was given to the point. The attempt to add this claimant was a late, and no doubt hurried, addition to the litigation. The claim form reveals that the address of the would-be claimants is Texas. The only mistake which one can identify is that MLB thought that the name in which the action should be brought is that originally pleaded. That is a bit thin in the present circumstances.

It now appears that there was a second mistake. The amendment proposes that it take the form of “The Trustees [etc]”. That is not a permissible way for a trust to bring an action. The trust itself is not an entity so the individual trustees have to be joined. The proposed amendment does not achieve that. The “Trustees” have sought to justify the use of the title rather than individual names of trustees, saying it is an acceptable form for describing the trustees as parties, and that certain reported authorities indicate that that is acceptable, citing Starbones v Trustees of the Royal

I disagree. A trust (if that is a correct description of this claimant) is not a legal entity. The relevant legal entities are the individual trustees. They are parties to contracts; they hold relevant property. They must be made parties individually to any litigation. There is no provision which allows trustees to be treated as sued individually even if described compendiously - contrast the position with partnerships where partnerships should be sued in the firm name - see now CPR 7APD, paras 5A.1, 5A.3. Mr Onslow did not show me any authority which decides otherwise.

His authorities were all matters of inference from the reports and how the defendant is described in them in each case. They do not work for him. The first example does not work because the Trustees of the Royal Botanical Gardens Kew is made a body corporate by statute (National Heritage Act 1983 Sched 1 Pt IV para 33). One cannot tell from the report of the Ove Arup case whether or not the heading to the report is an abbreviation or not; and the EE case does not work because the heading to the report on Bailii shows that the individual trustees were indeed joined individually and described as trustees of the Trust (I am puzzled as to how this authority came to be presented to me in the first place). So the originally proposed amendment should not be allowed because it does not seek the joinder of any entity.

However, shortly before the hearing the claimants’ solicitors provided the names of the trustees to be substituted in this case. I shall consider the matter on the footing that that is the amendment application of the claimant in question, though I myself have not been provided with the names.

Mr Onslow seeks to amend under all three heads. The application under 17.4(3) fails because in the circumstances described above it cannot be said there was no reasonable doubt as to the identity of the party in question. Until shortly before the hearing even the claimant did not know the proper identity.

I consider that for the purposes of the application under 19.5(3)(a) there is a mistake of name for reasons analogous to the Category 2 cases. The chosen name, which names something that does not exist, is in fact closer in concept to the intended claimants (or those that ought to have been intended) than the two companies in TRW

Once again, I do not consider that element (iii) of the Adelson formulation in relation to RSR Ord 20 r 5 is one that necessarily has to be fulfilled. If it did, the application would miss by a very long way in this case. I consider that there is no way in which the defendants could have appreciated who the true claimants were supposed to be when it only occurred to the claimants themselves close to this hearing and when defects had been pointed out to them.

In relation to 19.5(3)(b), again I shall not decide the point, but if I had had to I would have decided it against the claimants for the same reasons as appear in Category 2.

I turn now to “APIF - Europe Fund”. The only explanation of what happened in relation to this claimant appears in the first witness statement Mr Warren-Smith where he explains:

In his second witness statement he adds no further information about the nature of the mistake or how it came to be made other than saying describing the amendment which was originally sought. No evidence is provided about how any mistake came to be made, and of course the change of tack in submissions is not evidenced at all.

In his fifth witness statement Mr Warren-Smith explains that this “entity” is managed by the Invesco Group. He goes on:

I consider that this application fails on an evidential basis. Some of the evidence about the other mistakes is fairly thin, but at least there is some. The reference to “clarity” is itself a bit of a mystery, because the statement by Mr Warren-Smith does not provide any. It seems to suggest that for reasons of “clarity” the wrong party was deliberately joined. A person seeking amendment of substitution under 17.4(3) or 19.5 must provide some suitable evidential backing in order to bring the case within the rules. That has not happened in this instance. This is not an unduly pedantic approach. This litigation involves a lot of claimants, large numbers of whom are seeking to amend their names. The number of claimants involved does not excuse any of them from explaining their respective cases properly. The developing nature of this amendment obscures, rather than explains, what (if anything) went wrong here. There is inadequate evidence of a mistake, who made it, and actually what it was.

