Roberts v Gill & Co Solicitors and others [2010] UKSC 22

Easter Term[2010] UKSC 22On appeal from:[2008] EWCA Civ 803JUDGMENT
Between
RobertsAppellantGill & Co Solicitors and othersRespondents
Before
Lord Hope, Deputy PresidentLord RodgerLord WalkerLord CollinsLord ClarkeDate 19 May 2010Hearing Heard on 22 and 23 February 2010Leslie Blohm QC, Guy Adams (Instructed by Chilcotts) for AppellantTom Dumont (Instructed by Barlow Lyde & Gilbert LLP) for Respondent

Introduction

[1]This appeal concerns a claim by a beneficiary under a will for negligence against solicitors who, he claims, allowed his brother, also a beneficiary and then the administrator of the estate, to acquire and dispose of land which should have been part of the residuary estate. The claim was begun in a personal capacity, but it is now accepted that a claim that the solicitors owed a duty of care to beneficiaries would be difficult to sustain, and the claimant seeks to amend the proceedings to claim in a representative capacity on behalf of the estate. The events of which the claimant complains happened 13 or 14 years ago.[2]The principal questions on this appeal relate to whether this is an appropriate case for a representative (or derivative) claim, which was the focus of the judge’s decision, and to the interpretation and application of section 35 of the Limitation Act 1980 and the rules of court which were enacted pursuant to it, first in the Rules of the Supreme Court, and now in the Civil Procedure Rules. The limitation issues were the main focus of the decision of the Court of Appeal. Section 35 was enacted following recommendations of the Law Reform Committee in 1977. It had two main objectives. The first was to enable a plaintiff to amend pleadings out of time so as to sue in another capacity, in particular to reverse the effect of such decisions as Ingall v Moran [1944] KB 160, which created a grave injustice where proceedings were instituted under the Law Reform (Miscellaneous Provisions) Act 1934 prior to letters of administration being taken out and the limitation period expired before proceedings were instituted in a representative capacity: the grant did not date back to the date of death, by then it was too late to issue fresh proceedings or to amend. The second objective was to enable parties to be added out of time, in cases where joinder of the new party was necessary if the plaintiff’s claim was to succeed, for example where the plaintiff was an equitable assignee and had omitted to join the assignor prior to the expiry of the limitation period.[3]But section 35 has been described as being “without doubt one of the most convoluted provisions in the entire law of limitations” (McGee, Limitation Periods, 5th ed 2006, para 23.003). That is no doubt why there have been more than 25 decisions of the Court of Appeal on section 35 and the rules of court.

The background

[4]Mrs. Alice Margot Roberts (“Mrs Roberts”), of Lower Hellingtown in Devon, made a will on March 6, 1992. In September 1994 a receiver was appointed by the Court of Protection to administer her affairs. She died on July 27, 1995. Her grandson, Mark Roberts, the appellant, is one of the three equal residuary beneficiaries of the estate of Mrs Roberts. The other residuary beneficiaries are his brother, John Roberts, and his aunt, Ms Jill Roberts.[5]The executors named in the will (Mrs Roberts’ solicitor and an accountant) renounced their right to probate and John Roberts was granted letters of administration with will annexed on February 16, 1996.[6]Mrs Roberts’ will provided in clause 7 that if John Roberts within a specified time (the earlier of one month from demand by the trustees or twelve months from death) either paid, or provided security or an indemnity to the will trustees for, all of the estate and other duty arising on her death in respect of her estate then(i) a piece of land known as the Coppice would pass to Mark Roberts and(ii) the remainder of the property known as Lower Hellingtown Farm would pass to John Roberts. By clause 8, if the payment was not made or security/indemnity given then the properties would fall into residue. Consequently (because of the value of the farm) if John Roberts complied with the condition in clause 7 the position would be much more favourable to him than if he did not, and the converse was true in relation to Mark Roberts.[7]John Roberts paid some inheritance tax in order to obtain the grant of letters of administration. He does not appear to have paid the remaining inheritance tax due, which may amount to some £60,000 and with interest would substantially exceed £100,000.[8]During the time John Roberts acted as administrator, he instructed two firms of solicitors, Gill & Co and Whitehead Vizard, the first and second defendants.[9]In July 1996 John Roberts, as personal representative, executed an assent to Lower Hellingtown Farm vesting in himself as beneficiary. The first defendants, Gill & Co, acted for him on the grant of letters of administration, and (it seems) on the preparation of the assent.[10]In or about 1997, Lower Hellingtown Farm was sold by John Roberts in two lots for a total of £305,166.19. Some £285,000 of the proceeds of sale were paid to John Roberts and the balance was used to discharge certain estate liabilities. Whitehead Vizard, the second defendants, acted for John Roberts on the sale of Lower Hellingtown Farm.[11]By order dated October 30, 2000, on the application of Mark Roberts, John Roberts was replaced as administrator of the estate by Mr Charles Sainter, a partner in the firm of solicitors then and now acting for Mark Roberts.[12]By a claim form dated November 27, 2002, issued in the Plymouth County Court, Mark Roberts started proceedings against the solicitors for breach of duty of care owed to him as beneficiary of Mrs Roberts’s estate. The particulars of claim alleged that:(1) the first firm, Gill & Co, were retained by John Roberts to advise him on matters arising from the appointment by the Court of Protection of a receiver for Mrs Roberts and, after her death, to obtain letters of administration and subsequently to assist and advise on the administration of her estate;(2) John Roberts instructed the second firm, Whitehead Vizard, from April 1997 to act on his behalf on the sale of Lower Hellingtown Farm;(3) inheritance tax payable by reason of the death of Mrs Roberts had never been paid and no security or indemnity for the same had been furnished by John Roberts;(4) inheritance tax on Mrs Roberts’ personal estate was paid at the beginning of February 1996, together with the first of ten annual instalments payable in respect of her real estate, but no further inheritance tax was paid thereafter, and the duty payable in respect of Mrs Roberts’ life interest under two will trusts had not been satisfied;(5) notwithstanding this, on July 23, 1996 a legal executive employed by Gill & Co prepared and witnessed the transfer of Lower Hellingtown Farm by John Roberts as trustee of the property to himself as beneficiary under the will;(6) in so doing, Gill & Co acted in breach of duty owed to Mark Roberts personally as beneficiary in the estate;(7) Whitehead Vizard were instructed by John Roberts to act on his behalf on the sale of the farm to a third party, which took place in 1997, and negligently and in breach of duty to Mark Roberts, Whitehead Vizard effected the sale of the farm when they knew, or ought to have known, that the inheritance tax had not been paid or secured and hence that John Roberts did not have good title to the farm;(8) by reason of the negligence and want of care of the two firms Mark Roberts suffered loss and damage in that but for their negligence the farm would have fallen into the residuary estate of which he is entitled to a one third share, and the remaining estate is insufficient to meet either the pecuniary legacies or to discharge the inheritance tax outstanding.[13]The foundation of the claim is that the estate had been administered on the false basis that the requirements of clause 7 of the will had been complied with so that clause 7 of the will operated and clause 8 of the will did not have the effect of putting the land referred to in clause 7 of the will into the residuary estate. There is no doubt that this was a claim by Mark Roberts personally for loss suffered by him as a beneficiary.[14]On January 30, 2003 an order was made staying the action to allow the pre-action protocol to be followed. On March 12, 2003 the solicitors for the defendants wrote to Mark Roberts’ solicitors rejecting the claim, because (they said) the law does not recognise a duty of care between a solicitor instructed by a personal representative and a beneficiary. In April 2003, the parties agreed a general extension of time for the filing of a defence to enable Mark Roberts to respond to the letter from the defendants’ solicitors. A defence has not been filed.[15]Any claim by the personal representative of Mrs Roberts became statute-barred at the latest during 2003, six years after the sale of the farm by John Roberts.[16]By an application notice dated August 25, 2006 Mark Roberts applied to amend the proceedings, three years after the expiry of the limitation period, in order to continue them both in his personal capacity and as a derivative action on behalf of the estate.[17]The proposed amendments(1) describe Mark Roberts as suing on his own behalf and as representing the estate of Mrs Roberts; and(2) plead that (a) Gill & Co acted in breach of the duty of care owed not only to Mark Roberts but also to the estate of Mrs Roberts; (b) Whitehead Vizard were instructed, not only to act for John Roberts on the sale of the farm, but also to assist in the administration of Mrs Roberts’ estate; and (c) as a result of both firms’ negligence, the estate has suffered loss and damage in that the farm would have fallen into the residuary estate.[18]The proceedings were transferred to the Chancery Division. Mark Roberts has the benefit of a funding certificate issued by the Legal Services Commission in relation to these proceedings.[19]On April 4, 2007 Mr Paul Morgan QC (now Morgan J), sitting as a deputy judge of the Chancery Division, dismissed the application to amend on the ground that there were no special circumstances which would entitle Mark Roberts to bring a derivative action. Had he found special circumstances, he would have concluded that, notwithstanding that the limitation period for the personal claim had expired, the court would have been able to authorise Mark Roberts to bring a new derivative claim in a representative capacity different from his personal capacity.[20]On appeal the Court of Appeal (Pill and Arden LJJ and Patten J, with Arden LJ giving the only judgment on this aspect) held unanimously that, if Mark Roberts’ application to amend so as to plead a derivative claim were allowed, the administrator had to be joined as a party: [2008] EWCA Civ 803, [2009] 1 WLR 531. Since the limitation period had expired, joinder of the administrator could only be permitted if the addition were “necessary” (CPR 19.5(2)(b)) to enable the existing action to be pursued. The addition of the administrator was not necessary for the existing, personal, claim to be properly carried on, and permission to amend to plead the derivative claim only (without joining the administrator) was refused since that amendment would not enable the claimant to proceed to judgment on the derivative claim because the relevant parties had not been and could not be joined. This was a point not taken before the judge. But the Court of Appeal by a majority (Arden LJ, with whom Patten J agreed, Pill LJ dissenting) disagreed with the judge on what had been the main holding at first instance, and held that if the combined effect of the Limitation Act 1980 and the CPR had not been to prevent the amendment, there would have been such special circumstances as to justify a derivative claim.

