Aktas v Adepta (A Registered Charity)

These conjoined appeals raise the issue whether a claim whose claim form has been issued towards the very end of a limitation period and has then been struck out owing to a failure to serve it in time (CPR Part 7.5(1) allows four months from issue of the claim form for its service) can be resurrected in a second action which invokes the discretionary provisions of section 33 of the Limitation Act 1980 (the “1980 Act”) relating to claims for personal injury. The judge in each case has said No, and, without reaching the Limitation Act issue, has struck out the second action as an abuse of process. In one case he has also gone on in the alternative to exercise his discretion against disapplying the basic limitation period, but it is common ground that if the claim survives into a second action the exercise of that discretion was flawed and that it is for this court to exercise it afresh.

To understand the arguments it is necessary to visit a number of separate lines of authority dealing variously with failure to serve in time, with abuse of process, and with section 33. At the heart of the issue, however, is a relatively recent decision of the House of Lords, Horton v. Sadler [2006] UKHL 23, [2007] 1 AC 307 (Horton). The importance of Horton is that it departed from an earlier decision of the House in Walkley v. Precision Forgings Ltd [1979] 1 WLR 606 which had cast its influence on previous jurisprudence. Walkley had decided that the discretion in section 2D of the Limitation Act 1939 (the “1939 Act”), the predecessor to section 33, could not be exercised in relation to a second action which had been commenced out of time where the first action had been commenced in time but had been lost due to a failure to serve in time. Horton departed from that decision, both as a matter of the true construction of section 33 and as a matter of the interests of justice – which it was necessary to take into account since the House was reviewing its own prior decision. Horton held that section 33 gave a wide and unfettered discretion in relation to the second action to disapply the time limit of three years contained in section 11 of the 1980 Act having regard to the degree to which the parties would be prejudiced and taking into account all the circumstances of the case.

Although the submissions of the parties ranged widely over the intricate jurisprudence cited to us, the central submission of the respondent defendants is that a failure to serve in time in the first action is so serious a misuse of procedure and so strictly regarded as an act of disrespect to the court as to constitute or be tantamount to an abuse of process, such as requires the striking out of the second action as a further abuse of process, irrespective of any question of the section 33 discretion, which thus is never reached and becomes irrelevant. The central submission of the appellant claimants is that this has never been held to be the case in any of the multitudinous cases which have traversed this problem, and that the so far successful attack on their second actions is an attempt to put the clock back to before Horton.

It is important to emphasise that, although the claimants, through their negligent solicitors, are undoubtedly at fault in failing to serve their respective claim forms in time, and have therefore fallen outside the basic limitation period of three years under section 11 because their claim forms were themselves issued only shortly before the end of that period, nevertheless they are not guilty of any other breach of rule or order of the court. Moreover, the defendants were well aware of the respective claims and had admitted liability before the debacle.

What therefore is primarily at issue in this appeal is the tension between on the one hand the strictness with which it is clear that mere failure to serve in time is regarded by rules of procedure and the courts, and on the other hand the statutory concession which Parliament has seen fit to allow to personal injury claimants under the provisions of section 33. As will appear, that concession is largely determined by the question of prejudice to the parties, although the discretion is wide and traverses all the circumstances of the case. The defendants submit that, in the face of the claimants’ abuse of process, the courts are entitled to protect their own procedure, irrespective of such questions. The claimants submit that, at any rate in circumstances such as appear in these cases, this is to ignore the will of Parliament. The defendants submit that if the judges below were wrong to have struck out the second actions, then there is no sanction for the serious failure to serve in time. The claimants submit that there is sanction enough in the failure of the first action (with attendant cost consequences and the need for a claimant to pass successfully over the hurdle of section 33). The defendants submit that, irrespective of any lack of relevant prejudice to them in allowing the second action to proceed, they should be freed of responsibility so that the court may discipline the claimants and their solicitors. The claimants’ loss may be mitigated by the new cause of action which they obtain against their solicitors.

The two cases with which we are here concerned arose out of personal injury claims. In one, Aktas v. Adepta, the claimant, Mrs Aktas, suffered her injury on 4 September 2004. Her letter of claim was sent almost two years later on 11 July 2006. Her first claim form was issued on 3 September 2007, the last day of the three year limitation period. However, on the very next day the defendant admitted liability. Time for service was extended, but even so service was only effected on 3 April 2008, the day after time expired. The claim form was set aside. On 22 August 2008 Mrs Aktas issued her second claim form. On 21 April 2009 District Judge Dignan struck out the second claim as an abuse of process. On 24 June 2009 HHJ Tetlow referred Mrs Aktas’ appeal to this court. On this appeal Mr Augustus Ullstein QC has appeared on behalf of Mrs Aktas and Mr Christopher Purchas QC has appeared on behalf the respondent Adepta.

In the other appeal, Dixie v. British Polythene Industries plc, the claimant, Mr Dixie suffered his injury on 27 February 2005. His letter of claim was sent promptly on 19 August 2005. On 5 October 2005, the defendant admitted liability. There were interim payments and an offer to settle. The first claim form was issued on 22 February 2008, shortly before the expiry of time. Time for service expired on 22 June 2008, but by oversight service was not effected until 7 July 2008. The claim was struck out by District Judge Manley on 13 August 2008. On 29 August 2008 Mr Dixie issued his second claim form and it was served on 5 September 2008. On 8 July 2009 HHJ Mitchell struck out the second claim (but granted permission to appeal). He did so on the basis that the second claim form was an abuse of process, alternatively he would have declined to exercise his section 33 discretion in Mr Dixie’s favour. On this appeal Mr Richard Methuen QC has appeared on behalf of Mr Dixie and Mr James Rowley QC has appeared on behalf of the respondent BPI.

Thus in both cases liability was admitted either before or soon after the issue of proceedings, but the first claim was lost because of failure by the solicitors concerned to observe the last date for service. In both cases the parties were in touch with one another both before issue and in between issue and service of proceedings, in mutual awareness of the issue of the claim form, dealing with medical evidence relating to quantum. In Mrs Aktas’ case, her first claim form was set aside. In Mr Dixie’s case, his first claim form was struck out. It is agreed that nothing turns on this verbal distinction. A fuller chronology of each case appears below.

Aktas v. Adepta

Dixie v. BPI

The 1980 Act provides a special time limit for actions in respect of personal injuries of three years from the date on which the cause of action accrued, or the date of the claimant’s knowledge, if later. Section 33 provides a discretion to disapply the section 11 time limit, ie to extend time for commencing a personal injury action, in the following circumstances:

Part 6 contains the rules for service, and Part 7 contains the rules on how to start proceedings. The critical rules are in CPR 7.5 and 7.6.

As at the dates of issue of the claims under appeal and until amendment by the Civil Procedure (Amendment) Rules 2008, CPR 7.5 (“Service of a claim form”) provided:

“(1)

After a claim form has been issued, it must be served on the defendant.

(2)

The general rule is that a claim form must be served within 4 months of the date of issue…”

CPR 7.6 (“Extension of time for serving a claim form”) provided:

“(1)

The claimant may apply for an order extending the period within which the claim form may be served.

(2)

The general rule is that an application to extend the time for service must be made –

(a)

within the period for serving the claim form specified by rule 7.5; or

(b)

where an order has been made under this rule, within the period for service specified by that order.

(3)

If the claimant applies for an order to extend time for service of the claim form after the end of the period specified by rule 7.5 or by an order made under this rule, the court may make an order only if –

(a)

the court has been unable to serve the claim form; or

(b)

the claimant has taken all reasonable steps to serve the claim form but has been able to do so; and

(c)

in either case, the claimant has acted promptly in making the application.

(4)

An application for an order extending the time for service –

(a)

must be supported by evidence; and

(b)

may be made without notice.”

Thus where the time for service has not yet passed, the rule provides an open discretion to extend time. Where, however, the time for service has passed, the rule is tightly and strictly drawn: there can be no extension unless either the court or the claimant has been unable to serve the claim form. Therefore, the fact that not only the claim form in question but the cause of action which it reflects may be lost (for reasons of limitation) is no excuse or mitigation. The jurisprudence, which will be discussed below, demonstrates the strictness with which both CPR 7.6(2), as a matter of discretion, and CPR 7.6(3), as a matter of the effect of the rule itself, has been applied. Thus an extension requested within time cannot be granted merely to save against incompetence. An extension requested out of time cannot be granted unless it falls within the exceptional cases stated in the rule itself. See Vinos v. Marks & Spencer plc [2001] 3 All ER 784 (CA) (on rule 7.6(3)) and Hashtroodi v. Hancock [2004] EWCA Civ 652, [2004] 1 WLR 3206 (on rule 7.6(2)).

There is also a rule which enables a claimant to apply to dispense with service: see CPR 6.9 (“The court may dispense with service of a document”). Some claimants who have failed to serve in time have sought a remedy under CPR 6.9 rather than CPR 7.6. However, the jurisprudence shows that CPR 6.9 does not provide a route around the limitations of CPR 7.6: see Godwin v. Swindon Borough Council [2001] EWCA Civ 1478, [2002] 1 WLR 997. In any event, as Civil Procedure 2008 explains (under para 7.6.4), the power to dispense with service is designed for a claimant who seeks to be excused from the need to prove service, rather than for seeking permission to serve out of time.

There is some uncertainty about the correct way to close down an action which cannot go forward because service is out of time or cannot be effected in time. In the Aktas case it was agreed that if the claimant were not permitted to dispense with service, the defendant’s application to set aside the claim form and its service had to succeed. In the event, Deputy District Judge Masheder’s order was to “set aside” the claim form and its service. In the Dixie case it similarly does not appear what application the defendant made with respect to the claimant’s first action. District Judge Manley’s order said that “The claimant case be struck out”. In these appeals, Mr Rowley (on behalf of BPI) submitted that the correct application is under CPR 3.4(2)(c), to “strike out a statement of case if it appears to the court…that there has been a failure to comply with a rule, practice direction or court order”. He accepted that the correct application was not within CPR 3.4(2)(b), to strike out on the basis “that the statement of case is an abuse of the court’s process”. (Nevertheless, he submitted that there was a fundamental failure leading to the same result, namely a strike out). Mr Purchas on the other hand (on behalf of Adepta) submitted that the correct procedure was under CPR 11(1), whereby “A defendant who wishes to…(b) argue that the court should not exercise its jurisdiction, may apply to the court for an order that…it should not exercise any jurisdiction which it may have.” The difficulty with Mr Rowley’s suggestion is that CPR 3.4 applies in terms to a statement of case rather than to a claim form.