There are 4 instances in this category. The exemplar is currently pleaded as “Invesco Global Select Equity Fund”. The amendment proposed is to “Invesco Funds SICAV for and on behalf of Invesco Global Select Equity Fund”.

All three bases of amendment are deployed in relation to this amendment.

A SICAV is said to be a Luxembourg entity with legal status which can sue and be sued as such. In his second witness statement Mr Warren-Smith explains that he has been informed by the claimants that what he describes as this “entity” is in fact a subfund of a SICAV, so the position is analogous to Category 2 and the OIECs. He asserts, without producing evidence of foreign law (which he expressly acknowledges he has not done) that the SICAV has sufficient personality to be the appropriate claimant. His third witness statement explains that his firm had direct contact with the underlying claimants (whatever that might mean in this particular case) and the list of entities provided by the claimant did not include the associated umbrella company for the sub-fund. His evidence ends, curiously:

This is puzzling, because the whole purpose of the proposed amendment is to include the umbrella company because it does own the shares. I assume (in his favour) that he means that the umbrella company was not recorded properly as the holder of the shares.

Despite all that, I am satisfied that an error was made at the level of the claimant (who probably did not think enough about it) and the solicitors (who accepted what their client said, no doubt wishing to act quickly against the limitation background). When the fund, rather than the SICAV, was identified as a claimant, there was a mistake by a relevant person.

I consider that the same result follows here as for Category 2. I do not consider that this is a mistake in the name, for the same reasons. I also consider that there would have been more than reasonable doubt as to the identity of the party in question. That deals with 17.4.

So far as 19.5(3)(a) is concerned, again the Category 2 reasoning applies with the same result. The same applies to 19.5(3)(b).

Mr Rabinowitz sought to make something from an absence of foreign law evidence as to the status of the SICAV. If its status had been seriously put in issue for the purposes of these proceedings then there might have been something in this point. However, it was not sufficiently put in issue. Mr Warren-Smith has given some hearsay evidence about its status, and I do not consider that on the facts of this case he has to go further. Mr Rabinowitz submitted that without that evidence I cannot be satisfied that the error had been made. I do not agree. If I am satisfied enough on the basis of the current evidence that the shares were technically held by the SICAV, then that gives rise to the error that was made. That is the current state of the evidence. Should it later transpire that the SICAV has no legal personality after all, then no doubt the claimant might be in difficulty, and might well not be given another opportunity to amend, but that would be a matter for the future.

I repeat my remarks above about the “for and on behalf of” part of the amendment. That seems to me to be inappropriate to describe this claimant for the reasons given above, but again it was not a point taken by Mr Rabinowitz and I shall not hold it against the claimant in question.

In this category I adopt the description given in argument without vesting it with any particular significance for the purposes of legal argument. There are 6 of these, of which the two exemplars are:

As to the first (Rothschild) the attempt to amend fails for evidential reasons about the mistake. All that is said in the evidence about the mistake is this:

That is Mr Warren-Smith’s third witness statement - curiously, it is not dealt with at all until then. All that is said in his second witness statement is that they wish to change the name under each of the the three alternative manners considered in this judgment. His third witness statement remark is repeated in his fifth witness statement.

This claim fails for want of evidence. The approach to the evidence is rather casual, and there is no adequate evidence of the sort of mistake necessary to come within any of the provisions in question. The claimant’s own evidence suggests that the position might still not be clear - “the intermediary is confirming the position”. No further confirmation or elaboration was put in evidence before me. That is not good enough. Mr Rabinowitz said that a Google search throws up a lot of Rothschild entities with similar names, which gives rise to reasonable doubt as to what was intended. That was not in evidence either, but Mr Rabinowitz does not need it because the claimant’s evidence does not get as far as establishing a relevant (or any) mistake.

The position regarding Ossiam is slightly different. Again the matter is not dealt with until late in the evidential chain, when Mr Warren-Smith says (in his third witness statement):

The position is different because there is vestigial evidence of a mistake in that explanation. However, it is not enough in the present situation. There is a burden on the amending party to explain more. To have bodies corporate without some national designation indicating a corporate status is unusual, and calls for more of an explanation than has been given because it looks odd, and the reference to confirmation by the intermediary being required (which, as far as I know, has not been given, and certainly was not evidenced) does not help in resolving that oddity. So there is a qualified averment of a mistake, without any identification of the person making it or any substantiation of it. I do not regard that as sufficient. One can imagine how this exercise might be fairly straightforward, and I confess that if it had been adequately explained I can well imagine how this would be a fairly straightforward mistake within 17.4(3), with not much doubt about the true identity, but the evidence does not allow that conclusion.