The nature of Mark Roberts’ new claim

[21]It is clear that the two firms of solicitors did not owe duties in the circumstances to Mark Roberts as a beneficiary under the will. It is equally clear that any claim by Mr Sainter as administrator is statute-barred. But Mark Roberts wishes now to proceed on behalf of the estate and to take advantage of the fact that (as Lord Nicholls of Birkenhead put it in Royal Brunei Airlines Sdn Bhd v Tan [1995] 2 AC 378, 391) “for the most part [professionals] will owe to the trustees a duty to exercise reasonable skill and care. When that is so, the rights flowing from that duty form part of the trust property. As such they can be enforced by the beneficiaries in a suitable case if the trustees are unable or unwilling to do so.”[22]Mark Roberts seeks to bring a derivative action in his own name on behalf of the estate against a third party. The action is a derivative action in which the beneficiary stands in the place of the administrator and sues in right of the estate, and does not enforce duties owed to him rather than to the administrator. It has often been said that a beneficiary can bring a derivative action only in special circumstances: Hayim v Citibank NA[1987] AC 730, to which it will be necessary to revert.[23]The question on this appeal is whether Mark Roberts should be permitted to amend so as to put his claim as a derivative claim. That involves two further questions. The first question is whether the amendment can be made notwithstanding expiry of the limitation period in respect of his personal claim. The second question is whether, even if the expiry of the limitation period is not a bar to the necessary amendments, the claim is bound to fail because there are no special circumstances justifying a derivative action.

The Limitation Act 1980 and the rules of court

[24]The old rule of practice was that an amendment would not be allowed if it would prejudice the rights of the opposite party as existing at the date of the amendment; and in particular, an amendment should not be allowed so as to allow a plaintiff to set up a cause of action which would otherwise be barred by the Statutes of Limitation: Weldon v Neal (1887) 19 QBD 394, at 395, per Lord Esher MR. This principle applied to amendments consisting of joinder (or substitution) of parties: Mabro v Eagle, Star and British Dominions Insurance Co Ltd [1932] 1 KB 485; Davies v Elsby Brothers Ltd [1961] 1 WLR 170; Lucy v W T Henleys Telegraph Works Co Ltd [1970] 1 QB 393; Liff v Peasley [1980] 1 WLR 781.[25]RSC Ord 20, r 5 was added in 1964, and prior to the changes in the rules following the Limitation Act 1980, provided that the court could give leave to amend a writ or pleading in a number of cases, including an amendment to alter the capacity in which a party sued. Ord 20, r 5(4) provided:
“An amendment to alter the capacity in which a party sues (whether as plaintiff or as defendant by counterclaim) may be allowed under paragraph (2) if the capacity in which, if the amendment is made, the party will sue is one in which at the date of issue of the writ or the making of the counterclaim, as the case may be, he might have sued.”
Ord 20, r 5(2) provided that where an application to the court for leave to make such an amendment was made after any relevant period of limitation current at the date of the writ had expired the court might nevertheless grant leave if it thought it just to do so. Ord 20, r 5(5) also allowed the addition or substitution of a new cause of action if it arose out of the same or substantially the same facts as a cause of action in respect of which relief had already been claimed in the action.[26]But the Court of Appeal decided that the fact that in certain cases under Ord 20, r 5 amendments were to be permitted although the statutory period had run did not mean that, in cases falling outside the rule changes, there was any relaxation of the principle in Weldon v Neal: Braniff v Holland & Hannen and Cubitts (Southern) Ltd [1969] 1 WLR 1533 and Brickfield Properties Ltd v Newton [1971] 1 WLR 862, not following Chatsworth Investments Ltd v Cussins (Contractors) Ltd [1969] 1 WLR 1, at 5, per Lord Denning MR.[27]In 1977 the Law Reform Committee (chaired by Orr LJ and including, among others, Griffiths and Walton JJ, Mr T H Bingham QC and Mr E G Nugee QC, and Professor A G Guest) issued a Final Report on Limitation of Actions: Cmnd 6923. The Committee had been invited in 1971 to consider what changes to the law relating to limitation of actions was desirable. Part V of the Committee’s report was headed “Procedure” and dealt with questions arising when it was sought to alter the character or scope of an action after the limitation period had expired.[28]After referring to the power of the court to correct misnomer of parties in the then RSC Order 20, r 5(3), the Committee referred (para 5.17) to what it described as “not wholly dissimilar cases” where the existing rule might cause injustice, where the plaintiff had made an error of law or procedure, the correction of which would not occasion anyone to be taken by surprise. For example where an equitable assignee of a debt sued the debtor without joining the assignor and the limitation period then expired, he could not amend his pleading so as to join the assignor: Hudson v Fernyhough (1890) 34 SJ 288. The Committee identified (para 5.20) these cases, among others, in which a new party should be capable of being added by way of amendment after the limitation period:(1) where the plaintiff was beneficially entitled in equity, and the person with the legal title was a necessary party to the action, for example, the equitable assignee of a chose in action, who could not sue without joining the legal assignor: Performing Right Society Ltd v London Theatre of Varieties Ltd [1924] AC 1;(2) the cestui que trust, who could not enforce a right of action against a stranger to the trust without joining the trustee: Harmer v Armstrong [1934] Ch 65;(3) where the plaintiff was a shareholder suing to enforce a right vested in the company and the company was a necessary party to the action: Spokes v Grosvenor and West End Railway Terminus Hotel Co Ltd [1897] 2 QB 124, 128; cf Wallersteiner v Moir (No 2) [1975] QB 373. The Committee pointed out (para 5.27) that the common factors in these examples were that (1) the plaintiff’s action was not properly constituted unless the new party were joined; and (2) the plaintiff was not seeking any substantive relief against the new party.[29]The Committee recommended (para 5.25) that the Rules Committee should be given power to cover by rule specific cases falling within a formula embodied in primary legislation. It rejected the solution of legislating for the specific cases. It accepted that the necessary formulation would not be easy, and it accepted that it had not been able to devise any entirely satisfactory formula. It summarised its conclusions (paras 46-48), so far as material, in this way:(1) No change was required in the rules which enabled a new cause of action to be added out of time;(2) a plaintiff should be able to amend pleadings out of time so as to sue in another capacity (including that of administrator) and the rule-making powers should be extended for that purpose;(3) the rule-making powers should be enlarged so as to confer power to enable parties to be added out of time, in specific cases if (a) the plaintiff’s action was not properly constituted unless the new party were joined; and (b) the plaintiff was not seeking substantive relief against the new party, or if substantive relief was sought against the new party, joinder of the new party was necessary if the plaintiff’s claim against the defendant was to succeed.[30]The result of these recommendations (on which see Millett LJ in Yorkshire Regional Health Authority v Fairclough Building Ltd [1996] 1 WLR 210, 219) was section 35 of the Limitation Act 1980 and the consequent changes to the Rules of the Supreme Court.

Limitation Act 1980, section 35

[31]So far as material to this appeal, section 35 provides: “35.—New claims in pending actions: rules of court.(1) For the purposes of this Act, any new claim made in the course of any action shall be deemed to be a separate action and to have been commenced— (a) in the case of a new claim made in or by way of third party proceedings, on the date on which those proceedings were commenced; and (b) in the case of any other new claim, on the same date as the original action.(2) In this section a new claim means any claim by way of set-off or counterclaim, and any claim involving either— (a) the addition or substitution of a new cause of action; or (b) the addition or substitution of a new party; …(3) Except as provided by section 33 of this Act or by rules of court, neither the High Court nor any county court shall allow a new claim within subsection (1)(b) above, other than an original set-off or counterclaim, to be made in the course of any action after the expiry of any time limit under this Act which would affect a new action to enforce that claim. …(4) Rules of court may provide for allowing a new claim to which subsection (3) above applies to be made as there mentioned, but only if the conditions specified in subsection (5) below are satisfied, and subject to any further restrictions the rules may impose.(5) The conditions referred to in subsection (4) above are the following— (a) in the case of a claim involving a new cause of action, if the new cause of action arises out of the same facts or substantially the same facts as are already in issue on any claim previously made in the original action; and (b) in the case of a claim involving a new party, if the addition or substitution of the new party is necessary for the determination of the original action.(6) The addition or substitution of a new party shall not be regarded for the purposes of subsection (5)(b) above as necessary for the determination of the original action unless either— (a) the new party is substituted for a party whose name was given in any claim made in the original action in mistake for the new party’s name; or (b) any claim already made in the original action cannot be maintained by or against an existing party unless the new party is joined or substituted as plaintiff or defendant in that action.(7) Subject to subsection (4) above, rules of court may provide for allowing a party to any action to claim relief in a new capacity in respect of a new cause of action notwithstanding that he had no title to make that claim at the date of the commencement of the action. This subsection shall not be taken as prejudicing the power of rules of court to provide for allowing a party to claim relief in a new capacity without adding or substituting a new cause of action.” … (a) in the case of a new claim made in or by way of third party proceedings, on the date on which those proceedings were commenced; and (b) in the case of any other new claim, on the same date as the original action. (a) the addition or substitution of a new cause of action; or (b) the addition or substitution of a new party; … (a) in the case of a claim involving a new cause of action, if the new cause of action arises out of the same facts or substantially the same facts as are already in issue on any claim previously made in the original action; and (b) in the case of a claim involving a new party, if the addition or substitution of the new party is necessary for the determination of the original action. (a) the new party is substituted for a party whose name was given in any claim made in the original action in mistake for the new party’s name; or (b) any claim already made in the original action cannot be maintained by or against an existing party unless the new party is joined or substituted as plaintiff or defendant in that action.[32]The structure of section 35 is such that only sections 35(1) and 35(3) lay down binding rules, and the remainder of the section provides for rules of court to be made permitting amendments, subject to conditions, by way of new causes of action and new parties. Rules of court Rules of Supreme Court