It appears from Hoddinott v. Persimmon Homes (Wessex) Ltd [2007] EWCA Civ 1203, [2008] 1 WLR 806 that definitive support is there to be found for the CPR 11(1) route, as Mr Ullstein brought to our attention in his reply. On that basis, a defendant who is served out of time has to file an acknowledgment of service and bring his application within 14 days, otherwise “he is to be treated as having accepted that the court has jurisdiction to try the claim” (CPR 11(5)). What happened in Hoddinott was that the defendant successfully applied to set aside the claimant’s ex parte success in obtaining a CPR 7.6(2) extension of time for serving his claim form, but in the meantime it had acknowledged service and failed to exercise its CPR 11 right to dispute jurisdiction. That was enough to defeat Persimmon, although this court also held that the judge had been wrong to set aside the extension of service. What, however, if the claim form is never served? In such a case the action appears to go into limbo, although it can of course be discontinued.

CPR 3.10 however states:

“Where there has been an error of procedure such as a failure to comply with a rule or practice direction –

(a)

the error does not invalidate any step taken in the proceedings unless the court so orders; and

(b)

the court may make an order to remedy the error.”

That might suggest by inference that the court may make an order invalidating either the issue and/or the service of a claim form where service has not been timeously effected. As Civil Procedure 2010 states (at 3.10.1):

“In these rules, as under the previous rules, non-compliance with a rule, or (under these rules) a practice direction, does not nullify the proceedings or any step taken in the proceedings unless the court so orders.”

On this basis, it might seem that Deputy District Judge Masheder’s order to “set aside” the claim form and its service might not seem to be wide of the mark. The relevance of this is to ask what support there might be within the CPR code itself for the respondents’ submission that a failure to serve in time, in breach of the rule requiring service within 4 months, is an abuse of process.

The defendants submit that the CPR regime is more stringent than the prior regime of the Rules of the Supreme Court (RSC). It is not however apparent that this is so. Thus the note to Civil Procedure 2008 at 7.6.1 began

“Whilst r.7.6 is self-explanatory and effects no change in practice it nevertheless provides welcome clarification to the previous provisions found in RSC 6, rr.8(2) et seq. and CCR O.7, r.20(2) and (3).”

This is not, however, to be found in the current 2010 edition.

It is therefore relevant briefly to consider the RSC regime. Of course, the CPR is a self-contained code and also contains entirely new provisions such as CPR Part 1 and the overriding objective. It is not in general to be interpreted by reference to the RSC and its jurisprudence. There is a new determination under the CPR that time limits should be observed, and that delay will be sanctioned. Nevertheless the question has been raised whether the strictness of CPR 7.6 breaks entirely new ground.

Under the rule in effect before the expiry of the RSC, Ord 6, rule 8, a writ was “valid in the first instance” for a period of 4 months. Prior to 1990 the period had been 12 months. Although the writ was expressed to be “valid” only for its duration, it does not appear simply to have lapsed into invalidity after its expiry, for the irregularity of service after expiry could either be waived or objected to: see The Supreme Court Practice 1997 at 6/8/3. The rule permitting extension (rule 8(2)) said that an extension could be granted “if an application for extension is made to the Court before that day [the day of original expiry] or such later day (if any) as the Court may allow”. The general principles of jurisprudence governing the grant of an extension included the following (ibid at 6/8/4):

“(1)

It is the duty of the plaintiff to serve the writ promptly. He should not dally for the period of its validity; if he does so and gets into difficulties, he will get scant sympathy.

(2)

Accordingly there must always be a good reason for the grant of an extension. This is so even if the application is made during the validity of writ and before the expiry of the limitation period; the later the application is made, the better must be the reason…

(6)

The application for renewal should ordinarily be made before the writ has expired. The court has power to permit a later application but it must be made within the appropriate period of the first expiry…

(7)

A writ will not normally be renewed so as to deprive the defendant of the accrued benefit of a limitation period. The strict view taken in Heaven v. Road and Rail Wagons Ltd [1965] 2 Q.B. 355…was approved by the Court of Appeal in Chappell v. Cooper (above) but must be read in the light of the decision of the House of Lords in Kleinwort Benson Ltd. v. Brabrak Ltd., The Myrto (No.3) [1987] A.C. 597…Possible exceptions are the good reasons in 4(a) or (b), above [where negotiations are proceeding; where legal aid is awaited], or very sharp practice by the defendants which has deceived the plaintiff into inactivity…”

It appears therefore that there is now some tightening of the rules under the CPR in limiting the discretion where the application is made out of time under CPR 7.6(3)).

Nevertheless, so far as the context of the expiry of a limitation period in a personal injury claim is concerned, the then leading cases, first of Firman v. Ellis [1978] QB 886 (CA) and then of Walkley v. Precision Forgings Ltd [1979] 1 WLR 606 (HL) demonstrate that, subject to the differing views there taken of the relevance of section 2D (the precursor of section 33), the mere failure to serve a writ in time was strictly sanctioned by the loss of the action concerned.

Thus in Firman, this court was concerned with three actions which had been commenced within the limitation period but which had failed for non-service of the writ in time (and a fourth action by way of third party proceedings which had been commenced one day out of time). Each of those four actions having failed (in Horton [2007] 1 AC 307 at [10] Lord Bingham of Cornhill described them as having “lapsed”), the plaintiffs started new proceedings and sought to rely on section 2D which had only recently come into effect under the Limitation Act 1975 (amending the 1939 Act). The background facts demonstrate that then, as now, the penalty under the rules of procedure for failure to serve in time was strict. An extract from the facts of one of the cases, given by Lord Denning MR in his judgment, illustrates the point (at 901B/C):

“May 23, 1973: the accident…July 26, 1974: protective writ issued and insurers notified of it. July 26, 1975: writ ceased to be valid unless renewed: see R.S.C., Ord. 6, r. 8…May 23, 1976: three years elapsed since the accident…October 25, 1976: insurers ask for evidence that the writ had been renewed. Plaintiff’s solicitors reply that it had not been renewed. November 2, 1976: insurers say “snap”…December 30, 1976: plaintiff’s solicitors apply to renew the writ and give history of negotiations in detail. March 10, 1977: registrar refuses leave to renew. July 4, 1977: judge affirms decision of registrar.”

In another of the cases, the decision affirming the judge was that of this court (at 902B).

The new statutory discretion under the 1975 Act was, however, another matter. Lord Denning described the background to the 1975 Act and continued (at 905D/906B):

“The granting of this discretion is a revolutionary step…The value of this wide discretion is well shown by the present series of cases. They all arise out of circumstances which the various committees never had in mind at all. In each of the three cases there were negotiations for a settlement, but the plaintiff’s solicitors, by the merest slip, allowed time to run out. They failed to renew the writ in time. This slip did not prejudice the defendant or his insurers in the least. Yet as soon as the insurers discovered it, they cried “snap” and broke off the negotiations. They said to the plaintiff: “You are statute-barred. We are not liable. You sue your own solicitors for negligence. Make their insurers pay. And not us.” All of the judges rejected this submission. Each of the judges exercised his discretion in favour of the plaintiff. I think they were quite right. As a matter of simple justice, it is the defendant’s insurers who should pay the plaintiff’s claim. They have received the premium to cover the risk of these accidents. They should not be allowed to foist their liability on to the plaintiff’s solicitors or their insurers by calling “snap” as if it were a game of cards.”

Lord Denning accepted, however, that when it came to the section 2D discretion, the negligence of the plaintiff’s solicitor and the plaintiff’s remedy against him was a relevant consideration, for it was one of the “circumstances of the case”. But it was not as critical as questions of prejudice (at 909B/D). Ormrod LJ agreed: he said (at 912E):

“I do not think, however, that it carries much weight in these cases. The court is not concerned solely with financial prejudice to the plaintiff. It is prejudicial to be forced to start another set of proceedings and against a party whom one does not particularly wish to sue and to be deprived of a good cause of action against the original tortfeasor.”

In Walkley, however, the House of Lords took a new and different view of section 2D. This was because they interpreted it as being irrelevant where a first action had been started within the limitation period. In such a case the plaintiff had not been prejudiced at all by the Act’s three year limitation period (then found in section 2A), prejudice which section 2D took as its premise. Lord Wilberforce put the matter thus (at 609):

“But if the plaintiff has brought his action within the three years, how has he been prejudiced by section 2A? This I fail to understand. If this argument is sound, the respondent’s case fails in limine. He brought his first action within the normal limitation period, and if he has suffered any prejudice, it is by his own inaction and not by the operation of the Act.”

As it happens, Walkley was not a case where the first action had failed for want of timely service, but it was a bad case of want of prosecution. The plaintiff had consulted his trade union in 1969 about a complaint affecting his fingers. He issued a writ in October 1971, which was served, but nothing further was done. In July 1973 the defendant said that if the action was pursued it would issue a summons to dismiss it for want of prosecution. Still nothing was done and the action just appears to have run into the sand. A second writ was issued in December 1976. The court of appeal held (by a majority) that even if the first action could and would have been struck out for want of prosecution, the plaintiff was entitled to seek an extension under section 2D. The point on the interpretation of section 2D itself was a new point in the Lords. In the light of that point as accepted by their Lordships, Mr Walkley had commenced a second action out of time and could not avail himself of section 2D. It followed that if his first action had never been started, he could have invoked section 2D. This was an anomaly which has plagued the Walkley doctrine ever since. It may be observed that it was not suggested that the statutory discretion was not reached because the conduct of the first action had amounted to an abuse of the process of the court.

In Thompson v. Brown [1981] 1 WLR 744 (HL) the House of Lords had considered the Walkley doctrine again. That was a case where there was no failed first action: there was merely a negligent failure by the plaintiff’s solicitor to issue a first writ until one month after the three years had run. The argument there was that section 2D could not apply because there was no prejudice where the plaintiff had an unanswerable claim against his solicitor. The House rejected the argument and remitted the case back to the trial court for it to consider the section 2D discretion. In Horton Lord Bingham at [17] said that Thompson was authority for three propositions: (1) Firman v. Ellis was wrongly decided in the light of Walkley, but had been right to say that the statutory discretion was unfettered; (2) a plaintiff’s remedy against his negligent solicitor was always a highly relevant consideration but did not preclude exercise of the section 2D discretion in the plaintiff’s favour; and (3) the anomaly referred to in my previous paragraph was acknowledged. Lord Diplock also acknowledged that where the writ was only briefly out of time, the defendant’s limitation defence might be regarded as a windfall (at 750). It appears that at this stage of the jurisprudence (Thompson v. Brown) it was not yet clear that Walkley excluded the applicability of the section 2D discretion even in a case of mere failure to serve the first writ in time.