The two mistake bases for amending therefore fail, and the claim under 19.5(3)(b) fails for the same reasons as given above.

There are said to be 9 of these, of which the tested exemplars are:

The Allianz case is presented in the evidence as falling under the category of claimants who have changed their names. The only evidence presented by Mr Warren-Smith is the following (in his third witness statement - his second witness statement merely refers to the desire to change the name):

That statement is thin but it effectively achieves various things. First, it explains the nature of the mistake. Second, it explains why the amendment is sought. Third, it offers some sort of explanation as to why the mistake was made and who made it - the person giving instructions had not realised that the entity had changed its name. Mr Rabinowitz seeks to challenge this last point on the basis of evidence given by Mr Warren-Smith about getting properly signed letters of instruction from the client, but his point does not raise a sufficient challenge.

That is evidence capable of supporting an amendment under 17.4(3). It is a genuine change of name, and no more (a rarity in this litigation, it seems). The change is sought to correct a mistake in the name. There is no evidence that the mistake would cause reasonable doubt as to the identity of the party in question. The claim is ostensibly brought by an entity with the first name in respect of shares held in that first name. There has been a change of name. Unless another entity has stepped in and taken on the first name (as to which there is no evidence) then provided the change of name of the entity is genuine then the “no reasonable doubt” provision if fulfilled.

This amendment is therefore allowable under this head. I shall not consider the other heads.

In relation to the above Invesco fund, it is said that a corporate entity needs to be added. It also appears that the intended fund is itself misdescribed in the preamendment claim form - the words “Perpetual GBL” ought not to have been there, and the word “Global” ought to have been there. Those errors are corrected and the trustee’s name is then added.

The evidence about this is again limited. In his first instance Mr Warren-Smith says: “The legal entity owning the assets of this fund is Invesco Pensions Limited, which is a named party to this action. For clarity, this fund has been listed as party to this action, even though it does not itself have separate legal personality.”

In his third witness statement Mr Warren-Smith says:

There is a specific reference to this exemplar:

There is therefore a triple mistake here. There is a wrongly pleaded former description of the fund; that description no longer applies because of a name change of the fund; and the “entity” pleaded does not hold the shares anyway because they are held by a trustee.

I accept that it might just be possible to describe all this as a mistake of name for the purposes of 17.4(3). However, even if that is done then this exemplar fails the reasonable doubt case. Even if the claimant itself is satisfied as to the correct entity now, looking at the matter as at the date of issue of the proceedings, or even as at the date of service, the chain of mistakes would give rise to reasonable doubt as to the identity of the person in question. It is all too puzzling. Mr Onslow pleaded for some understanding since this was a “vast and complex” organisation which does not resemble a human and in which mistakes would be made. I do not doubt that the Invesco group is a complex organisation, in which some mistakes will be made, but these mistakes, taken with the others that have been made, are serious and extensive and, frankly, they ought not to have been made. One would expect the organisation to be able to keep records of changes of names of funds, and transmit names to solicitors accurately - indeed, the names of its funds must be important for all sorts of reasons beyond these proceedings. I am afraid I am unable to extend the sympathy sought.

Turning to 19.5(3)(a), the position might be thought to be analogous to the Category 2 cases where it is in fact an umbrella company holding the assets (though Invesco Pensions Ltd is not said to be an umbrella company). However, I do not think that that works for the claimant in this particular case. In Category 2 one can take the description of the fund, and the pleading of a shareholding, and treat the failure to plead the legal owner as a mistake as to name within the generosity of the test. The present case is different because of the first pleading, which was of a fund which did not own the shares in question. One therefore starts from a different point. I do not consider that the generosity of the test extends as far as this case.

19.5(3)(b) fails for reasons which will by now be apparent. That provision cannot be taken to be intended to extend to the chain of mistakes in this case. If it did it would really amount to a provision which always allows a substitution no matter what has gone before if that is what is necessary to save the action, and that cannot be right.