Rules of court

[33]Following the Limitation Act 1980, RSC Ord 15, r 6 was amended in 1981 and immediately prior to the CPR provided, so far as material, as follows: “(5) No person shall be added or substituted as a party after the expiry of any relevant period of limitation unless either – (a) the relevant period was current at the date when proceedings were commenced and it is necessary for the determination of the action that the new party should be added, or substituted … … In this paragraph “any relevant period of limitation” means a time limit under the Limitation Act 1980 … (6) … the addition or substitution of a new party shall be treated as necessary for the purposes of paragraph (5)(a) if, and only if, the Court is satisfied that – (a) the new party is a necessary party to the action in that property is vested in him at law or in equity and the plaintiff’s claim in respect of an equitable interest in that property is liable to be defeated unless the new party is joined, or … (d) the new party is a company in which the plaintiff is a shareholder and on whose behalf the plaintiff is suing to enforce a right vested in the company… …” (a) the relevant period was current at the date when proceedings were commenced and it is necessary for the determination of the action that the new party should be added, or substituted … … (a) the new party is a necessary party to the action in that property is vested in him at law or in equity and the plaintiff’s claim in respect of an equitable interest in that property is liable to be defeated unless the new party is joined, or … (d) the new party is a company in which the plaintiff is a shareholder and on whose behalf the plaintiff is suing to enforce a right vested in the company…[34]RSC Ord 20, r 5 was also amended in 1981, but the only relevant change was to permit amendment to a party’s capacity not only to a capacity which the party had at the date of the commencement of the proceedings, but also to a change to a capacity which the party had since acquired. This gave effect to a recommendation of the Law Reform Committee, enacted as section 35(7), to deal with the anomaly that, where probate was granted to a person as executor, leave to amend to make a claim on behalf of the estate could be given because the title related back to the death, but where the plaintiff was subsequently granted letters of administration in such cases, the title related back to the date of the grant, which would have been after the issue of the writ. This had the effect of removing the grave injustice caused by such decisions as Ingall v Moran [1944] KB 160 (CA); Hilton v Sutton Steam Laundry [1946] KB 65 (CA); Burns v Campbell [1952] 1 KB 15; Finnegan v Cementation Co Ltd [1953] 1 QB 688 (CA).

Civil Procedure Rules

[35]The Civil Procedure Rules were introduced in 2000 to replace the Rules of the Supreme Court. By CPR 17.4, as amended by rule 7 of the Civil Procedure (Amendment) Rules 2001 (SI 2001/256):
“(2) The court may allow an amendment whose effect will be to add or substitute a new claim, but only if the new claim arises out of the same facts or substantially the same facts as a claim in respect of which the party applying for permission has already claimed a remedy in the proceedings. … (4) The court may allow an amendment to alter the capacity in which a party claims if the new capacity is one which that party had when the proceedings started or has since acquired (Rule 19.5 specifies the circumstances in which the court may allow a new party to be added or substituted after the end of a relevant limitation period.)”
[36]CPR 19.5, as amended by rule 8 of the Civil Procedure (Amendment) Rules 2001, provides so far as far as material as follows: “(1) This rule applies to a change of parties after the end of a period of limitation under –(a) the Limitation Act 1980;(b) the Foreign Limitation Periods Act 1984; or(c) any other enactment which allows such a change, or under which such a change is allowed. (2) The court may add or substitute a party only if – (a) the relevant limitation period was current when the proceedings were started; and (b) the addition or substitution is necessary. (3) The addition or substitution of a party is necessary only if the court is satisfied that – (a) the new party is to be substituted for a party who was named in the claim form in mistake for the new party; (b) the claim cannot properly be carried on by or against the original party unless the new party is added or substituted as claimant or defendant; or (c) the original party has died or had a bankruptcy order made against him and his interest or liability has passed to the new party. … (Rule 17.4 deals with other changes after the end of a relevant limitation period).” (a) the Limitation Act 1980; (b) the Foreign Limitation Periods Act 1984; or (c) any other enactment which allows such a change, or under which such a change is allowed. (a) the relevant limitation period was current when the proceedings were started; and (b) the addition or substitution is necessary. (a) the new party is to be substituted for a party who was named in the claim form in mistake for the new party; (b) the claim cannot properly be carried on by or against the original party unless the new party is added or substituted as claimant or defendant; or (c) the original party has died or had a bankruptcy order made against him and his interest or liability has passed to the new party.[37]In 2001 the Law Commission, Limitation of Actions (Law Com 270) recommended that the addition of new claims made between parties to existing proceedings after the expiry of the limitation period relevant to the new claim should be permitted where(1) the new claim arises out of the conduct, transaction or events on which a claim in the existing proceedings is based; and(2) the existing proceedings are commenced within the relevant limitation period: para 5.11 and draft Bill, clause 25(2). The Law Commission recommended that there should be no reform in relation to the addition of new claims to existing proceedings where the new claim involved the addition or substitution of new parties: para 5.19 and draft Bill, clause 25(3). The draft Bill contained among the conditions for amendment to add or substitute new parties, that “(c) the addition or substitution is necessary for the determination of a … civil claim previously made in the proceedings (‘the existing claim’), and (d) the existing claim was not made after the end of any applicable limitation period …” In November 2009 the Government announced that it would not be introducing legislation to implement the Law Commission’s proposals.

Effect of the Limitation Act 1980 and the CPR

[38]The effect of the provisions, so far as relevant on this appeal, in the Limitation Act 1980 and the CPR can be summarised in this way:(1) A new claim means a claim involving either (a) the addition or substitution of a new cause of action; or (b) the addition or substitution of a new party: section 35(2).(2) Any new claim made in the course of an action is deemed to have been commenced on the same date as the original action: section 35(1).(3) No such new claim may be made after the expiry of any applicable limitation period, except as provided by rules of court: section 35(3).(4) Rules of court may provide for allowing a new claim, but only (a) in the case of a claim involving a new cause of action, if the new cause of action arises out of the same facts or substantially the same facts as are already in issue on any claim previously made in the original action; and (b) in the case of a claim involving a new party, if the addition or substitution of the new party is necessary for the determination of the original action (i.e. any claim made in the original action cannot be maintained by an existing party unless the new party is joined as claimant or defendant): section 35(4), (5), (6). The relevant rules of court are in CPR 17.4 and 19.5.(5) CPR 17.4(2) has the effect that a new claim may be added by amendment but only if the new claim arises out of the same facts or substantially the same facts as the original claim.(6) CPR 19.5(2), (3) have the effect (among others) that a new party may be added only if the limitation period was current when the proceedings were started, and the addition of that party is necessary in the sense that the claim cannot properly be carried on by the original party unless the new party is added.(7) Rules of court may allow a party to claim relief in a new capacity: section 35(7). The relevant rule is CPR 17.4(4), by which the court may allow an amendment to alter the capacity in which a party claims if the new capacity is one which that party had when the proceedings started, or has since acquired.[39]The original claim for the purposes of these proceedings was the personal claim by Mark Roberts. The relevant limitation period for his claim was current when he started his proceedings in 2002. Mr Sainter’s claim as administrator is statute-barred. Mr Sainter could not apply to be joined as a claimant because his joinder is not “necessary” for the purposes of section 35(6)(b) as put into effect by CPR 19.5(3)(b). That is because his joinder is not necessary for the purposes of the original action, namely Mark Roberts’ personal claim.[40]But an amendment to treat Mark Roberts’ claim as a representative claim rather than a personal claim would be an amendment to alter the capacity in which he claims: Haq v Singh [2001] EWCA Civ 957, [2001] 1 WLR 1594, at [19], per Arden LJ. CPR 17.4(4) permits such a change to be made if the new capacity is one which he had when the proceedings started or has since acquired. Mark Roberts has throughout had the capacity of beneficiary. It is not necessary to decide whether the representative capacity is one which he has had in theory at all times, since there is no doubt that the court has power to allow the amendment to alter the capacity in which he sues.[41]The representative claim is a claim involving a new cause of action, since the capacity in which Mark Roberts makes the claim is an essential part of the claim: Oates v Consolidated Capital Services Pty Ltd [2009] NSWCA 183, at [105]. The court has power to allow the amendment because the new representative claim arises out of the same facts or substantially the same facts as the existing claim: CPR 17.4(2). Consequently it is not necessary to burden this discussion with a sterile analysis of the learning on what constitutes a cause of action. It is sufficient to quote what Robert Walker LJ said in Smith v Henniker-Major & Co (A firm) [2002] EWCA Civ 762, [2003] Ch 182 (CA) at [96]. He referred to the classic definitions by Brett J in Cooke v Gill (1873) LR 8 CP 107, 116 as “every fact which is material to be proved to entitle the plaintiff to succeed”, and by Diplock LJ in Letang v Cooper [1965] 1 QB 232, 242-243 as “simply a factual situation the existence of which entitles one person to obtain from the court a remedy against another person,” and went on:
“…in identifying a new cause of action the bare minimum of essential facts abstracted from the original pleading is to be compared with the minimum as it would be constituted under the amended pleading. But in applying section 35(5)(a) the court is concerned on a much less abstract level with all the evidence likely to be adduced at trial: see Goode v Martin [2002] 1 WLR 1828, 1838, approving Hobhouse LJ’s observation in Lloyds Bank plc v Rogers The Times, 24 March 1997; Court of Appeal (Civil Division) Transcript No 1904 of 1996: ‘The policy of the section is that, if factual issues are in any event going to be litigated between the parties, the parties should be able to rely upon any cause of action which substantially arises from those facts’.”