The loss of a first action due to mere failure to make timely service of its writ was considered under the Walkley doctrine in Deerness v. John R Keeble & Son (Brantham) Ltd [1983] 2 Lloyd’s Rep 260. Walkley was applied so as to prevent the plaintiff in his second action from invoking the statutory discretion. In Walkley Lord Diplock had raised the possibility of “most exceptional circumstances”, which the trial judge had utilised in extending time, but in Deerness Lord Diplock said that only estoppel could suffice, and that otherwise the doctrine brooked no exceptions. He again acknowledged the anomaly. See Lord Bingham in Horton at [18]. Thus the Walkley doctrine was a doctrine relating to the scope of section 2D, and not relating to a doctrine of disciplinarianism. It applied to serious cases of want of prosecution and to mere failures to serve a valid writ in time alike.

Over the years different situations have tested the Walkley doctrine and its difficulties. It has been disliked because it has put all manner of cases where a first action fails and a second action has been started out of time into a straitjacket. It has therefore been distinguished where the courts have felt they could. Thus where the first action has failed because of the formal lack of a condition precedent, such as notice to the MIB (the situation in Horton itself), the straitjacket could be avoided, because the first action was not valid. In Horton Lord Bingham said this (at [19]):

“The detailed grounds relied on to distinguish Walkley in these cases are not in my opinion important: the significance of the decisions lies in the Court of Appeal’s unwillingness to regard the ratio of Walkley as applicable to any case not on all fours with it and in the disfavour with which the reasoning and decision in Walkley has in some of these cases been regarded.”

Ultimately the House of Lords in Horton decided that the Walkley doctrine was a misreading of the statute and wrong. Lord Bingham put it in this way:

“21 (1) Counsel for the appellant submitted that…the question for the court under section 33 is always whether it is equitable or inequitable as between the parties to override the time bar which, if relied on by the defendant, will, unless disapplied by order of the court, defeat the action which the plaintiff has ex hypothesi brought out of time.

22

This analysis is, as I think, plainly correct. But the appellant contended that it could not readily be reconciled with the reasoning in Walkley. It was held there that the plaintiff, having issued a writ within the three-year period could not be prejudiced by section 11. But the action timeously brought by the plaintiff in Walkley, as in the Firman v Ellis [1978] QB 886 cases, could not be effectively pursued. The plaintiff could succeed only in his second action. To that section 11 provided a bar which prejudiced him by defeating his action. Section 11 did not prejudice the plaintiff in his first action, brought in time, but that was not the relevant action. When in Thompson [1981] 1 WLR 744 it was said, more than once, that section 11 did not affect Mr Walkley at all, this overlooked the fact that section 11 affected him, unless disapplied, by defeating his second action. The negligence of his solicitors gave rise to the need for a second action, but it was the time bar which meant that that action was bound to fail.

23

Despite the great eminence of those who gave judgment in Walkley, explained and distinguished that decision in Thompson and applied it in Deerness [1983] 2 Lloyd’s Rep 260, I do not think there is any answer which can be given to the appellant’s criticisms and counsel for the MIB provided none.

24

(2) The appellant relied on the fine distinctions drawn by the Court of Appeal in some of the cases cited above as evidence of the court’s reluctance to apply Walkley in any case where it was not strictly constrained by that authority to do so, and in particular on the distinction, acknowledged by Lord Diplock to be anomalous, between cases where the plaintiff had issued proceedings within the primary limitation period, which had thereafter foundered for some procedural reason and cases where no proceedings had been issued at all. There was no rational ground for concluding that a defendant should be vulnerable in the latter case, where the plaintiff’s solicitor’s negligence would probably be greater but not in the former case where it would probably be less. This was an anomaly caused not by the rule that proceedings are brought when the writ is issued, and not by the language of the statute, but by the decision in Walkley. In the result, the distinction lacked any principled justification.

25

I can, again, see no answer to this criticism, and I do not think counsel for the MIB was able to advance one.

26

(3) In reliance on Firman v Ellis, strongly affirmed on this point by the House in Thompson, the appellant contended that the legislation conferred a wide and unfettered discretion and that it was accordingly inconsistent with that interpretation to constrain the exercise of the section 33 discretion by technical rules such as that laid down in Walkley. This was to subvert the intention of Parliament to be derived from the terms of the legislation.

27…Lord Denning’s exposition of the statutory language in Firman v Ellis is to my mind persuasive. It was endorsed by the house in Thompson. It has never to my knowledge been questioned.

28 On this point also I would accept the thrust of the appellant’s criticism. Mr Walkley’s case for seeking an exercise of the section 33 discretion in his favour was, as Waller LJ rightly held, of the weakest. Thus the outcome of the case was clearly right. But I feel bound to conclude that the reasoning of the decision was unsound, that it has given rise to distinctions which disfigure the law in this area and that the effect has been to restrict unduly the broad discretion which Parliament conferred.”

Lord Bingham then considered whether it would be legitimate for the House to depart from its previous decisions. He considered that it would be right to do so. He concluded in this respect –

“31.

There is force in these arguments. But this is not a case in which contracts, settlements of property or fiscal arrangements have been entered into on the faith of a settled legal rule. The criminal law, where certainty is particularly important, will be unaffected if the House departs from Walkley. There will be no detriment of public administration. While injustice may bear more hardly on individuals, I do not regard the concept as inapplicable to judgments affecting corporations with competing interests. I would in the result depart from Walkley for three reasons, taken together: that it unfairly deprives claimants of a right Parliament intended them to have; that it has driven the Court of appeal to draw distinctions which are in my opinion correct but which are so fine as to reflect no credit on this area of the law; and that it subverts the clear intention of Parliament.”

At the end of his speech, Lord Bingham addressed the question of “Abuse” as follows:

“37 It was argued below and in the House that it was an abuse for the appellant to bring a second action while his first action was still extant. The judge accepted that in the ordinary way it is an abuse to pursue two actions against the same defendants in respect of the same subject matter and indicated that if he were giving permission for the second action to continue he would require the first to be discontinued. This was, I think, the correct response. As it was, he dismissed both actions.”

I have cited extensively from Lord Bingham’s speech, with which Lord Rodger of Earlsferry, Lord Carswell and Lord Brown of Eaton-under-Heywood agreed, because Horton is the central authority cited to us on these appeals, and because it is submitted on behalf of the respondents that it was exclusively concerned with the case of an MIB defendant to whom proper notice had not been given. On the contrary, Lord Bingham’s reasons show that they are concerned with every aspect of the problem, and that the fine distinctions which had been drawn in some cases in this court were of no credit to the law. They also show that Firman, which was concerned with three cases of lack of timely service, has been fully reinstated. Indeed, on the subject of second actions which are out of time and therefore raise the issue of the section 33 discretion in circumstances where the first action lapsed because of the failure to serve in time, Firman is once again the leading case, subject to any critical distinction which the CPR regime may have introduced. In that connection the respondents submit that recent CPR authority such as that discussed below, and in particular Hashtroodi which Mr Purchas described as the key authority, was not cited to the House. It remains to be seen what the learning of such cases is. The fact remains, however, that in a decision in which the House of Lords was concerned to set a new path, albeit that path also led back to Firman, and in which the question of abuse was considered, even if not over a wide spectrum, there is no support whatsoever for a doctrine which would exclude the claimant in his (or her) second action from arguing the section 33 discretion on the ground that the first action had been struck out. Even Walkley itself, where the first action must effectively be regarded as if it had failed for want of prosecution, is referred to as a case where, albeit the result was rightly in favour of the defendant, the rationale, which sought to exclude the application of the statutory discretion, was wrong (see Lord Bingham at [28]).

The judgments below proceeded on the basis that the loss of the claimants’ first actions constituted an abuse of the process of the court such that the commencement of their second actions was itself a further abuse of process which entitled the court to strike out such actions without any consideration of the section 33 discretion. In effect, the second actions were struck out, not because they were out of time, but because they were abusive.

In Aktas, District Judge Dignan put the matter in this way:

“3.

In summary this is an application to strike out this claim as being an abuse of process on the basis that to allow this claim to continue would bring the Administration of Justice into disrepute amongst right thinking people: the application relies on this argument only…

12.

The argument is that

(1)

The Rules for serving a Claim Form, particularly where limitation has expired, are critical and that a failure to comply should result in the claim being lost.

(2)

If the second claim here were not struck out it would have the effect of undermining all of the CPR 7.5 and Anderton authorities. In effect there would be no sanction, save for costs, and second claims after a strike out could be issued with impunity. That is the abuse. There are no exceptional/special circumstances to rescue this second claim…

13…to counter the above…the main attack centres around the effect of Horton…In a claim such as this where service was one day late and liability has been admitted it is not difficult to envisage an application under s.33 to disapply s.11 and to allow the claim to continue would succeed. Therefore it is the nub of the argument to decide if an Abuse of Process application can be considered and succeed without any reference to the s.33 criteria…

27.

I am satisfied it is open to me to find in an appropriate case that where a claim has been struck out as a result of a failure to comply with CPR 7.5 that it is an abuse of process to issue a second claim outside the Limitation Period notwithstanding there is an extant application under s.33. Indeed, logically there should be no distinction between inside and outside the period…”

In Dixie, HHJ Mitchell put the matter thus:

“6.

The circumstances require me to decide how to balance the court’s powers of management, on the one hand, with the wide discretion given to the courts under s.33…

and then, after reviewing the authorities, concluded:

“21.

The court has management powers and the authority is clear that a failure to serve through oversight means that time for service is not extended. In my judgment, the reality is that s.33 of the Limitation Act 1980 cannot be used to put right the failure of the claimant’s solicitors to get proceedings started, or, indeed, continued…

25.