This exemplar therefore fails.

It is not clear to me how many are in this case, but there are two exemplars:

Again, all three bases are relied on in the alternative in relation to both of these instances.

So far as the first is concerned, the evidence is very limited. In his second witness statement Mr Warren-Smith merely asserts that this is an instance of an amendment or substitution sought, without details. In his third witness statement, it is to be inferred that he refers to this as an error:

The only particulars appear in a table where it is said:

His fifth witness statement repeats that formulation and provides the name of the intermediary and the individuals involved (or at least that is who I assume they are).

CPR 17.4(3) does not avail Mr Onslow. The thin evidence does not describe the process as one in which a name was provided but needs correcting. It describes a process in which one entity was named in place of another. The correction of that does not involve correcting the name of a party. It involves the complete substitution of one party for another. That cannot be done under 17.4(3). It also fails the reasonable doubt test. There is no evidence which allows one to conclude that, objectively speaking, there was no reasonable doubt that in referring to the one entity the claimant intended a reference to the other. If it is suggested that the defendant ought to have rootled around in its register and worked out what the claimants were referring to, I reject that suggestion. It is for the claimant to put in evidence dealing with this point, and there is none.

In this case 19.5(3)(a) does not work for Mr Onslow either. This is a complete mistake as to identity, not name. The original entity (which I assume to be one with legal personality entitling it to sue and hold property) has been replaced by another “Fund” (whether or not it has legal personality does not appear) which is expressed to hold the property for another “Fund”. This is not like the case of a single landlord, tenant or shareholder, with an established identity as such, being given the wrong name in the claim form. It is a complete switch of identity of claimant, assuming (which has not been clearly established) that it has corporate personality. As far as one can tell from the thin evidence, there was a deliberate decision to claim in the name of the original LLC; there was a deliberate intention that that entity should make the claim. The desire to switch is not a change of name even on the generous interpretation given to that concept. It is a complete switch of identity.

The mistake is not particularly satisfactorily described either. It is not clear whether the shareholding was recorded under the wrong name, or whether there was an error in ascertaining which entity had the claim which overlooked the correct name in the records, or whether there was some other error. At the end of the day a mistake is a mistake, and it may not matter much precisely what it was, but a cursory description such as that given is not likely to fulfil the amending party’s obligation to establish a relevant mistake to the satisfaction of the court.

For the reasons appearing above, 19.5(3)(b) does not get Mr Onslow home either. Wherever the boundaries of this provision may lie, it cannot operate in favour of someone who says: “We have sued in the name of A. We can now see that A has no claim, and that B has a claim. Unless B is substituted for A, we will lose. Therefore we seek the amendment.” Furthermore, were it necessary to do so I would find that Lloyd LJ’s approach to discretion would apply to rule out the invocation of this provision.

Turning to the MML entity, the first reference to a change of name is in Mr WarrenSmith’s second witness statement. In a table he describes in general terms names that wishes to substitute for existing names. MML Investment Advisers LLC (the name in the claim form) appears twice in the table. Against one instance the proposed new name is “Mass Mutual Select Funds”, and against the other is “MML Series Investment Fund”. No further explanation is given. The explanation comes in the third witness statement:

That explanation is repeated in his fifth witness statement.

This application fails for similar reasons to the other application in this category, but even more clearly because of the attempt to substitute two for one. The 17.4(3) application fails because this is not a correction of a name. It is a case of pleading the wrong claimant as a deliberate act, and then seeking to substitute two names who (it is said) have turned out to be people whom the intermediaries now identify as having shares in respect of which they need to sue. The error, as described, is not just a misstatement of a name, or anything like that. Furthermore, there is plenty of doubt beyond the reasonable.

The application under 19.5(3)(a) fails because the mistake was not just a mistake of name, even viewed broadly. It was plainly a mistake of identity. The intermediary proposed one entity as a shareholder of shares in G4S. The solicitors are likely to have shared the mistake that that was the case. Then the intermediaries discovered it was not a shareholder, but two other entities were. That is what the evidence seems to say. It is impossible to say that the first entity was “named in mistake for the [two] new [parties]”. It was named because there was a mistake as to whether it was a shareholder at all, when apparently the other two entities were in no way in mind. I do not see how that falls within 19.5(3)(a).