Joinder of the administrator

[42]The next question is whether Mark Roberts needs to join the administrator as a defendant. If he does need to join the administrator, there would be a new claim for the purposes of the Limitation Act 1980 because “a new claim means … any claim involving … the addition … of a new party” (section 35(2)(b)). Rules of court may allow such an addition after the limitation period for the original claim has expired, and the condition in CPR 19.5(2)(a) is fulfilled, namely that the relevant limitation period was current when the proceedings were started (i.e. by Mark Roberts in his personal capacity).[43]But if the administrator has to be added at the same time as Mark Roberts changes the capacity in which he sues, Mark Roberts must satisfy the requirements of CPR 19.5(2)(b) and CPR 19.5(3)(b) (giving effect to section 35(5)(b) and 6(b)), namely that the addition of the administrator is necessary in the sense that “the claim cannot properly be carried on by … the original party unless the new party is added ...” But if it were necessary to join the administrator in order for the representative action to be carried on, Mark Roberts would not be able to satisfy those requirements because he would not be able to show that the original claim could not properly be carried on by Mark Roberts in his personal capacity against the solicitors unless the administrator were added as a party. That is because there is no possible basis for any suggestion that the administrator would be a necessary or proper party to the personal claim.[44]Consequently, the only way in which the action could proceed would be(a) if the joinder of the administrator were not necessary at all; or(b) if it were not necessary at the time Mark Roberts changes the capacity in which he sues, but could be done at a later stage. The latter point arises because it is suggested on behalf of Mark Roberts that the combined effect of section 35(1)(b), section 35(6)(b) and CPR 19.5(3)(b) is that (a) the change in Mark Roberts’ capacity is deemed to take effect as at the date of the original proceedings; (b) the joinder can be effected after the change in Mark Roberts’ capacity, and would then be “necessary” for the continuance of what would then be regarded as the original claim, namely the claim in a representative capacity.