I appreciate Mr Roy’s point about prejudice. It is difficult to imagine there could be any prejudice to the defendants in this case. They admitted liability, their argument is about the length of time and the future loss, as I understand it, that might or might not have been occasioned in this case. But, in my judgment, s.33 is for that class of cases where the claimant may be unaware of the nature or extent of his injuries or illness or his medical condition. It is a very different matter from the exercise of case management powers by the court. It is likely I would not have exercised my discretion in favour of the claimant.”

The leading modern case on abuse of process is of course Johnson v. Gore Wood & Co [2002] 2 AC 1. It was not concerned with delay in the conduct of litigation, but, more generally, with the problem of successive civil actions arising from the same facts. In an already classic passage, Lord Bingham said this (at 31B/D):

“I would not accept that it is necessary, before abuse may be found, to identify any additional element such as a collateral attack on a previous decision or some dishonesty, but where those elements are present the later proceedings will be much more obviously abusive, and there will rarely be a finding of abuse unless the later proceeding involves what the court regards as the unjust harassment of a party. It is, however, wrong to hold that because a matter could have been raised in earlier proceedings it should have been, so as to render the raising of it in later proceedings necessarily abusive. That is to adopt too dogmatic an approach to what should in my opinion be a broad, merits-based judgment which takes account of the public and private interests involved and also takes account of all the facts of the case, focusing attention on the crucial question whether, in all the circumstances, a party is misusing or abusing the process of the court by seeking to raise before it the issue which could have been raised before.”

In the context of delay in the conduct of litigation, there is a line of cases which goes back at least to Janov v. Morris [1981] 1 WLR 1389. That case concerned a claim for damages for breach of contract. Following the service of a defence, the action went to sleep. An application to strike out for want of prosecution was made on the specific ground that there was an abuse of process (under RSC Ord 18, r 19), and the judge made an unless order to strike out unless a summons for directions was taken out by a certain date. The plaintiff neither complied with the order nor explained his failure to comply. The first action was therefore struck out. A few months later the plaintiff started a new action, still within the limitation period. The defendant applied to strike out the second action. This court acceded to that application. The issue was whether a second action within the limitation period could be struck out for abuse in such circumstances. That point had been left open, with conflicting views expressed, in Tolley v. Morris [1979] 1 WLR 592 (HL), see Lord Diplock at 603 and Lord Edmund-Davies at 604. Dunn LJ pointed out that the power of the court to strike out for want of prosecution –

“should be exercised only where the court is either satisfied that there has been an intentional and contumelious default – for example, disobedience of a peremptory order of the court – or that there has been inordinate and inexcusable delay” (at 1392G).

In Janov, such abuse had been found in the first action, indeed the plaintiff was “still in contempt of court” (at 1395C). Watkins LJ agreed with Dunn LJ and added –

“To behave in such a way is in my judgment to treat the court with intolerable contumely. This is a matter which can properly be taken into account in the exercise of the court’s discretion” (at 1395H).

Watkins LJ also said (at 1395F):

“A prospective litigant must be deemed to know that upon taking out a writ endorsed with a claim for monetary or other relief, his conduct of the action thereby brought into being will be governed thereafter by rules and orders of the court. A failure to conform to any one of these may cause him to be penalised even to the extent of having his action struck out.”

The defendants rely upon the generality of those words. However, there is nothing in the judgments to suggest that it is open to strike out a case for abuse of process in the absence of intentional and contumelious default or inordinate and inexcusable delay. In Janov there were both.

That was the very matter in issue in Gardner v. Southwark London Borough Council (No 2) [1996] 1 WLR 561 (CA), where Janov was not followed. Gardner concerned three personal injury claims. Each had been struck out under CCR Ord 17, r. 11(9), whereby an action would be automatically struck out where there had been no request for a hearing for 15 months from close of pleadings. This was a novelty introduced to deal with “the scandal of delay” (at 566G). Each of the plaintiffs had commenced new actions within the limitation period, but each of the new actions had been struck out in turn, in reliance on Janov. The defendants’ submission to this court was similar to the submission before us with reference to CPR 7.6. Waite LJ records the submission there as follows (at 566H):

“He argues, further, that the categories of conduct which is frivolous, vexatious or abusive of the process are never closed (and for that, as I would have thought, incontestable proposition he cited a well-known authority). This is an opportunity, he says, to demonstrate the truth of that, by backing the new regime with a stern use of the court’s powers to expand the definition of abuse of process by condemning any attempts to renew an action which has deservedly suffered a discontinuance after the very generous period down to the guillotine date allowed under the rule has expired.”

That submission was roundly rejected, by Waite LJ at 567B/D and by Sir Thomas Bingham MR in his short judgment at 569/570, all of which is worth noting, while the following is particularly pertinent:

“The object of the new procedural regime, as counsel for the defendants have urged, is quite plain. It has been described in earlier cases. It is intended to encourage the expeditious conduct of litigation and strongly to discourage delay. But, as it seems to me, a plaintiff who for reasons of negligence, dilatoriness, lethargy or mistake fails to apply for a hearing date before the guillotine date and so suffers the consequences of Ord. 17, r. 11(9), cannot be treated as if he were guilty of wilful or contumacious disobedience. The rules do not vary the ordinary rules which the court has habitually observed, and nothing short of a clear provision should, in my judgment, deprive a plaintiff of what is otherwise a potentially important right.”

Other than for the facts (a) that failure to request a hearing during a fifteen month period following the close of pleadings seems to me an arguably worse example of negligence or dilatoriness, and a more serious misuse of the court’s process, than a failure to serve within the four months allowed at the very outset of litigation, and (b) that the existence of a still running limitation period there has to be contrasted with the recent expiry of the limitation period accompanied by the statutory discretion of section 33 here, those three cases are close to the essential issue raised by these appeals. Indeed, the respondents have had to submit that where, as here, their case is that the abuse of process on which they rely is essentially a fault of the litigant vis-à-vis the court, it ought to make no difference whether the second action is commenced in time or not. In either case, the abuse involved in having lost the first action by reason of failure to serve in time is what must destroy the right to commence suit again. That argument would not, I think, have appealed to this court in Gardner.

Arbuthnot Latham Bank Ltd v. Trafalgar Holdings Ltd [1998] 1 WLR 1426 (CA), decided on the eve of the CPR regime, reflects CPR considerations in its application of the Janov principle to a second action, commenced within time, after a first action had been struck out. Like Janov, it did not concern personal injury. However, it is noticeable that this court did not assume that abuse of process in the first action would necessarily lead to the loss of the second action, although some special reason would need to be shown to justify it being allowed to proceed. It also recognised that for an action to be struck out as an abuse required either an intentional and contumelious default or inordinate and inexcusable delay, albeit the former category might be supplied by a “wholesale disregard of the rules”. Lord Woolf MR, the architect of the new regime, said this (at 1436G/1437A):

“We think that the change in culture which is already taking place will enable courts to recognise for the future, more readily than heretofore, that a wholesale disregard of the rules is an abuse of process as suggested by Parker L.J. in Culbert v. Stephen G. Westwell & Co. Ltd. [1993] P.I.Q.R P54.

While an abuse of process can be within the first category identified in Birkett v. James [1978] A.C. 297 it is also a separate ground for striking out or staying an action (see Grovit v. Doctor at pp. 642-643) which does not depend on the need to show prejudice to the defendant or that a fair trial is no longer possible. The more ready recognition that wholesale failure, as such, to comply with the rules justifies an action being struck out, as long as it is just to do so, will avoid much time and expense being incurred in investigation [of] questions of prejudice, and allow the striking out of actions whether or not the limitation period has expired. The question whether a fresh action can be commenced will then be a matter for the discretion of the court when considering any application to strike out that action, and any excuse given for the misconduct of the previous action: see Janov v. Morris [1981] 1 W.L.R. 1389. The position is the same as it is under the first limb of Birkett v. James. In exercising its discretion as to whether to strike out the second action, that court should start with the assumption that if a party has had one action struck out for abuse of process some special reason has to be identified to justify a second action being allowed to proceed.”

Securum Finance Ltd v. Ashton [2001] Ch 291 (CA) is the first of the CPR cases to consider the question of abuse in the context of a second action. It concerned a claim for the repayment of a loan. Janov was cited but not referred to in the judgments. The first action had been struck out for want of prosecution and abuse of process. A second action, within the limitation period, raised both the claim for repayment of the loan which had been litigated in the first action, but also a claim to enforce the security under a legal charge which had not and could not have been the subject matter of the first action. This court, agreeing with the judge below, declined to strike out the second action.

Chadwick LJ referred to Lord Woolf’s observations in Arbuthnot (and many others since) about a “change in culture” (at 308). He continued:

“34 For my part, I think that the time has come for this court to hold that the “change of culture” which has taken place in the last three years – and, in particular, the advent of the Civil Procedure Rules – has led to a position in which it is no longer open to a litigant whose action has been struck out on the grounds of inordinate and inexcusable delay to rely on the principle that a second action commenced within the limitation period will not be struck out save in exceptional cases. The position, now, is that the court must address the application to strike out the second action with the overriding objective of the Civil Procedure Rules in mind – and must consider whether the claimant’s wish to have a “second bite of the cherry” outweighs the need to allot its own limited resources to other cases. The courts should now follow the guidance given by this court in the Arbuthnot Latham case.”

In a subsequent passage, Chadwick LJ further explained (at 315E): “It is an abuse because it is a misuse of the court’s limited resources”. Rattee J agreed.

The respondents relied heavily on Arbuthnot and Securum. However, the question is whether the failure to serve in time is really comparable to a case where the first action has been struck out for want of prosecution and abuse of process. If Chadwick LJ intended his remarks to cover a much wider range of case in which the first action has lapsed, then that proposition will have to be made good. He was expressly referring to inordinate and inexcusable delay. In such a case the claimant is truly attempting a second bite at the cherry. In our case, the claimants have indeed invoked the court’s jurisdiction, but without even serving the claim form it might be said that they had not managed even a single bite.

In Aldi Stores Ltd v. WSP Group plc [2007] EWCA Civ 1260, [2008] 1 WLR 748 and Stuart v. Goldberg Linde [2008] EWCA Civ 2, [2008] 1 WLR 823, the facts of which are not relevant, the question arose, in the context of an appeal, as to whether the finding of abuse of process is a matter of discretion or not. It was held that, although the decision whether to strike out a claim on the ground of abuse is ultimately a matter for the court’s discretion, the finding of abuse was a judgment which was either right or wrong, albeit it involved an assessment of a large number of factors and might for that reason be hard for an appellate court to dislodge: see Thomas LJ at [16] and Longmore LJ at [38] in Aldi Stores, and Lloyd LJ at [24] and Sir Anthony Clarke MR at [81] in Stuart. The finding of abuse is of course likely to result in a discretionary decision to strike out for abuse, but the dicta in Arbuthnot and Securum indicate and the latter case illustrates the fact that, where a first action has been struck out for abuse of process, it does not necessarily follow that a second action would be.