The application under 19.5(3)(b) fails for reasons which are the same as apply in the case of the other case in this category.

So far, with limited exceptions in relation to CPR 19.5(3)(b), I have not dealt with the issue of discretion. If the claimants get over the hurdle of satisfying one of the mistake tests in 17.4(3) or 19.5(3)(a) they still have to satisfy the court that it should exercise its discretion in their favour. The arguments on discretion were mounted on global basis, so I consider them here on that basis even though, as has appeared, it does not arise in relation to cases where I have decided that the claimants have not got over the first hurdle of demonstrating a relevant mistake.

The principal factor relied on by Mr Onslow in support of his application is the balance of prejudice. He submits that rejecting the attempt of the claimants to amend themselves would be to rule out £90m of claims (the bulk of the value of the claims), whereas allowing the amendments will not cause the defendant any real prejudice. Mistakes happen and the jurisdiction is there to allow them to be corrected. He accepted an accusation levelled at his client, that the defendant had no prior warning of the issue of the claim form, but there was a 9 month long attempt to negotiate. While it took some time to mount an actual application to amend, the proposals were foreshadowed some time before the application.

Mr Rabinowitz pointed up three factors which he said pointed against the exercise of discretion in relation to any amendment:

The action was started at the last minute (and beyond the last minute in relation to many claimants) without any prior intimation of a claim. It was incumbent on the claimants to get their identities right and they should not have left their claims until the last minute, and they should themselves have sought permission for joinder and did not do so.

The response in the correspondence to being told that there were issues with the names of claimants was not to accept there was an issue and to set about resolving it at that stage, but was to attack Herbert Smith in correspondence, dispute that there was any issue and criticise G4S for failing to do the necessary research. It took a month to get an application launched, and evidence came in very shortly before the hearing date.

The claimants, or their solicitors, had not been transparent when the identity of their clients, the nature of the mistake and the capacity to give instructions, was queried, and insisted that proper instructions had been given even when it then turned out that half a dozen had not existed and therefore could not have given instructions.

A number (17) of the proposed amendments were made without warning and without the defendants making an application to strike out because the claimants did not know, at the time, that they were mis-named. At some point MLB must have discovered that those claimants were mis-named but did not inform the defendant of that (until the application itself).

The claimants have dragged their heels and refused to provide proper particulars of reliance and loss.

Not all these points (made in the skeleton argument) were pursued in oral argument. Mr Rabinowitz confined himself to (a), (b) and (c).

The most comprehensive review of the authorities on discretion appears in the judgment of Stuart-Smith J in Jalla, to which I was referred. He refers to Horne-

He therefore exercised his judicial discretion in favour of the amendment of the name of the party.

In Adelson at para 57 the Court of Appeal observed:

The question of discretion was considered briefly in Best Friends in the Court of Appeal. Simon LJ said:

In Jalla Stuart-Smith J concluded”

On the authorities, therefore, the following material points can be extracted:

The quality of the mistake can be relevant. An accidental slip that is easily made may be more remediable than other more serious forms of mistake.

The speed with which corrective action is taken is relevant. A speedy application will be looked on more favourably than a tardy one.

Prejudice to each party is relevant.

The fact that a claim will be extended to a claimant who would otherwise be time-barred is not, by itself, sufficient prejudice to justify a refusal of the exercise of discretion. That is logical - the ability to pursue a claim which could otherwise not be pursued is built into the legislation and the rules.

The state of knowledge of the claim on the part of the defendant is relevant. If the defendant knows of a number of similar claims already, and the amending claimant just adds one, then the prejudice to the defendant is not that great (Horne-Roberts). By contrast, if a whole batch of “new” claimants seek to come in, then that may well be different.

It is of assistance to a claimant that the defendant knows of the claim and of the mistake in advance of the proceedings. By contrast, it is relevant the other way if the defendant does not have that knowledge.

The jurisdiction is not intended to be punitive of the maker of the mistake. The court understands that honest mistakes can be made (see also Insight at para 106 and TRW).