The trustee as a necessary party and the “special circumstances” rule

[45]To avoid repetition, it is convenient to treat together two aspects of the leading authorities. One is the requirement for special circumstances as a condition of beneficiaries being entitled to sue on behalf of the estate. The second is the light they throw on whether the administrator is a necessary party to a claim by a beneficiary to recover property from a third party.[46]The cases go back to the eighteenth century, and many of them were reviewed in Hayim v Citibank NA [1987] AC 730 (PC). The special circumstances which were identified in the earliest authorities as justifying a beneficiary’s action were fraud on the part of the trustee, or collusion between the trustee and the third party, or the insolvency of the trustee, but it has always been clear that these are merely examples of special circumstances, and that the underlying question is whether the circumstances are sufficiently special to make it just for the beneficiary to have the remedy: In re Field, decd [1971] 1 WLR 555, 560-561, per Goff J; cf. Barker v Birch (1847) 1 De G & Sm 376, 63 ER 1112; Daniell’s Chancery Practice, 7th ed 1901, p 176.[47]In all of the early cases the trustees were co-defendants with the third party debtor. In Bickley v Dorrington (1737) West T Hard 169, 25 ER 877 the bill was brought by creditors, and by one of the residuary legatees of the testator, against his executors, the other residuary legatee, and the former partner of the testator to recover from the former partner money owing to the estate. Lord Hardwicke LC said that the bill was totally improper as against the debtor, and inconsistent with the principles of law and the rules of the court:
“No action or suit can be brought against a debtor to the estate but by the executor or personal representative of the testator. The whole management of the estate belongs to him. The right of it is vested in him, and cannot be taken from him by creditors or legatees. If he release a demand and is solvent, it is a devastavit in him, and he is personally answerable for the sum released. In cases of collusion or insolvency it may be proper to come here for satisfaction against the debtor; but there must always be some special case …”
(pp 171, 879)[48]Sixty-five years later in Alsager v Rowley (1802) 6 Ves Jun 748, 31 ER 1289, Lord Eldon LC said (at 749-750):
“The established rule of the Court is certainly … that in ordinary cases a debtor to the estate cannot be made a party to a bill against the executor: but there must be, as the cases express it, collusion or insolvency. That very principle admits, that, if there is solvency, the executor must pay: if there is collusion, both are liable. … Lord Hardwicke there in the judgment [Beckley v Dorrington] does not state any thing as to negligence. That is in the argument by the Counsel; and in Newland v Champion (1 Ves sen 105) delay in the representative is also stated as one of the special cases, as well as collusion: but no notice is taken of the former in the judgment. If the general principle will not allow you to bring a bill against both the executor and a debtor in the given case, the same principle will apply to the case, where you bring a bill against the executor and a creditor improperly paid by the executor: that is, that, if there is no collusion, or special case, if the executor is not insolvent, he stands the middle man, responsible to the residuary legatee for the property, misapplied by paying a man as a creditor, who was not a creditor, as in the other case for the property outstanding in a debtor.”
[49]The following three cases involve claims by legatees who were concerned that executors were not properly pursuing the testator’s partners for an account of partnership dealings. In each case the executors were defendants. In Bowsher v Watkins (1830) 1 Russ & Myl 277, 39 ER 107, residuary legatees brought suit against executors and a surviving partner of the testator for an account. In answer to the argument that there were no special circumstances justifying the action by legatees, Sir John Leach MR held that collusion between the executor and the partner was not an essential condition. In Davies v Davies (1837) 2 Keen 534, 48 ER 733, a bill was filed by residuary legatees against the executor and the surviving partner of the testator for an account of partnership transactions. Lord Langdale MR held that, in the absence of a charge of fraud or collusion, there were no special circumstances justifying the legatees’ claim. In Travis v Milne (1851) 9 Hare 141, 68 ER 449, a similar case, Turner V-C held that a suit by parties beneficially interested in the estate of a deceased partner could not be maintained against both his executors and surviving partners, in the absence of special circumstances; but collusion was not the only ground for such a suit; and the suit might be maintained where the relation between the executors and surviving partners was such as to present a substantial impediment to the prosecution by the executors of the rights of the parties interested in the estate as against such partners.[50]Yeatman v Yeatman (1877) 7 Ch D 210 was another case involving a partnership. The executors were named defendants. It was an action by a residuary legatee against her mother-in-law’s executors and another member of the family, who it was alleged, had failed to account for partnership monies to the mother-in-law. A mere refusal by a personal representative to sue for recovery of a debt owed to the estate would not by itself suffice but there must be additional circumstances, such as a case where the trustee refused to sue and the court was satisfied that it would have given liberty to the trustee to bring proceedings even though there was no certainty that the proceedings would be successful, then “it would be a proper case to allow a party to sue in his own name” (at 216, per Hall V-C).[51]In In re Field, decd [1971] 1 WLR 555 Mr Field’s widow obtained letters of administration. His former wife (who was not a beneficiary) obtained an order for maintenance to be paid out of the estate, and independent administrators were appointed. The former wife sued her former husband’s employers and the administrators in her own name for the recovery of capital sums on insurance policies which she claimed the employers had wrongly paid to the widow. As Goff J said (at 558):
“She does not and, indeed, cannot ask for payment to herself, but she asks for payment to the administrators who are added as defendants for the purpose of regularising the proceedings and, by her writ and statement of claim, she expressly disclaims any relief as against them.”
It was held that there were special circumstances entitling the former wife to make the claim, particularly because there were no other beneficiaries and the alleged asset had been paid to the widow on the footing that it was not part of the estate. Consequently “justice requires that the plaintiff, who is the only other person interested, should be allowed to have this question properly tried by the court” (at 561). See also Bradstock Trustee Services Ltd v Nabarro Nathanson [1995] 1 WLR 1405, 1412-1413, on the relevance of legal aid for beneficiaries.[52]In Hayim v Citibank NA [1987] AC 730 the plaintiffs were the testator’s sons, who were beneficiaries under his American will. He also executed a Hong Kong will under which the residue of his property outside the United States was to be held on trust for sale on the trusts of the American will. The terms of the trust enabled the trustee of the American will to give directions to the trustee of the Hong Kong will in respect of the retention of a house in Hong Kong in the interests of the elderly residents of the house. The plaintiffs began proceedings in Hong Kong against the first defendant, the trustee of the American will, and the second defendant, the trustee of the Hong Kong will, for an order that the house be sold and for damages to be awarded against the second defendant for breach of the trusts of the Hong Kong will by the delay of the second defendant in selling the house. No relief was sought against the first defendant. It was held that there were no special circumstances entitling the plaintiffs to bring proceedings directly against the second defendant, but that in any event no breach of the trusts of the Hong Kong will had been committed by the second defendant in implementing the lawful instructions of the first defendant.[53]Lord Templeman, giving the advice of the Privy Council, said (at 747):
“…when a trustee commits a breach of trust or is involved in a conflict of interest and duty or in other exceptional circumstances a beneficiary may be allowed to sue a third party in the place of the trustee. But a beneficiary allowed to take proceedings cannot be in a better position than a trustee carrying out his duties in a proper manner. …” and (at 748) (after citing, among other cases, Travis v Milne; Yeatman v Yeatman; and In re Field, decd) “These authorities demonstrate that a beneficiary has no cause of action against a third party save in special circumstances which embrace a failure, excusable or inexcusable, by the trustees in the performance of the duty owed by the trustees to the beneficiary to protect the trust estate or to protect the interests of the beneficiary in the trust estate.”
[54]The law in Scotland is similar: Wilson and Duncan, Trusts, Trustees and Executors, 2nd ed 1995, para 10-13, who give an example of the possibility of an action by a beneficiary for recovery of a sum paid by trustees to a third party in error, “provided that the action is brought against the trustees as well as the third party.” The authority cited for this proposition is Armour v Glasgow Royal Infirmary 1909 SC 916, where the Lord Ordinary, Lord Skerrington said (at 920) that in the ordinary case the action could be brought only at the instance of the trustees. But the testamentary trustees had been called as defenders, and concurred with the Infirmary in maintaining that the money was properly paid. Consequently in those circumstances “the pursuers have a good and sufficient title to maintain this action to the effect of demanding that the money shall be repaid to the trustees.” It is only with the greatest hesitation that I would differ from Lord Hope on the content of Scots law, but in my judgment neither of the cases which he cites lends support to the view that there is a qualification to the rule that the trustee must be joined. In Morrison v Morrison’s Executors, 1912 SC 892, 893, Lord Skerrington went on to say after the passage quoted by Lord Hope:
“A decree in such an action would be res judicata, provided always that the whole trustees and beneficiaries had been called as defenders.”
In the second case mentioned by Lord Hope, Rae v Meek (1889) 14 App Cas 558, the beneficiaries under a settlement made pursuant to a marriage contract sued a trustee for having lost trust money which had been lent on the security of unfinished houses in a building speculation. The trustee was held liable to restore the trust fund. The beneficiaries also sued the trustees’ solicitor, who had advised the trustee that there was no objection to the investment. The passage quoted by Lord Hope in Lord Herschell’s speech (at 569), which deals with the liability of the solicitor, is simply re-stating the rule that in the exceptional case of a failure by trustees to act, “the beneficiaries might compel them to do so, or even enforce the right themselves.” He went on to say that no such question (that is, of a failure by trustees to act) was raised by the averments in relation to the claim in that case by the beneficiaries against the solicitor, who (in any event) was not liable because he had not been retained by the trustees to advise on the sufficiency of the security. But the trustees were parties to the action, and the decision is not authority for any suggestion that the beneficiary can assert the claim without joining the trustee.[55]So also the law in the United States has the same result, although it is put somewhat differently. See Restatement (Second), Trusts, 1959, section 282(2); and Scott and Ascher, Trusts, 5th ed 1995, chapter 28, section 28.1, where it is put in this way: “It is the trustee rather than the beneficiary who is entitled to maintain actions against third parties who commit torts with respect to the trust property or fail to pay debts held in trust. If the trustee improperly fails to bring such an action, the beneficiaries can compel the trustee by a suit in equity to do so, and, in order to settle the whole matter in a single suit, they can join the third party as a co-defendant”.