It is now necessary to consider the line of jurisprudence which CPR 7.6 has engendered. In effect, the respondents’ submission is that it is the combination of two lines of authority, the Janov line concerning abuse of process, and the Vinos line concerning the strictness of the CPR rules relating to service, that produce the answer in their favour which they are defending here. The Vinos line demonstrates that a failure to serve within the four months allowed is or is tantamount to an abuse of process in the first action which debars the claimant from any right to rely on section 33 in a second action.

In Vinos v. Marks & Spencer plc [2001] 3 All ER 784 (CA) a personal injury claimant’s solicitor issued a protective claim form, after lengthy negotiations with the defendant, a week before the expiry of the limitation period, and then failed to serve it within the four months allowed. Service was 9 days late. The claimant applied, out of time, for an extension of time. He recognised that CPR 7.6(3) could not assist him, but he relied on CPR 3.10 as an alternative route. He failed. CPR 3.10 could not be used to circumvent the plain words of CPR 7.6(3). There is nothing surprising about that. However, May LJ supported the inevitable result with these observations (at 790):

“The merits of Mr Vinos’ particular case are not relevant to that question…Criticism of Mr Vinos’ solicitors in this case may be muted and limited to one error capable of being represented as small; but there are statutory limitation periods for bringing proceedings. It is unsatisfactory with a personal injury claim to allow almost three years to elapse and to start proceedings at the very last moment. If you do, it is in my judgment generally in accordance with the overriding objective that you should be required to progress the proceedings speedily and within time limits. Four months is in most cases more than adequate for serving a claim form. There is nothing unjust in a system which says that, if you leave issuing proceedings to the last moment and then do not comply with this particular time requirement and do not satisfy the conditions in r.7.6(3), your claim is lost and a new claim will be statute-barred. You have had three years and four months to get things in order. Sensible negotiations are to be encouraged but protracted negotiations generally are not.”

There was no reference there to section 33: but assuming May LJ had section 33 in mind, this was of course the era of Walkley and therefore loss of the first action at the end of the limitation period was the end of the line. If that was May LJ’s view of Walkley, then that was that. In any event, as Firman showed back in 1978, subject to section 33, it had always been the case that a failure to serve in time, at the end of the limitation period, was disastrous.

Godwin v. Swindon Borough Council [2001] EWCA 1478, [2002] 1 WLR 997 followed not much later and related to the technical question of when postal service was deemed to be effective under CPR 6.7. The defendant received the claim form in time, but deemed service was out of time. The rule prevailed and the claim was lost. The details of the discussion about rule 6.7 are of no concern for our cases. It was a personal injury claim. May LJ adapted his remarks from Vinos somewhat, saying –

“27.

This appeal is, of course, centrally concerned with limitation. The relevant time period after which the present action may not be brought is three years from the date on which the cause of action accrued, that is the date of the accident. For relevant purposes, the claim is brought by issuing the claim form. This was done in the present case just before the expiry of the three-year limitation period. In addition, the claim form has to be served in accordance with the rules. A failure to do so which cannot be rectified by one means or another will mean that the claim cannot proceed and often that a new claim will be statute-barred. Those who delay serving their claim to the last moment risk disaster.”

Thus in Vinos May LJ said that with the failure to serve in time “a new claim will be statute-barred” and in Godwin he said “often…a new claim will be statute-barred”. Perhaps he was thinking in Godwin about the exceptional cases in which Walkley had been distinguished. In any event, there is nothing here, in my judgment, any more than in Vinos that goes beyond the effect of the rules themselves, together with the background of Walkley which rendered the section 33 discretion irrelevant. There is certainly nothing here to suggest that failure to serve in time is tantamount to abuse. On the contrary, in Vinos May LJ had said that criticism of Mr Vinos’ solicitor may be muted and the single error capable of being represented as small. What he was at pains to point out however was that the consequences could be large.

The next case to be relied on by the respondents was Anderton v Clwyd County Council (No 2) [2002] EWCA Civ 933, [2002] 1 WLR 3174. This concerned five personal injury cases where there had been various problems with timely service. Several of them concerned technical aspects of the CPR 6.7 rules for deemed service, as in Godwin. Two of the claimants succeeded on the technical argument as to whether Saturdays and Sundays were excluded from the days of deemed service. One concerned a claim where service had been dispensed with, a dispensation which was upheld. For one reason or another, all five claimants succeeded. Reference was nevertheless made to the strictness of the rules. Lord Phillips of Worth Matravers MR said (at 3184E):

“Now that the disputed interpretations of the Civil Procedure Rules have been resolved by Godwin’s case and by this judgment, there will be very few (if any) acceptable excuses for future failures to observe the rules for service of a claim form. The courts will be entitled to adopt a strict approach, even though the consequences may sometimes appear to be harsh in individual cases.”

As for rule 6.9 and the ability to dispense with service, this court there concluded that the rule was wide enough to entitle a court to dispense even retrospectively with service, even though most applications would be prospective and most retrospective applications would be caught by the reasoning of Godwin’s case, that other rules could not be used to circumvent CPR 7.6(3). The exceptional kind of case illustrated by Lord Phillips’ judgment is one where a claimant has for some reason invalidly served the defendant by a permitted method of service but the defendant cannot deny that the claim form has been brought, in time, to his attention. “The basis of his application to dispense with service is that there is no point in requiring him to go through the motions of a second attempt to complete in law what he has already achieved in fact” (at [58]).

I can find nothing in Anderton to support the central thrust of the respondents’ thesis. Rule 7.6(3) is strict, and the consequences of failure to serve in time may therefore sometimes appear harsh. This is still in the era of Walkley. It is simply assumed that a new claim will not be possible. There was no attempt to bring one. All the argument concerned whether the first claims failed or not. None of them did. There is no suggestion of any abuse on the part of claimants.

Wilkey v. British Broadcasting Corpn [2002] EWCA Civ 1561, [2003] 1 WLR 1 concerned a claim in libel. By the deemed service rules, the claim form was served one day late, but the claim form had in fact been received by the defendant in time. Service was therefore dispensed with, and Anderton applied. However, this was limited to pre-Anderton cases, for after the clarification of rule 6.7 provided in Anderton, there was no further room for dispensation from the useful certainty of the deemed service rules. As Carnwath LJ said (at [28]]: “the rules of service in the CPR constitute a strict regime, introduced in the interests of certainty”. This has, to my mind, nothing to do with any question of abuse of process.

Hashtroodi v. Hancock [2004] EWCA Civ 652, [2004] 1 WLR 3206 is said by Mr Purchas to be the or at least a linchpin of his case: and he submits that Horton is undermined because Hashtroodi was not there cited to the House of Lords. The claim was for personal injury, but was lost because, through solicitor’s negligence, the claim form had not been served in time. The difference in that case, however, was that there was an application within the four months, just, to extend time, which was granted ex parte. Even so, service was not effected, for the documents got lost when sent by DX within the time so extended. However, the case was decided on the basis that the claimant had no good reason for requesting an extension of time in the first place, and the ex parte order extending time was subsequently set aside. Therefore there was no service within the original four months and no extension. The faults of the solicitor were held to be “particularly egregious” (at [36]) because, it seems, his failure to serve within time was not by oversight, but he deliberately left it until the last moment to ask for an extension in the hope that he could finally succeed in persuading the defendant’s insurers to appoint solicitors. He knew the defendant’s address, and even at that late stage could have served the defendant at home, within time.

The decision is contained in the following reasoning in the judgment of the court given by Dyson LJ:

“35 It follows that this is a case where there is no reason for the failure to serve other than the incompetence of the claimant’s legal representatives. Although this is not an absolute bar, it is a powerful reason for refusing to grant an extension of time. Despite this, Mr Gore submits that an extension of time should be granted. In relation to the application of the overriding objective, he relies on the following factors. First, the claim is very substantial. Secondly, the issues in the case were identified early on, so that a short extension of time would not undermine the case management process. Thirdly, the extension of time would not put the parties on a more or less equal footing than they would have been if the extension were not granted. Fourthly, the extension would not increase the cost of the litigation. Fifthly, it would be disproportionate to refuse the extension. Finally, the defendant has not suffered any prejudice as a result of the extension, since at the date of the claimant’s application, the defendant had not yet acquired an accrued limitation defence.

36 We are in no doubt that the time for serving the claim form should not be extended in this case. The absence of any explanation for the failure to serve is, on the facts of this case, decisive. Sadly, the errors on the part of Mr Pike were particularly egregious. The other factors identified by Mr Gore are not sufficient to outweigh the complete absence of any reason which might go some way to excusing the failure to serve in time. If we were to grant an extension of time in the present case, it seems to us that the rule stated in CPR r. 7.5 would cease to be the general rule. Moreover, there would be a real risk that statements made by this court about the importance of the need to observe time limits would not be taken seriously. That would be most unfortunate.”

Thus the facts in Hashtroodi were held to be at the extreme end of a negligent failure to serve. In the circumstances the court did not need to and did not rely on the ultimate failure to serve within the time extended: see at [38]. The defendant had applied for that latter reason to strike out the claim under CPR 3.4, but that application was not reached by this court. It had been refused by the deputy master. It was sufficient that the extension of time for service was held to be made in error and set aside (see at [39]).

Dyson LJ had also introduced the issue with a more general discussion of the topic of extending time under CPR 7.6(2). One of the questions discussed was whether the RSC doctrine of requiring a “good reason” as a condition precedent for an extension was similarly required under CPR 7.6(2). This court rejected that submission, in essence on the basis that the CPR should be interpreted as a new code on its own, without recourse to cases on or the doctrine of the RSC. However, the ultimate position reached was that, although a good reason for an extension was not formally required, in practice an extension would not be granted without one, and in this respect a deliberate, but incompetent, failure to serve could not suffice, but on the contrary militated against extension. Thus by the time of Hoddinott [2008] 1 WLR 806, Dyson LJ was able to say (at 821E) that –

“the court is unlikely to grant an extension of time for service of the claim form under CPR r 7.6(2) if no good reason has been shown for the failure to serve within the four months’ period.”