It is said that a defendant who is notified of the claim after the expiry of the limitation period is in a better position than one who knows about it before the limitation period has expired. This is justifiable on the basis that in cases like Horne-Roberts and TRW the amended-against defendant is only being put in the same position as he thought he was in before the limitation period expired. The position is otherwise if the defendant knew nothing of the claimants or the claims made until after the period had expired.

I therefore turn to apply those factors, and other factors arising out of the facts of this matter, to the applications before me. Doing so, insofar as it falls to me to exercise my discretion in relation to any matter, I would exercise it against the claimants, apart from one limited exception, for the following reasons.

This is nothing like the case of an understandable error in relation to known prelimitation period litigation being corrected, on a one-off basis, after the limitation period. The defendant was not notified of the claim until after the expiry of the limitation period. That presents immediate difficulties to the claimants on the basis of the above factors, and is one of the indicators of the great (and unnecessary) rush in which the claimants apparently were.

Assuming the sample instances placed before me were typical (which was the purpose of the sampling) it is clear enough what is likely to have happened. At an extraordinarily late stage someone in the Invesco group, or perhaps an intermediary, realised there was or might be a claim. They did not realise early enough, or action it early enough, to enable an orderly marshalling of the various claims and the orderly identification of claimants, followed by a letter before action and a pre-action protocol. There was therefore a rush to get some claims issued, and then a continuing exercise in identifying claimants. Those responsible for the exercise did not carry out enough researches to understand who the claimants should be. That would, I accept, be a tiresome exercise, but it needed to be done. Instead, the identity of claimants was passed to the solicitors who themselves probably rushed through the exercise of adding batches of claimants. I say nothing about the culpability of the solicitors in failing to check (if they did - Mr Warren-Smith said they raised queries when they had any), but the fact that batches of amendments were carried out on one particular day speaks to the haste of the exercise. That haste should have been unnecessary. The source of the error may have been the lack of understanding of intermediaries, or the lack of understanding of Invesco of its own corporate structures, but either way it is not a meritorious position when it occurred on such a grand scale.

A large part of this exercise went on after the limitation period had (at least arguably) expired but before the claim was intimated to G4S. It was in this period that most of the mistaken identities are pleaded. It is said that the exercise was difficult because of the need to go back into a lot of records going back a number of years, and that contributed to the mistakes. That may be the case, but it ought not to have been happening when it was. If it was complex it ought to have started earlier. If it had been, and mistakes had been made, there would not have been a limitation problem.

In practically all mistake cases within this amending jurisdiction the amending party is, of course, the author of its own misfortune, so that is not a reason for disallowing the amendments. However, in this particular instance the authorship went not merely to who made the mistakes, but also to when and in what circumstances they were made. They were made close to or (in most cases) after the end of a limitation period of which the claimants were aware - Mr Warren-Smith in substance refers to the fact that the activities were taking place close to a perceived limitation end date. That is apparently because the decision to sue was taken very late. Where that is the case, and the source of the error is a failure to carry out sufficient investigation to understand who should be a claimant, that is a strong factor against the exercise of the discretion.

A further factor pointing the same way is the late stage at which G4S was told of the claim - 3 months after the issue of proceedings and a similar period after the end of the limitation period. When it was told of the claim there were a significant number of misidentified claimants. At that point the claim can be described as being in something of a mess, with a lot of misdescribed claimants and a large number of claimants who were subsequently removed (about 90). Group claimants have a certain obligation to make their claim clear, not confused, and to do so before the claim is issued.

So far as the speed of correction is concerned, some mistaken references were corrected by purported amendments; others were not done until after the defendant pointed out the problem, and even then only with a degree of reluctance. In its Defence, served on 13^th July 2020, G4S took the unidentified claimants point, and it issued an application seeking to strike out the claims of 64 defendants as being improperly identified. It was not until 14^th September that the claimants said they would, absent consent to amendment, apply to amend to correct names, and then did not make an application until 23^rd October. This is not a particularly prompt response to the point being taken, and if it be said that the scale of the exercise is the cause of any delay, then the riposte is that that scale was caused by the scale of the errors in the first place, and is not a particularly good justification.