[56]That joinder of the trustees in a beneficiaries’ derivative action is required is supported by the analogy of shareholders’ derivative actions, where the wrongdoers are themselves in control of the company, and the aggrieved minority may bring a minority shareholders’ action. In Nurcombe v Nurcombe [1985] 1 WLR 370 (CA) Browne-Wilkinson LJ said (at 378) that a minority shareholder’s action, where the courts of equity permitted a person interested to bring an action to enforce the company’s claim, was analogous to that in which equity permitted a beneficiary under a trust to sue as plaintiff to enforce a legal right vested in trustees, which right the trustees will not themselves enforce, the trustees being joined as defendants.[57]A derivative action is brought in representative form, and the company is joined as a defendant in order for it to be bound by any judgment and to receive the fruits (if any) of the judgment, and because the action has not been authorised by its board or general meeting: Spokes v Grosvenor and West End Railway Terminus Hotel Co Ltd [1897] 2 QB 124, which is the leading authority on the joinder of the company in derivative actions. A L Smith LJ said (at 126): “That in the circumstances of this case the company are necessary parties to the suit I do not doubt, for without the company being made a party to the action it could not proceed”.[58]Chitty LJ said (at 128-129): “To such an action as this the company are necessary defendants. The reason is obvious: the wrong alleged is done to the company, and the company must be party to the suit in order to be bound by the result of the action and to receive the money recovered in the action. If the company were not bound they could bring a fresh action for the same cause if the action failed, and there were subsequently a change in the board of directors and in the voting power. Obviously in such action as this is, no specific relief is asked against the company; and obviously, too, what is recovered cannot be paid to the plaintiff representing the minority, but must go into the coffers of the company. It was argued for the appellants that the company were made a party for the purpose of discovery only, and authorities were cited to shew that when no relief is asked against a party he cannot or ought not to be compelled to make discovery. But this argument and these authorities have no bearing on the present case, where, as already shewn, the action cannot proceed in the absence of the defendant company, and the defendant company are interested in and will be bound by the results”.[59]Since Part 11 of the Companies Act 2006 came into force in 2007 shareholders’ derivative claims have been put on a statutory basis. CPR 19.9 is headed “Derivative Claims – how started”. It does not apply to derivative claims of the type in issue on this appeal, but it illustrates the general principle that in derivative actions the entity on whose behalf the claim is brought is a necessary party to the derivative claim. CPR 19.9 applies in terms only to derivative claims by members of companies, other bodies corporate, and trade unions, and provides in CPR 19.9(3) that:
“The company, body corporate or trade union for the benefit of which a remedy is sought must be made a defendant to the claim.”
[60]The expression “derivative action” in the context of shareholders’ actions has been used in the United States since the nineteenth century and was first used in that context in England by Lord Denning MR in Wallersteiner v Moir (No 2) [1975] QB 373, 390 et seq. In the United States it is equally established that the corporation is a necessary party in any shareholder derivative action, although (as in a derivative claims by beneficiary) it is sometimes analysed as two claims, one against the company for failure to take action and the other being the claim by the company against the wrongdoer: Nurcombe v Nurcombe [1985] 1 WLR 370, 378; Konamaneni v Rolls Royce Industrial Power (India) Ltd [2002] 1 WLR 1269, at [50]. In 1873, in Davenport v Dows, 85 U.S 626 (1873) Justice Davis, delivering the opinion of the Supreme Court, said, at p 627: “These rights the individual shareholder is allowed to assert in behalf of himself and associates, because the directors of the corporation decline to take the proper steps to assert them. Manifestly the proceedings for this purpose should be so conducted that any decree which shall be made on the merits shall conclude the corporation. This can only be done by making the corporation a party defendant. The relief asked is on behalf of the corporation, not the individual shareholder, and if it be granted, the complainant derives only an incidental benefit from it. It would be wrong, in case the shareholder were unsuccessful, to allow the corporation to renew the litigation in another suit involving precisely the same subject matter. To avoid such a result, a court of equity will not take cognizance of a bill brought to settle a question in which the corporation is the essential party in interest unless it is made a party to the litigation”.[61]Almost a hundred years later, in Ross v Bernhard, 396 US 531 (1970) at 538, the Supreme Court said:
“The corporation is a necessary party to the action; without it the case cannot proceed. Although named a defendant, it is the real party in interest, the stockholder being at best the nominal plaintiff. The proceeds of the action belong to the corporation and it is bound by the result of the suit”
. Another reason for joinder has been said to be that “the decree must protect the directors against any further suit by the corporation, and this will not be true unless it be a party to the suit:” Philipbar v Derby, 85 F 2d 27 (2d Cir 1936) at 30.[62]Consequently it has been the consistent practice (noted in Annual Practice 1887-8, p 223; Harmer v Armstrong [1934] Ch 65, 93, per Romer LJ) for almost 300 years that, where a beneficiary brings an action in his own name to recover trust property, the trustees should be joined as defendants. Daniell’s Chancery Practice, 7th ed 1901, p 176 states:
“…such an action cannot, however, be maintained without the personal representative being a party”
. To put it differently, it would be “procedurally improper to continue without the addition … which is proposed”: McGee, Limitation Periods, 5th ed 2006, para 23.025. The purpose of joinder has been said to ensure that they are bound by any judgment and to avoid the risk of multiplicity of actions: Lewin, Trusts, 18th ed 2008, para 43-05. But joinder also has a substantive basis, since the beneficiary has no personal right to sue, and is suing on behalf of the estate, or more accurately, the trustee.[63]The conclusion that in a beneficiary’s derivative action the trustee must be a party is not undermined by those cases in which it has been held, or assumed, that an action by an equitable assignee of property (such as a debt, or intellectual property) can proceed, or is properly constituted, without the joinder of the assignor at the outset of proceedings.[64]The starting point is that if an equitable assignee sues a third party, the assignor must be joined as a defendant: E M Bowden’s Patents Syndicate Ltd v Herbert Smith & Co [1904] 2 Ch 86, 91 (Warrington J); William Brandt’s Sons & Co v Dunlop Rubber Co Ltd [1905] AC 454, 462 (Lord Macnaghten); Performing Right Society, Ltd v London Theatre of Varieties Ltd [1924] AC 1, 13-14 (Viscount Cave LC), 19-20 (Viscount Finlay), 29 (Lord Sumner); Vandepitte v Preferred Accident Insurance Corporation of New York [1933] AC 70, 79 (Lord Wright); Harmer v Armstrong [1934] Ch. 65, 82 (Lord Hanworth MR).[65]But it is not an invariable rule: Performing Right Society Ltd v London Theatre of Varieties Ltd [1924] AC 1, 14 (“there may be special cases where it will not be enforced” per Viscount Cave LC). In that decision it was held that an equitable assignee may obtain interlocutory relief but was not entitled to obtain a final injunction without joining the legal owners. Viscount Cave LC said (at 14):
“That an equitable owner may commence proceedings alone, and may obtain interim protection in the form of an interlocutory injunction, is not in doubt; but it was always the rule of the Court of Chancery, and is, I think, the rule of the Supreme Court, that, in general, when a plaintiff has only an equitable right in the thing demanded, the person having the legal right to demand it must in due course be made a party to the action … Further, under Order XVI., r. 11, no action can now be defeated by reason of the misjoinder or non-joinder of any party; but this does not mean that judgment can be obtained in the absence of a necessary party to the action, and the rule is satisfied by allowing parties to be added at any stage of a case. Subject to these observations, I think that the general rule is still operative …”
[66]William Brandt’s Sons & Co v Dunlop Rubber Co Ltd [1905] AC 454 was a case in which an assignee was allowed to proceed to judgment without joining the assignor. That was because the whole focus of the litigation was on the question whether instructions given by the bank’s customer to purchasers of rubber to pay its bank direct amounted to an equitable assignment of debts, so that the bank could sue for their recovery. The bank sued the purchasers directly without joining its customer, the assignor. The fact that the assignor was not a party seems to have been overlooked until the House of Lords held that there had been an equitable assignment. Lord Macnaghten said (at 462):
“Strictly speaking, [the sellers], or their trustee in bankruptcy, should have been brought before the Court. But no action is now dismissed for want of parties, and the trustee in bankruptcy had really no interest in the matter. At your Lordships’ bar the Dunlops disclaimed any wish to have him present, and in both Courts below they claimed to retain for their own use any balance that might remain after satisfying Brandts.”
Lord James said (at 464):
“The defect in the parties to the suit can be remedied.”
[67]In more modern times it has been held that, although the practice was to join the assignor, the requirement is a procedural one, the absence of which can be cured. The assignor must be joined before a final judgment can be obtained by the assignee, but the action is validly constituted without joinder, so that if the assignee sues without joining the assignor, the action is in time for the purposes of limitation: Central Insurance Co Ltd v Seacalf Shipping Corpn (The Aiolos) [1983] 2 Lloyd’s Rep 25, 34, per Oliver LJ; Weddell v JA Pearce & Major [1988] Ch 26, 40, per Scott J; and cf Robinson v Unicos Property Corpn Ltd [1962] 1 WLR 520, 525-526, per Holroyd Pearce and Harman LJJ; Three Rivers District Council v Governor and Company of the Bank of England [1996] QB 292, 309, 313, per Peter Gibson LJ. For criticism see Tolhurst, Equitable Assignment of Legal Rights: A Resolution to a Conundrum (2002) 118 LQR 98, at 111-116.[68]What distinguishes these cases from the present one is that in the case of an equitable assignment the assignee is the true owner and the assignor is a bare trustee. I agree with Lord Walker that there is no real analogy between an equitable assignee and a beneficiary interested in an unadministered estate.[69]I would not rule out the possibility that there may be circumstances in which justice would require that joinder of the administrator be dispensed with. But the mere fact that there were “special circumstances” justifying an action by the beneficiary, or the fact that non-joinder would defeat a limitation defence, would not be sufficient.[70]It follows that the limitation problem cannot be overcome by amendments in separate stages so as to procure the result that the addition of the administrator would be necessary “for the determination of the original action” for the purposes of section 35(5)(b). The argument for amendments in separate stages is this. First, there would be an amendment to change the capacity in which Mark Roberts sues from his personal capacity to a representative capacity under CPR 17.4(4). Second, this would have the effect that the new claim in the representative capacity is deemed to commence on the same date as the original action: section 35(1)(b); section 35(2)(a). Third, the addition of the administrator would be necessary because the claim in the original action (i.e. the back-dated representative claim) cannot be maintained against the solicitors unless the administrator is joined: section 36(5)(b); section 36(6)(b).[71]This procedural device cannot overcome the limitation problem, since it is plain that, other than in the most exceptional circumstances such as existed in William Brandt’s Sons & Co v Dunlop Rubber Co Ltd, even in the case of an equitable assignment the assignee cannot proceed to judgment without joining the assignor. It would be an abuse of process for the amendments to be made in separate stages. As Arden LJ said (at [36]), it would be contrary to principle for the court to grant permission to amend the claim merely to reflect a change of capacity which would not enable Mark Roberts to proceed to judgment. But, in any event, in a representative action, the administrator must be joined at the outset.[72]The result is that the Court of Appeal was right to conclude that this was not a case where permission to amend to plead the derivative claim should be given.