In Hashtroodi, however, this court was anxious to emphasise that the circumstances of failure relied on were (not merely not a good reason but) a bad reason for extending time for service. Thus Dyson LJ said this:

“19 Whereas, under the previous law, a plaintiff who was unable to show a good reason for not serving in time failed at the threshold, under the CPR a more calibrated approach is to be adopted. If there is a very good reason for the failure to serve the claim form within the specified period, then an extension of time will usually be granted…The weaker the reason, the more likely the court will be to refuse to grant the extension.

20 If the reason why the claimant has not served the claim form within the specified period is that he (or his legal representative) simply overlooked the matter, that will be a strong reason for the court refusing to grant an extension of time for service. One of the important aims of the Woolf reforms was to introduce more discipline into the conduct of civil litigation. One of the ways of achieving this is to insist that time limits be adhered to unless there is good reason for a departure. In the Biguzzi case [1999] 1 WLR 1926 Lord Woolf MR said, at p 1933: “If the court were to ignore delays which occur, then undoubtedly there will be a return to the previous culture of regarding time limits as being unimportant.”

21 It is easy enough to take the view that justice requires a short extension of time to be granted even where the reason for the failure to serve is the incompetence of the claimant’s solicitor, especially if the claim is substantial. But it should not be overlooked that there is a three-year limitation period for personal injury claims, and a claimant has four months in which to serve his or her claim form. Moreover, the claim form does not have to contain full details of the claim. All that is required is a concise statement of the nature of the claim: see CPR r 16.2(1)(a). These are generous time-limits.”

Dyson LJ then referred to what May LJ had said in Vinos, cited above.

However, in this respect, given that the negligent failure to serve in time was a bad rather than a good reason for an extension of time, there was no essential difference between the jurisprudence of the CPR and that of the RSC regimes, as the three cases in Firman as well as the cases in Walkley demonstrate.

The next case in this series was Collier v. Williams [2006] EWCA Civ 20, [2006] 1 WLR 1945, where four further cases concerning errors in service were considered. One of them was Leeson v. Marsden (at 1981/7). Mr Leeson’s appeal did not succeed, and so his first action failed. I shall refer to the fate of Mr Leeson’s second action below. This court, in its judgment again given by Dyson LJ, reiterated its approach to the question of time extensions in Hashtroodi (at 1986/7) and commented (at [133]): “It is no answer to say, as District Judge Singleton said in this case, that to insist on service within the time specified is to adopt a “technical” approach. In a sense, all rules may be said to be “technical”.”

It is at this point in the story that Horton was decided in the House of Lords, on 14 June 2006. Thereafter no further cases about the strictness of the service rules have been cited to us. On the contrary, we have been referred to a number of post-Horton cases on the position of second actions. It is to those that I now turn in the next section of this judgment. But just before I do so, I ask what is to be learned, for the purposes of the issue in these appeals, from the Vinos line of authorities which I have just reviewed.

In my judgment, there is nothing in that line to support the respondents’ submission that failure to serve a claim form in time is an abuse of process, or tantamount to one. It is nowhere said to be. What is said is that the rules are strict and will be strictly applied. The negligence of a claimant’s solicitor is no excuse. It is not a good reason for an extension, even where the extension is applied for in time. It is a bad reason, a reason for declining an extension. The strictness derives from the wording of CPR 7.6(3) with its expressly limited reasons (“only if”) for allowing an extension out of time, but also from the philosophy that time limits are to be strictly enforced. Otherwise there is nothing in the rules themselves to require such strictness that the balance of prejudice to the parties, and thus it might be said the ultimate justice of the case, are to be ignored. That is a matter of public policy, in the belief that more damage will in the long run be done in the generality of cases not only to the administration of justice as a whole, let alone to judicial resources, but also to the fair adjudication of disputes. It is a policy which, like the law of limitation itself, is in the public interest, even if it produces unfortunate results in individual cases. It is mitigated by the claimant’s remedy against his solicitor, where he acts through one. Because it is a strict policy, it may become technical. Where negligence is the cause of the service failure, therefore, there is no need for an individual balancing of the factors elsewhere spoken of in the CPR themselves, whether of the factors in rule 1.1(1) or rule 1.4(2), or in rule 3.9(1). But it should not be supposed that this attitude of the courts is new, although there may be differences at the margin. The overall public policy towards negligent failure of service has not changed since the pre-CPR days of Firman and Walkley. Nevertheless, no authority has ever suggested that such conduct was an abuse of process in itself. Even May LJ was prepared to regard a typical case of negligent oversight of the need to serve in time as capable of giving rise to only “muted” criticism.

As for second actions where the first has failed due to lack of timely service, these cases say nothing that goes beyond an explicit acceptance that if the claimant, having lost the benefit of his first action, is then out of time to start a second action, he has lost his claim. There is nothing whatsoever in these cases to suggest that, if there is still time to start a new action, it cannot be done. All the background, discussed in Birkett v. James and the Janov line of authority, is to the effect that loss of a first action for reasons otherwise than on the merits is no bar to a second action within time save where there has been conduct which can be described as an abuse of process: whether such conduct is intentional and contumelious, or a want of prosecution, or wholesale disregard of rules of court.

So, in the post-Horton universe, the clash, as it seems to me, where negligent failure to serve in time is in question, is not between an abusive claimant and the section 33 discretion, but between on the one hand a judicial doctrine, founded in strict rules governing service, which does not take account of the individual circumstances of each case or of the respective prejudice to each party, and on the other hand a statutory discretion granted to a personal injury claimant the essence of which is to take account of all the circumstances of the individual case including in particular the respective prejudice to each party. It is common ground that where no first action has ever been started within the limitation period, section 33 is operative. The respondents submit nevertheless that where a first action has been lost through a negligent failure to serve the claim form in time, the section 33 discretion, although theoretically operative, is never reached. That, it seems to me, takes us straight back to the anomaly of Walkley, with its windfall for the defendant,which Horton corrected, but in circumstances where section 33 does operate on its own terms, as Horton now determines.

In these circumstances, it is interesting to see how the post-Horton cases have dealt with these issues.

We have been cited three cases which follow Horton.

The first is Leeson v. Marsden [2008] EWHC 1011, 13 May 2008, Cox J. Susan Leeson was one of the claimants who lost their first actions in the authority which goes under the name of Collier. She started a second action. Her injury was suffered in December 2000. Pre-action disclosure was sought in March 2001. In August 2002 there was a formal, detailed, letter before action. Her first action was commenced on 24 November 2003, but the claim form was served a day late on 25 March 2004. In the meantime the defendant had been notified of it. In January 2006 she lost her appeal in Collier. She gave instructions to sue her solicitors. Then on 14 June 2006 the House of Lords gave judgment in Horton. She issued her second claim on 19 December 2006, three years out of time. Her second claim form was served on 25 January 2007. Two issues were raised, on Ms Leeson’s application to disapply limitation under section 33, and on the defendant’s application to strike out the second action as an abuse of process. The judge dealt with the issue of abuse first.

The defendant’s primary reliance was on Securum and the misuse of court resources. The Vinos/Hashtroodi line of authority was not cited. The judge asked defendant’s counsel what Ms Leeson should have done differently. The answer was that she should have abandoned her first action as soon as the error was discovered and not sought through litigation to persuade the courts to extend time (at para 84). The defendant’s counsel said it was a matter for the court’s discretion whether the second claim should be struck out as an abuse (at para 88). The judge knew that the finding of an abuse itself was not a matter of discretion, because she cited Aldi and Stuart. Her conclusion was that (a) there was no abuse, and (b) in any event the discretionary question of striking out for abuse could not be decided independently of the section 33 discretion (para 110). She said:

“91 The Defendants do not submit, in this case, that the mere issuing of a second claim was itself an abuse. Their argument is based entirely on the Securum principle. This requires the Court to consider whether the second claim is an appropriate use of the Court’s resources in circumstances where she has already had a share of those resources in the first claim and where her right to bring a second claim must be weighed in the balance with the rights of other litigants to use those resources. There is, in my view, nothing to prevent the Court having regard to those public interest considerations under section 33 which, as was said in Firman and re-emphasised in Horton, gives the Court the widest discretion in deciding whether to allow a claim to proceed, upon which further judicial fetters should not be imposed…

93 In any event as Mr Oppenheim pointed out Parliament has decided, through section 33, that the Court should have an unfettered discretion, subject only to considering those matters spelt out in sub-section (3), in deciding whether to allow a personal injury claim to proceed, notwithstanding the fact that it has been brought out of time. Section 33 therefore provides the means by which the Court protects itself from abuse arising from such claims. The Defendants acknowledge that, even if they succeeded in striking out this claim as an abuse, a further, professional negligence claim against her previous solicitors, in which the Claimant will be allocated a further share of Court resources, will follow. The real question is whether the Claimant should be permitted to continue with this claim against these Defendants. That question should be determined, in my judgment, under section 33 and not as a separate question relating to abuse of process.”

The judge also decided the section 33 discretion in Ms Leeson’s favour.

The defendants submit that this is a flawed, and wrong, authority. They say (1) that the judge was wrong to treat abuse as a matter of discretion; (2) that the Vinos/Hashtroodi line of authority was not deployed; and (3) that their case on abuse is wider than that advanced there and essentially depends on abuse within the first action, and they do not say that the mere issue of a second claim is not an abuse.

In my judgment, these submissions are wide of the mark. (1) Cox J did not treat abuse as a matter of discretion. She applied the learning in Aldi and Stuart. She found no abuse. (2) The Vinos/Hashtroodi line of authority was not relied on, but it does not assist the defendants. (3) On essentially identical facts, the same essential claim to abuse is made: that a misuse of court resources in the first action constitutes an abuse both because the rule about service within four months was disregarded and improperly defended and because the claimants should not have a second bite at the cherry. I therefore regard Leeson as being an authority in the appellants’ favour, although not of course binding on this court. The argument there ranged over many days, employed leading counsel, and the judgment is detailed and careful.