Mr Onslow’s main point, as I have said, is lack of prejudice to the defendant. Mr Rabinowitz did not particularly rely on prejudice to his client and it is true that if the amendments are not allowed then a large number of claimants owning a large part of the claim will fall away. That is prejudice. However, on the facts of this case it is less compelling than might be the case in other litigation. All applications of a limitation period cause prejudice. It is so significant in this case because of the scale of the errors that occurred. It arises not because what happened is of the kind of accident that will happen (to use a cliché), but because of a failure to address important points, and doing everything very late. I do not ignore the prejudice but it has to be weighed with the other relevant matters and is nothing like determinative.

So far as 19.5(3)(a) is concerned, there is also the factor of reasonable doubt, of lack of knowledge of the intended identity on the part of the defendant, which is capable of coming back in at this stage. There is no way the defendant can realistically have understood what was intended in relation to all these claimants. This factor weighs against the claimants when it comes to discretion.

Taking all the above factors into account in this case, and taking the discretion point globally, as did the parties, I consider that the correct course is to exercise my discretion against the claimants. These were multiple mistakes borne of haste, casualness and a failure to understand a group’s own structures and/or entities. They were done in the course of activities conducted in full knowledge of an approaching limitation period. If the litigation (which is complex, and required proper attention) had been put in train earlier, either there would not have been so many mistakes, or there would have been time to correct them. As it is the lateness has led to the mistakes and the expiry of a limitation period. Such disorderly litigation is not to be encouraged, and is certainly not to be assisted by the exercise of the court’s discretion as to the amendments sought. There is prejudice to the claimants in this conclusion, but that is what happens when the Limitation Act applies.

I would allow one limited exception to this. In the Allianz exemplar in category 6 I would exercise my discretion in favour of the amendment. A straight historic change of name of a corporate entity which has not been properly recorded is a straightforward matter and where there is no scope for reasonable doubt I would allow

This last point arises out of the fact that the claim form limits claims that are made to claims arising out of statements made by G4S “from 2011”. The Particulars of Claim, however, make claims in relation to publications by G4S from 2006 onwards. The claimants accept that without an amendment to the claim form to take the claim back to 2006 the pre-2011 claims cannot be maintained. They further accept that since a limitation defence arguably arises then any amendment application hase to satisfy the provisions of CPR 17.4(2):

The assessment of whether a claim involves “substantially the same facts” involves a value judgment and:

I find that the claimants cannot satisfy the requirements of this provision. The starting point is the claim form. That pleads the following elements:

From September 2011 the defendant made statements about its compliance with certain standards.

The statements were untrue. G4S had in fact engaged in wrongful billing practices (period unspecified).

As the truth emerged G4S’s share price suffered a decline, reflecting the fact that by reason of the wrongful statements the shares had been trading during the “Relevant Period” (undefined) at an artificially elevated price which was corrected as the truth came out.

That the claimants have suffered loss as a result of their purchasing or continuing to hold shares in the “Relevant Period”.

So the gist of the claim involves (a) false statements made after 2011, which (b) inflated the share price which (c) caused the claimants to pay too much for shares or to hold on to them for too long. Everything relevant therefore took place from September 2011 onwards, with the exception of matters underlying the falsity which might have taken place before September 2011. Crucially, it is those post-2011 matters which give rise to the cause of action pleaded in the claim form.

In the light of that analysis it is impossible for the claimants to maintain that the new proposed claim (which has not actually been provided in draft in claim form style) arises out of the same or substantially the same facts. True it is that the facts giving rise to the falsity may have to be investigated in the prior period, but the rest of the claim involves identifying the statements, identifying the falsity, identifying purchases made by claimants as a result of share price exaggeration since September 2011, and identifying the extent to which any claimant held on to shares which it might otherwise have parted with since September 2011. Any investigation prior to 2011 involves identifying pre-2011 statements, identifying their falsity, identifying the effect on the market in that period, and identifying the purchases or holding-on of claimants in that period. Apart from the falsity point those are entirely different facts which would have to be investigated from those requiring investigation in the 2011onwards period. They are not substantially the same facts as requred by CPR 17.4(2). They might be the same type of facts as some of those currently pleaded, but they are very different facts. The value judgment on this point is, in my view, plain.

It follows that claims based on pre-2011 statements and purchases fall to be struck out of the Particulars of Claim, and I would refuse permission to amend to include them if such an application were formulated.