Special circumstances

[73]Consequently, the question whether the judge was right to decide that in any event there were no special circumstances justifying the derivative claim (on which the Court of Appeal was divided) does not arise. It can therefore be dealt with shortly.

The judge set out the relevant circumstances as follows:

[74](1) When Mr Sainter was appointed as administrator in 2000, he was still in time to bring any claim which the estate was able in law to bring against the solicitors;(2) at the present time, a claim by the estate against the solicitors is statute-barred;(3) Mr Sainter was appointed by the court as administrator on the application of Mark Roberts;(4) Mark Roberts did not apply for himself to be appointed as an administrator;(5) Mark Roberts did not procure by way of an assignment or by way of an assent, the vesting of the estate’s cause of action against the solicitors into himself before the limitation period ran out;(6) if Mark Roberts procured the vesting of the estate’s cause of action in himself at the present time then he would not be able to assert that cause of action, by reason of limitation;(7) there was no reason to think that Mr Sainter would not have been prepared to vest the estate’s cause of action in Mark Roberts;(8) there was no basis for any allegation of any breach of trust against Mr Sainter;(9) there was no conflict of duty or interest involving Mr Sainter;(10) Mr Sainter’s decision not to sue the solicitors had not been said to be open to any criticism;(11) if John Roberts had remained the administrator then there might at that time have been special circumstances arising out of the allegations being made as to the involvement of John Roberts in the matters complained of;(12) any special circumstances which existed during the time that John Roberts was administrator ceased to exist when Mr Sainter became administrator in October 2000;(13) Mark Roberts was not the sole beneficiary;(14) the Court had no specific evidence as to the attitude of Mrs Roberts’ sister or the Inland Revenue;(15) the proceedings against the solicitors were far from straightforward, although the judge did not base his decision on any assessment of the precise prospects of success in those proceedings;(16) in the absence of argument on the point, he left out of account the question whether Mr Sainter as administrator might be liable to pay the costs if a derivative action were permitted and proceeded and failed;(17) Mark Roberts had legal services funding to bring the present proceedings and it might very well be the case that he had or would obtain legal services funding to bring a derivative claim;(18) the court had power under CPR 17.4 to give Mark Roberts permission to amend the present proceedings to add a derivative claim (if special circumstances existed) and thereby defeat a limitation defence.[75]The judge took the view that, although the list of “special circumstances” was not closed and “special circumstances” had never been exhaustively defined, the circumstances as to legal services funding and limitation were of a different character from anything contemplated in the cases as to special circumstances. The circumstances in (17) and (18) were not special circumstances which would justify the court in permitting Mark Roberts to bring a derivative claim against the defendants. Arden LJ (with whom Patten J agreed) seems to have taken the view that the judge was wrong (among other reasons) because he had not given sufficient weight to the fact that the derivative claim would enable an asset to be realised, which otherwise could not be realised, and because Mark Roberts had legal aid the estate would not have to fund his costs.[76]If the point had arisen for decision, I (in agreement with Pill LJ) would have taken the view that this was a case where the judge had a wide latitude in evaluating what were special circumstances, that he took all the relevant circumstances into account, and that he conducted the enquiry in a way with which an appellate court should not have interfered.[77]I would therefore dismiss the appeal on the ground that the Court of Appeal was right to hold that the amendment to pursue a derivative claim was not permitted by the CPR after the expiry of the limitation period.[78]I agree with all my colleagues that, for the reasons that Pill LJ in the Court of Appeal and Lord Collins in this court have given, the judge at first instance was fully entitled to hold that the appellant has failed to show that there were special circumstances justifying the derivative claim which he seeks to bring at this late stage. Contrary to the views of the majority in the Court of Appeal, I would hold that it would not have been permissible for that court to interfere with his decision. I would dismiss the appeal on this ground because, like Lord Clarke, I would prefer not to reach a final conclusion on the question whether, if special circumstances had been made out, the court would have had power to give the appellant permission to amend to introduce the derivative claim.[79]I am not convinced that the rule that the administrator must be joined is quite as absolute as Lord Collins indicates in his judgment. He has referred in para 54 to the law of Scotland as explained in Wilson and Duncan, Trusts, Trustees and Executors, 2nd ed (1995), para 10-13. It is stated in that paragraph that if the trustees refuse to sue to recover a debt due to the trust estate, they can be forced to lend their names to the beneficiaries to enable them to raise the action: Blair v Stirling (1894) 1 SLT 599; Brown’s Trustees v Brown (1888) 15 R 581. There is no doubt that this is the ordinary rule. As Mackenzie Stuart, The Law of Trusts (1932), p 210, explains, if the beneficiaries insist on action being taken, the trustees must lend their name and authority to the beneficiaries in order that they may have a formal title to sue. This explanation supports the view that Lord Collins has expressed in para 62 of his judgment that joinder has a substantive basis, as the beneficiary has no personal right to sue for the recovery of trust property. This is not, as Scots law would see it too, simply a matter of procedure.[80]How Scots law would see the procedural issue is, as Lord Rodger says, not free from difficulty. But there are some indications as to how it deals with the question which is of real interest in this case, which is whether proceedings can be raised without joining the trustees at the outset.[81]In Morrison v Morrison’s Executrix 1912 SC 892, 893 the Lord Ordinary, Lord Skerrington, said that it was certainly logical that no one should be allowed to sue an action unless that right sought to be enforced had been duly transferred to him, and that any injustice that this rule of law might operate was obviated by the further rule that the person who has the beneficial interest may compel the person who has the formal title to lend his name on receiving security against expenses. But he went on to indicate that this was not an absolute rule:
“I am of opinion that where justice absolutely requires it, the action may, in spite of the legal technicalities, be allowed to proceed at the instance of the party who has the beneficial interest.”
He referred, in support of that proposition, to a passage in Lord Herschells’ speech in Rae v Meek (1889) 14 App Cas 558, 569, where he said:
“The alleged duty, if it existed at all, was to the trustees, and not to the beneficiaries. If there has been a breach of it, the trustees and not the beneficiaries are the parties to sue. There may be cases where, if trustees failed to call to account those who were under liability in respect of acts injurious to the trust estate, the beneficiaries might compel them to do so, or even enforce the right themselves.”
[82]The last six words in this quotation from Lord Herschell’s speech may seem a rather slender foundation on which to qualify a rule based on a substantial point of principle. But I take them to indicate that it would be unwise to regard this rule as one which will always be enforced. Viscount Cave LC’s observation in Performing Right Society Ltd v London Theatre of Varieties [1924] AC 1, 14 that there may be special cases where the rule that the person with the legal right must be joined will not be enforced provides further support for this approach.[83]It is true, as Lord Collins points out, that in Morrison v Morrison’s Executrix Lord Skerrington went on to point out that a decree in such an action would not be res judicata unless all the trustees and beneficiaries were called as defenders. Teulon v Seaton (1884) 12 R 971 is an example of such a case where the title of the beneficiary to sue a debtor to a trust was sustained, but the precaution had been taken of calling the trustees as defenders to the action at the outset. But two points should be noted about this qualification. First, it is for the party against whom the action has been brought to take this point, by a plea of all parties not called. It has never been suggested that the court can compel the pursuer to do this as a condition of raising his action. Second, this step can be taken at any time before extract of the final decree: Maclaren, Court of Session Practice (1916) p 478-479; Maxwell, The Practice of the Court of Session (1980), p 281.[84]The procedure which Scots law uses to cure the absence of a personal right in the beneficiary is different from that which is under discussion in this case. But there is much common ground. The beneficiary has no personal right to sue. The requirement that the personal representative must be joined is more than just a matter of procedure. Yet the rule is not an absolute one. It may be departed from if this is necessary to avoid an injustice. An action which is raised on this basis is not to be regarded as bad from the outset, although the personal representative may have to be joined at a later stage. Like Lord Clarke, however, I think that the appellant would find it very hard to justify a departure from the rule in the circumstances of this case.[85]At the end of the hearing I was inclined to think that it might have been possible for the claimant to make the amendments in separate stages, as outlined by Lord Collins in para 70 of his judgment. But, having studied what he says, I am satisfied that this would really be an artificial device: it would be to permit the claimant to do in two steps something which the statute and CPR do not permit him to do directly in one step.[86]For the rest, I agree with Lord Walker and Lord Collins and would dismiss the appeal for the reasons which they give.[87]I am reluctant to get drawn into a discussion of a tangential point of Scots law which was not argued and is not free from difficulty. Unquestionably, the general rule is that the beneficiary of a trust cannot sue a debtor of the trust: the relevant right of action is vested in the trustees and it is for them to enforce that right by raising an action, if appropriate. Where the trustees decline to take proceedings but the beneficiary insists, he can require them to assign the right of action or to permit him to use their name, provided that he gives them an indemnity for any liability for expenses.[88]Lord Shand stated the position in absolute terms in Raes v Meek (1888) 15 R 1033, 1050-1051:
“If the trustees do not think fit to raise an action against the debtors for certain debts, having doubts it may be how far they may be certain of success, is it for a beneficiary or beneficiaries to do so in their own name? I think they have no such right. And I do not think this is a matter of mere form; it is, in my view, a matter of substance, because if the law were otherwise, then the debtors of trust-estates, including amongst them law-agents who may have been employed by the trustees, would be liable to actions at the instance of many different persons – of anyone having a beneficial interest in the trust-estate – requiring them to pay the amount of their debts to the trustees. I think such actions are not competent, and that the only persons who can maintain actions to recover debts due to an executry or trust-estate are the administrators of the estate, the trustees or the executors. A beneficiary could not discharge the liability for a claim due to the trustees and I do not see that a judgment in an action at the instance of a beneficiary could be res judicata in a question with the trustees. It appears to me that the law would get into extreme confusion if we were to sanction actions of this kind raised by a beneficiary against one with whom he had no contract. The beneficiary has his rights against the trustees, for the trustees are in direct relation with him because of their having undertaken a trust for his behoof. But if beneficiaries seek to enforce by action a claim of any kind against a debtor to the trust, it appears to me that they must either compel the trustees to raise the question directly in their own names, or get authority to use their names, or get an assignation to the claim, and thereupon sue as assignees.”