The second authority is Cain v. Francis [2008] EWCA Civ 1451, [2009] QB 754. That concerned two separate personal injury claims in which the claimants relied on section 33. One of the claims, McKay v. Hamlani was made by way of second action after the loss of the first due to the failure to serve the claim form in time. The other involved a first action, one day out of time. Both claimants succeeded in this court. No abuse of process argument was raised in the McKay claim. This was despite the citation to the court of Leeson. Rightly or wrongly, no one therefore considered that the argument of abuse was there to be made. The authority is useful for illustrating the power of section 33 in a post-Horton world and the return to Firman principles. Where the defendant knows about the claim, and the prejudice is little, the incidence of a limitation defence can properly be described as a windfall. Lady Justice Smith said:

“57 It appears to me that there is now a long line of authority to support the proposition that, in a case where the defendant has had early notice of the claim, the accrual of a limitation defence should be regarded as a windfall and the prospect of its loss, by the exercise of the section 33 discretion, should be regarded as either no prejudice at all (see Firman v Ellis [1978] QB 886) or only a slight degree of prejudice: see Donovan v Gwentoys Ltd [1990] 1 WLR 472. It is true that, in Thompson v Brown [1981] 1 WLR 744, Lord Diplock said that the accrual of the defence might be regarded as a windfall only where the delay in issuing proceedings was short. However, with great respect, it does not seem to me that the length of the delay can be, of itself, a deciding factor. It is whether the defendant has suffered any evidential or other forensic prejudice which should make the difference.”

Thirdly, McDonnell v. Walker [2009] EWCA 1257 also concerned a second action. The first action was lost because the solicitor incorrectly diarised the last date for service as 26 August, when it should have been a few days earlier. Service was in fact effected on 23 August, one day late. Lord Justice Waller said:

“11.

I should say at the outset that there is, as it seems to me, a slight tension between CPR 7.6 and its stringent terms not entitling a court to extend time in the first action even if no forensic prejudice has been suffered by a defendant, and having the power nevertheless to allow a second action to be commenced by using its discretion under s.33. But since the decision in Horton there is no doubt that there have been cases including McKay v Hamlani (considered by the Court of Appeal with Cain v Francis [2009] 3 WLR 551) in which time has been extended under s.33 in second actions where CPR 7.6 prevented an extension of time for service of a first action. Thus it cannot be said that in a CPR 7.6 case an extension of time for bringing a second action should never be granted, but it seems to me to be a relevant context and to at least show that it should not be easy for a claimant to commence a second action and obtain a disapplication of the limitation period under s.33.”

Subsequently, Waller LJ contrasted the type of case like Cain, where the defendant could not show any forensic prejudice at all with a case where forensic prejudice is suffered by a defendant who has not for many years been notified of a claim in any detail so as to enable him to investigate it (at paras 21/22). In McDonnell the discretion was exercised against the claimant: the defendant had received in the end claims quite different from anything notified before, some seven years after the accident, where there was an excuse for only 17 months of that delay. The defendant was clearly forensically disadvantaged by the claimant’s unexcused delay.

Both sides rely on these remarks. The defendants rely on Waller LJ’s cautionary words cited above. The claimants submit that McDonnell, on its facts, just as Cain and McKay on their facts, show how the section 33 discretion is well able to take account of all relevant circumstances, such as prejudice, delay and inexcusable delay, in arriving at the justice of the matter. I agree that these cases show section 33 working well, in both kinds of case. As for Waller LJ’s more general remarks about CPR 7.6 and section 33, I would be somewhat circumspect. CPR 7.6 is a general provision, whereas section 33 is specific to personal injury claims with their shorter time limit. If there is a tension, Horton, rather than the post-Horton cases like McKay, rules, on the highest authority, that the statute must be preferred. How, of course, the discretion will be exercised, will depend on the individual facts of each case. There is no reason to think that the defendant will be more prejudiced where the second action is out of time than where the first action is out of time, rather the contrary. And the claimant’s or his solicitor’s fault, generally speaking, is liable to be greater where the first action is out of time than where the second action is, at any rate where the second action is lost only by reason of an error over service.

Although it is a case in the Vinos/Hashtroodi line of authority, Hoddinott (which I have discussed at para 18 above) was not relied on by the defendants, being cited by Mr Ullstein. It also falls in the post-Horton era but, because it did not concern a personal injury claim, Horton has no particular relevance. Nevertheless, it is an interesting authority because, uniquely of the Vinos/Hashtroodi line cited, it concerned a failure to serve in time where the limitation period had not expired, indeed the period in question was 12 years. As explained above, the claimant succeeded in preserving his claim because, although he should not have been allowed to extend time for service, the defendant had waived the jurisdictional defect. Nevertheless, Dyson LJ ventured some observations on the significance for a CPR 7.6(2) extension application of the fact that the claimant could start afresh.

It is worth citing Dyson LJ’s remarks in extenso, for they throw light on the contention that a failure to serve in time is tantamount to an abuse which should prevent a second action even where that action would be in time (just as in the Janov line of authority). He said:

“52 It is clear beyond doubt that the claim for breach of contract is not yet time-barred and will not be time-barred for several years. There is no basis for a contrary argument and the contrary does not seem to have been argued. Where there is doubt as to whether a claim has become time-barred since the date on which the claim form was issued, it is not appropriate to seek to resolve the issue on an application to extend the time for service or an application to set aside an extension of time for service. In such a case, the approach of the court should be to regard the fact that an extension of time might “disturb a defendant who is by now entitled to assume that his rights can no longer be disputed” as a matter of “considerable importance” when deciding whether or not to grant an extension of time for service: see Hashtroodi’s case, para 18.

53 But where it is clear that an extension of time beyond the four months’ period will not extend the time to a date when the claim has become time-barred, the considerations are quite different. In such a case, an extension of time does not deprive the defendant of any limitation advantage. Nevertheless, in our view the fact that a claim is clearly not time-barred is a relevant consideration to be taken into account in favour of the claimant when the court decides whether to grant an extension of time. But it is not determinative.

54 It is tempting to ask: what is the point in refusing to extend the time for service if the claimant can issue fresh proceedings? But service of the claim form serves three purposes. The first is to notify the defendant that the claimant has embarked on the formal process of litigation and to inform him of the nature of the claim. The second is to enable the defendant to participate in the process and have some say in the way in which the claim is prosecuted: until he has been served, the defendant may know that proceedings are likely to be issued, but he does not know for certain and he can do nothing to move things along. The third is to enable the court to control the litigation process. If extensions of time for serving pleadings or taking other steps are justified, they will be granted by the court. But until the claim is served, the court has no part to play in the proceedings. A key element of the Woolf reforms was to entrust the court with far more control over proceedings than it had exercised under the previous regime. The rules must be applied so as to give effect to the overriding objective: this includes dealing with a case so as to ensure so far as is practicable that cases are dealt with expeditiously and fairly: CPR r 1.1(2)(d). That is why the court is unlikely to grant an extension of time for service of the claim form under CPR r 7.6(2) if no good reason has been shown for the failure to serve within the four months’period.

55 Thus, there are good reasons why the court should scrutinise with care applications to grant an extension of time for service of the claim form. If an application for an extension of time is made after the end of the four months’ period, the application must be dismissed unless the three conditions specified in CPR r 7.6(3) are satisfied. In such a case, the fact that the claim is clearly not yet time-barred is irrelevant. It would be strange if, where an application is made before the end of the four months’ period, the fact that the claim is clearly not time-barred means that the court must grant an extension of time. In our view, the better view is that, where the application is made before the end of the four months’ period, the fact that the claim is clearly not yet time-barred is a relevant consideration. The district judge was right on this point.”

This is a closely reasoned passage, but for our purposes it seems to me that the following emerges. (1) Dyson LJ clearly envisages the possibility that a first action will fail because an extension of time to serve the claim form ought not to be granted, but the claim will survive because a new action can be started within the relevant limitation period: see, for instance, the first sentence of para 54. On the defendants’ submissions, this ought to be impossible, for the second action will always be an abuse of process, a second bite of the cherry, even if it is still in time. (2) Nevertheless, where and because a second action could be commenced within the limitation period, the court considering whether to extend time should take that matter into account. (3) The function of service and the rules governing it are most comprehensively explained (at para 54). Nothing there suggests that timely service, although important to the administration of justice generally and therefore something for strict control, is such that a mere, short lived, oversight amounts to an abuse of the process of the court. Indeed, Dyson LJ even goes so far as to say that until service “the court has no part to play in the proceedings”. That is not conducive to a line of reasoning that the interference with judicial resources is of such importance that mere oversight in timely service is an abuse of process. (4) Where it is clear that a time bar has not expired, that fact may be taken into account. Where, however, it is uncertain, the status of limitation should be left to the second action, if any. Where, however, it is clear that limitation has expired with the failure of service, subject to any extension, that fact should count in the defendant’s favour.

It follows to my mind that where there is a section 33 possibility in a second action, that matter should always be left to the second action. Moreover, since the second action may yet be in time because of section 33, there is an uncertain dividing line between the second action which is in time and the second action which is out of time. The defendant is entitled to say in the first action, however, that, subject to any extension, he is entitled to retain any section 11 time-bar which has prima facie accrued, and that should count in his favour on any question of an extension. If, however, the claimant does not need section 33 because he is still in time to start a second action, then the loss of his first action should not in any event debar him from beginning again.

It is time to draw the strings together. In my judgment, the abuse of process argument relied upon by the defendants to deny the claimants an opportunity to rely on section 33 of the 1980 Act for the purpose of bringing within time a second action which would otherwise be time barred under section 11 of that Act fails. It does so for the following reasons, which have been more or less anticipated in the preceding analysis.

A mere negligent failure to serve a claim form in time for the purposes of CPR 7.5/6 is not an abuse of process. It has never been held to be in any of the many cases cited to this court, nor in my judgment should it be described as such, nor as being tantamount to such. I say a “mere” negligent failure to serve in time in order to distinguish the typical case of such failure to be found in these appeals and many other cases in the reports from any more serious disregard of the rules; but not in order to be in any way dismissive of the proper strictness with which a failure to serve in time, without good reason for doing so, is and has been rigorously dealt with by the courts, whether under the CPR or under the previous regime of the RSC. However, all the cases make clear that for a matter to be an abuse of process, something more than a single negligent oversight in timely service is required: the various expressions which have been used are inordinate and inexcusable delay, intentional and contumelious default, or at least wholesale disregard of the rules.

The reason why failure to serve in time has always been dealt with strictly (even if CPR 7.6(3) represents a still further tightening of the rules where a retrospective request for an extension is made out of time) is in my judgment bound up with the fact that in England, unlike (all or most) civil law jurisdictions, proceedings are commenced when issued and not when served. However, it is not until service that a defendant has been given proper notice of the proceedings in question. Therefore, the additional time between issue and service is, in a way, an extension of the limitation period. A claimant can issue proceedings on the last day of the limitation period and can still, whatever risks he takes in doing so, enjoy a further four month period until service, and his proceedings will still be in time. In such a system, it is important therefore that the courts strictly regulate the period granted for service. If it were otherwise, the statutory limitation period could be made elastic at the whim or sloppiness of the claimant or his solicitors. For the same reason, the argument that if late service were not permitted, the claimant would lose his claim, because it would become time barred, becomes a barren excuse. But even where the claimant is well within the limitation period despite his delay in serving, there is a clear public interest in the rules and the courts curtailing the efficacy of a claim form which, because it has not been served, is not very different from an unposted letter. Therefore, the strictness with which the time for service is supervised has entirely valid public interest underpinnings which are quite separate from the doctrine of abuse of process. It is sufficient for the rules to provide for service within a specified time and for the courts to require claimants to adhere strictly to that time limit or else timeously provide a good reason for some dispensation. There is no need for that procedure to be muddled up with the different doctrine of abuse of process.

There is of course the (possibly) new argument in the era of the CPR which emphasises the importance of any misuse of court resources. It is well to be aware of the important public interest bound up in the efficient use of those limited resources. However, to seek to turn that proper concern, in such a case as these, into a surrogate for the doctrine of abuse of process is to my mind a disciplinarian view of the law of civil procedure which risks overlooking the overriding need to do justice. Certainly, the authorities have not gone that far, and there is nothing in the CPR themselves to indicate that a mere failure to serve in time is to be regarded as an abuse requiring or deserving anything further than the failure of the claim form itself – with the vital consequence in the absence of section 33 of losing a claim which has become time barred. Moreover, it should not be forgotten that one of the great virtues of the CPR is that, by providing more flexible remedies for breaches of rules as well as a stricter regulatory environment, the courts are given the powers and the opportunities to make the sanction fit the breach. That is the teaching of one of the most important early decisions on the CPR to be found in Biguzzi v. Rank Leisure Plc [1999] 1 WLR 1926 (CA).

Be that as it may, however, Parliament has in any event enacted that personal injury claims should be treated in a special way. They have a shorter limitation period than the general run of tortious and contractual claims, but, at the same time, there is a special discretion, now to be found in section 33 of the 1980 Act, to displace the three year time limit. That discretion arises, if invoked, whenever a personal injury claim is commenced outside that three year limit. For these purposes it does not matter whether the claim in question is brought in an action for the first time commenced outside the three year limit, or is brought outside that limit in a second action after the first action has failed because, although commenced in time, the claim form was served too late or indeed never served: that is the teaching of Horton. In these circumstances, to say that a second action can never reach the section 33 discretion because the mere bringing of it is an abuse of process is to ignore the will of Parliament: that is what Firman taught in the era of the RSC and Horton has reverted to Firman in the era of the CPR. Moreover, the disputed submission would have the effect of resurrecting the Walkley anomaly.

By seeking to apply the Janov line of authority to a mere failure to effect timely service, the respondents acknowledge that it should make no difference, on the logic of their submission, whether there is still time, after the failure of the first action, to commence a second action before the expiry of a relevant limitation period. However, dicta in Gardner, Anderton, and Hoddinott are inconsistent with the view that a second action could not be started within a limitation period which was still running after the failure of a first action for lack of timely service. Gardner is particularly interesting in this respect, for the automatic strike-out generated by a failure to apply for a hearing after a delay of no less than 15 months from close of pleadings would appear to arise out of a much more serious breach than a failure to serve within four months of issue. The submission which was rejected in that case was very close to the submission which the respondents make in these appeals.

It is suggested that it is a sufficient reason to ignore the will of Parliament that otherwise there is no or no sufficient sanction for the failure to serve a claim form within the lime allowed. The submission is misconceived. It is for Parliament to say that the section 33 discretion cannot be invoked in a second action after the failure to serve in time in a first action. In Walkley the House of Lords considered that that was what Parliament had enacted: but in Horton the House ruled that that interpretation was in error and had, in justice, to be corrected. In any event, the failure of the first action is a sufficient sanction in itself, for it comes accompanied by costs, and, where a claim has in the meantime become time barred and is not covered by section 33, the loss of the claim itself.

It is suggested that the alternative remedy against a negligent solicitor supports the respondents’ doctrine. That may be an argument about the balance of prejudice, and it is clear from Horton itself that it is relevant to the section 33 discretion: but it cannot turn what is not an abuse of process into some kind of quasi-abuse. In any event, a remedy against negligent solicitors, although a necessary long-stop for the victim of negligence, is a generally unsatisfactory way of litigating the claimant’s claim, and does little or nothing for the courts’ limited resources.

I would therefore allow Mrs Aktas’ appeal in Aktas v. Adepta and remit her case to the trial court for a decision on the section 33 discretion. I would also allow Mr Dixie’s appeal against the decision to strike out his second action in Dixie v. British Polythene Industries plc and go on to consider the section 33 in his case, below.

A question might arise, although it does not in either of these cases, as to whether a second action, in a Janov situation of real abuse, could be stopped in its tracks by being struck out, or whether even so it would be necessary to filter that question through the overall section 33 discretion. In my judgment, in an appropriate case, the second action could be struck out for abuse of process without entering on the section 33 discretion. The courts are entitled to control access to them in a situation of real abuse, and logic suggests that if there is or has been abuse which, having been found, ought to disentitle a claimant from proceeding with his claim, then the courts are entitled, in the exercise of their discretion, to say so. Such an action would be struck out for abuse of process, and the section 33 issue, which is an issue in that action, would never be reached. If a second action which is started within the limitation period can be struck out for abuse of process in the first action, it ought to follow that a second action which is commenced out of time can also be struck out in limine for the same reason. However, in Arbuthnot Lord Woolf seems to suggest that the fact that there has been abuse in the first action is not an automatic bar to the commencement of a second action, although “some special reason has to be identified to justify a second action being allowed to proceed” (at 1437A). If, therefore, despite the finding of abuse of process, the exercise of the court’s discretion, whether to strike out the second action or not, becomes something which cannot be determined in the abstract, without considering the section 33 circumstances as a whole, then it might become necessary to consider the question of abuse of process as part and parcel of the “all the circumstances of the case” under section 33 (as Cox J did in Leeson v. Marsden).

In the event, Mr Rowley on behalf of BPI did not press his opposition to the exercise of the section 33 discretion in Mr Dixie’s favour to any great extent. The judge, HHJ Mitchell, rightly said that it was “difficult to imagine there could be any prejudice to the defendant in this case”. Mr Dixie had suffered a back injury at work on 27 February 2005. BPI was immediately aware of the injury as Mr Dixie went off work and BPI was obliged to investigate and report as soon as Mr Dixie had been off work for five days. There was a formal letter of claim within 6 months of the accident. Within a short period of seven weeks after that letter BPI admitted liability and requested Mr Dixie to make no further liability investigations. By the time that the first claim form was issued on 22 February 2008 there had been an exchange of medical reports (Mr Dixie had been examined by BPI’s medical expert) and a settlement offer from BPI. Although the protective claim form was not served, BPI was immediately informed of its issue. Time ran out for the service of the claim form about a week after Mr Dixie had been examined again on behalf of BPI.

The main focus of BPI’s argument before the judge was on the abuse of process point. Its skeleton argument below addressed the section 33 point only in its last two paragraphs. The judge, despite finding no prejudice, said he was likely to have decided the section 33 discretion against Mr Dixie, but not on the facts of the case but on the basis, which it is common ground was flawed, that section 33 was really concerned with a case where the claimant was unaware of the nature or extent of his injuries. But for that error, it is hard to think that the judge would not have exercised his discretion in Mr Dixie’s favour.

In BPI’s skeleton argument to this court, it has sought to suggest that the delay in Mr Dixie’s proceedings has had an effect on the cogency of the evidence in circumstances where Mr Dixie sought to return to work and suffered other incidents in May 2005 and thereafter (referred to in Mr Dixie’s GP notes) and where BPI’s consultant orthopaedic surgeon has suggested pre-existing degenerative changes. It is submitted that Mr Dixie’s claim for a 2-3 year acceleration of symptoms from a pre-existing condition is undermined by the further incidents post the February 2005 injury: and that this issue is bound to be more difficult to evaluate after 5 years, during which most of the delay has been the fault of Mr Dixie.

At the hearing, Mr Rowley submitted that in the light of Mr Dixie’s faults in litigation – even if those faults did not quite amount to abuse of process – even a small amount of forensic prejudice to BPI would suffice to have the section 33 discretion resolved in its favour. There was an issue whether Mr Dixie’s difficulties were really down to the February 2005 incident or to the May 2005 incident. Mr Methuen riposted to the effect that it was a simple case where the claimant tried to go back to work after his injury and found the work too heavy. The defendant was seeking to make a new case of prejudice “out of straw”.

In my judgment, apart from the loss of a limitation defence, there is no real prejudice to BPI in having to defend Mr Dixie’s claim. It is a good example of the “windfall” which may occur for the benefit of a defendant where the claimant fails to observe the necessary time limit for service. The only real delay which has occurred has been while the parties have been fighting the abuse of process point, which BPI has now lost. The case is nearly on all fours with McKay v. Hamlani which was considered as one of the two cases decided in this court under the title of Cain v. Francis. It is not like the case of McDonnell v. Walker, where the discretion was exercised against the claimant for the very reason that the claim as it had emerged seven years after the accident was quite different from anything that had been notified before and where the defendant was clearly disadvantaged by the claimant’s unexcused delay.

I would therefore exercise the section 33 discretion in Mr Dixie’s favour.

I would allow both appeals and set aside the orders for the striking out of the appellants’ second actions. In Mrs Aktas’ case, the section 33 discretion will have to be considered in the trial court. In Mr Dixie’s case, I would exercise the section 33 discretion in his favour and disapply the section 11 time bar.

It is a pity the law in this area has become so complicated since the introduction of the CPR, but I am in complete agreement with my Lord’s analysis and conclusions.

I also agree with Rix LJ’s conclusions and reasons. As Longmore LJ observes, it is a pity that this area of the law had become so complicated since the CPR. But the principles to be adopted in the present type of case should now be clear and simple to apply following Rix LJ’s comprehensive analysis.