By contrast, Lord Young, in a characteristic, freewheeling judgment, argued, at pp 1058-1059, that, since everyone was in the action already, the beneficiaries should be able to proceed against the law agents.[89]Against the background of Lord Shand’s careful statement of the position, the passage in Lord Herschell’s speech on the appeal quoted by Lord Hope at para 5 is, as Lord Hope says, “a rather slender foundation on which to qualify a rule based on a substantial point of principle” Indeed, one might wonder whether Lord Herschell had in mind anything more than the use of the trustees’ name or the taking of an assignation, which are the well recognised ways in which beneficiaries can take proceedings, if they wish to do so. Assuming, however, that a beneficiary can take proceedings against a debtor of the trust, the question is: can he do so without joining the trustees as defenders?[90]In Morrison v Morrison’s Executrix 1912 SC 892, 893, the Lord Ordinary (Skerrington) did indeed take Lord Herschell’s comment in Raes as support for the view that, where justice absolutely required it, an action against a debtor to a trust might be allowed to proceed at the instance of a beneficiary. Lord Skerrington also cited Teulon v Seaton (1885) 12 R 971 in which he had acted as counsel for the pursuer, who was the administratrix of the estate of her mother, a residuary legatee under a settlement. The First Division was prepared to allow her action against the truster’s widower, who was alleged to have intromitted with the trust estate, to go ahead - but only if she found caution for the expenses. She failed to do so and the defenders were assoilzied: (1885) 22 SLR 786. According to the report, 12 R 971, 973, in the Inner House the first defender directed his principal argument to the issue of forum non conveniens. On the other hand, Lord President Inglis appears to have focused on the fact that the pursuer was a married woman with no estate independent of her husband. Which is presumably why counsel ended up by successfully moving that the pursuer should be ordained to find caution for expenses.[91]In Morrison Lord Skerrington recalled, 1912 SC 892, 894, that the pursuer’s title to sue had not been much considered in the Inner House in Teulon, but he treated the case, “as it stands”, as authority for the proposition that “in exceptional cases, a beneficiary may sue a debtor to the trust.” His comments were obiter, however, since he sustained the defenders’ plea of no title to sue, on the view that there was no reason why the pursuer should not bring his action in the ordinary way in the name of the executor. The First Division, including Lord President Dunedin, counsel for the first defender in Teulon, considered that Lord Skerrington’s judgment was “quite right” and so found the pursuer liable in expenses “because the progress of the action so far was quite useless”: 1912 SC 892, 895. On condition that the pursuer first paid the expenses, the Division allowed him to amend to put the deceased’s executor in as pursuer – on consignation of a sum to cover his liability in expenses.[92]To summarise. In Raes, Teulon and Morrison the relevant trustees or executrix had actually been called as defenders and there were conclusions for payment to be made in their favour. In Raes and Morrison the pursuers were found to have no title to sue. So neither case can be regarded as any real authority on whether such an action could have been brought without joining the trustees or executrix as defenders. There is no sign of the point having been considered in Teulon.[93]Since in these cases the trustees and executors were parties, there was no room for the defenders to rely on a plea of all parties not called. That plea would indeed have been appropriate if they were necessary parties. But the defenders were in any event relying on the more fundamental argument that the pursuers had no title to sue. Assume, however, that, in some situation, a beneficiary under a trust could raise an action in his own name against a debtor to the trust. If the position were that the trustees would have to be joined as defenders before there could be any effective decree, the position would appear to be much the same as in England.[94]This appeal is concerned with the amendment of pleadings after the expiry of the limitation period. The amendment of pleadings is part of the law of procedure and practice which has traditionally been regarded as the province of the Court of Appeal rather than the House of Lords (or, now, the Supreme Court). Interventions into this area by the highest appellate tribunal have not always received universal approbation (see for instance the trenchant remarks of Sir Henry Brooke in the last chapter of The Civil Procedure Rules Ten Years On ed. Dwyer (2009) pp 453-459).[95]I am in full agreement with the judgment of Lord Collins. His judgment is so comprehensive that I am doubtful whether I can usefully add anything to it, particularly in the circumstances mentioned in the preceding paragraph. I shall add a few observations, but they are not intended to be in conflict with Lord Collins’ reasoning and conclusions.[96]As the English legal system has developed statutes of limitation and procedural rules of court have both become more elaborate, but for a long time there was very little direct interaction between them. In relation to causes of action founded on tort or simple contract, statutes of limitation referred, as the Limitation Act 1980 still does, to the date on which a cause of action accrued. That was the date from which, in the normal case, the statutory period started to run. When a question arose of amending pleadings or adding a new party after the period had run, that was regarded as one factor (but usually a decisive factor) influencing the exercise of the Court’s discretion whether or not to permit the amendment. The Court acted on the principle that it would not be just to deprive the defendant of a vested right of defence, and it often expressed its reasons in a very summary way: see for instance Doyle v Kaufman (1877) 3 QBD 7; Weldon v Neal (1887) 19 QBD 394; Hudson v Fernyhough (1889) 61 LT 722 (in the last case the amendment was allowed, but only on terms that it was not to prejudice the defendant’s defence).[97]For much of the 20th century the same approach was taken. In the much-cited case of Mabro v Eagle, Star and British Dominions Insurance Co Ltd [1932] 1 KB 485, 487 Scrutton LJ said:
“In my experience the Court has always refused to allow a party or a cause of action to be added where, if it were allowed, the defence of the Statute of Limitations would be defeated. The Court has never treated it as just to deprive a defendant of a legal defence.”
The same uncompromising rule of practice was applied even in cases (such as Ingall v Moran [1944] KB 160 and Finnegan v Cementation Co Ltd [1953] 1 QB 688) where the result was to shut out a meritorious claim, arising from a fatal accident, on what many would regard as a technicality. Indeed in the latter case Singleton LJ (at p 699) described the point as “a blot upon the administration of the law.”[98]These hard cases turned on the technical but long-established distinction between the position of an executor (whose standing relates back to the deceased’s death, and is merely confirmed by probate) and an administrator (whose title depends on, and dates from, the grant of letters of administration). That distinction has ceased to be relevant, for present purposes, because of section 35(7) of the Limitation Act 1980 and CPR r 17.4(4) (made pursuant to section 35(7)). Rule 17.4(4) (replacing the former RSC Order 20 r 5(4)) provides:
“The Court may allow an amendment to alter the capacity in which a party claims if the new capacity is one which that party had when the proceedings started or has since acquired.”
Here it is the last four words (added to Order 20 r 5(4) in 1981) that made the important change.[99]Section 35 of the Limitation Act 1980 was enacted to give partial effect to the recommendations of the 21st Report (Final Report on Limitation of Actions) of the Law Reform Committee (Cmnd 6923), published in September 1977. It was the first statutory provision which (by the prohibitory terms of subsection (3), and subject to the important exceptions in subsections (4) to (8)) made it the Court’s duty, and not merely a long-standing rule of practice, to refuse amendments which subvert an established defence based on the Limitation Act. In cases where the amendment was not prohibited, the Court retained its traditional discretion whether or not to permit an amendment, that discretion being exercisable by reference to what was just (embodied, since 1999, in the “overriding objective” in CPR 1.1).[100]Although section 35 of the Limitation Act 1980 was the first provision in a limitation statute which referred to rules of court, rules of court (made under section 99 of the Judicature Act 1925) had already started to acknowledge the existence of limitation statutes. As Millett LJ pointed out in Yorkshire Regional Health Authority v Fairclough Building Ltd [1996] 1 WLR 210, 216, RSC Order 20 r 5 (which apart from r 5(4) was in force long before section 35 came into force on 1 May 1981) gave a limited power to amend pleadings even after the expiration of the limitation period. As early as the mid-1960s questions had been raised as to whether Order 20 r 5 was intra vires: Rodriguez v RJ Parker (Male) [1967] 1 QB 116 (Nield J); Mitchell v Harris Engineering Co Ltd [1967] 2 QB 703 (Court of Appeal). In the latter case Russell LJ said at p 721, after referring to Mabro and some of the other earlier cases:
“But I take these cases to have been decided on grounds of settled practice, albeit attributable to the parties’ position vis a vis the Statute of Limitation. So far as I am aware, no judge said that it would be outside the jurisdiction of the Court to allow the amendment in question: and if it were thought to be a question of substantive law, this would surely have been the immediate and short answer to the application to amend.”
The attacks on the vires of Ord 20, r 5 were therefore rejected. Further insights into the history of these developments can be obtained from the judgments of Hobhouse J in Payabi v Armstel Shipping Corporation (The Jay Bola) [1992] QB 907, 922–928; Staughton LJ in Hancock Shipping Co Ltd v Kawasaki Heavy Industries Ltd [1992] 1 WLR 1025, 1028–1030; and Mance J in Industrie Chimiche Italia Centrale v Alexander G Tsavliris & Sons Maritime Co (The Choko Star) [1996] 1 WLR 774.[101]I shall come back to some wider issues arising on section 35 and the Yorkshire Regional Health Authority case, but first I want to look more closely at the concept of capacity which is the focus of section 35(7) and CPR r 17.4(4). In this context “capacity” is “… being used in the sense of legal competence or status to bring or defend a claim. It is a competence that one may have in one’s own right or on behalf of another person. ... In my judgment the same meaning of capacity must apply in CPR r17.4(4). This means that the alteration in capacity which is referred to is an alteration from a representative capacity, or personal capacity, to another representative capacity, or (in the case of a representative claim) to a personal capacity.” That is how it was put by Arden LJ in Haq v Singh [2001] 1 WLR 1594, paras 18-19, and I agree that that is the right meaning in the context. The best example of a representative capacity is that of an executor or administrator of a deceased person, both offices being included in the compendious expression “personal representative”.[102]When a personal representative takes proceedings as such, he is making a claim in a representative capacity, and his claim form must state what that capacity is (CPR r 16.2 (3)). In the usual case where the deceased’s estate has not been fully administered, the personal representative conducts the proceedings for the benefit of all those interested, whether as creditors or as beneficiaries, in the assets of the estate. No residuary beneficiary has an equitable interest in the assets, only the right to have them properly administered. It is worth emphasising this elementary point because much of the argument addressed to the Court on behalf of the appellant was based on a supposed analogy between the relationship between a properly constituted personal representative and a residuary beneficiary, on the one hand, and an assignor and an equitable assignee, on the other hand. In my opinion that supposed analogy is misleading. As Holroyd Pearce LJ said in Robinson v Unicos Property Corporation Ltd [1962] 1 WLR 520, 526, where the plaintiffs wished to amend to plead an assignment:
“In no sense is the nature of the action altered. The plaintiffs still wish to claim that which they claimed in the beginning. Nor are they suing in a different capacity. Although they now wish to claim by virtue of their right as equitable assignees of the benefits of the principal to the original contract, they still sue in their personal capacity as principals through the same agency on the contract albeit through an assignment of the benefit to them.”
Where one person acquires property as a bare trustee or nominee for the benefit of one or more other persons who are absolutely entitled beneficially, the analogy with an equitable assignment is obviously closer (see for instance Harmer v Armstrong [1934] Ch 65).[103]Just as there is no real analogy between an equitable assignee and a beneficiary interested in an unadministered residuary estate, so there is in my opinion no real analogy between the assignee of a pending cause of action and a residuary beneficiary or a minority shareholder who seeks (under the general law, and not under special statutory provisions) to bring a derivative action on behalf of a deceased person’s estate, or a company. Where, after an action has been commenced, the cause of action is assigned or transmitted by operation of law, the assignment or transmission is not part of the cause of action. It makes it necessary to join the new claimant as a party and it needs to be recorded in the pleadings, but it does not amount to a new cause of action. Not so with a claimant who is not invested with the office of personal representative, but nevertheless seeks to bring a derivative action. As Campbell JA observed, writing for the Court of Appeal of New South Wales in Oates v Consolidated Capital Services Pty Ltd [2009] NSWCA 183, para 105:
“To summarise, a plaintiff who seeks to bring a derivative action under the general law must allege, in the initiating process, facts that show that he or she falls within a recognised exception to the prima facie rule that the proper plaintiff in an action in respect of a wrong alleged to be done to a corporation is the corporation itself. If the initiating process fails to make those allegations, it is liable to be struck out if the defendant chooses to apply to have it struck out … But there is no requirement under the general law relating to derivative actions for leave to be obtained before a plaintiff commences such an action.”
So while he need not obtain prior leave from the court, he must plead the special circumstances entitling him to the court’s indulgence. Those special circumstances are part of his cause of action.[104]In the ordinary case of a simple assignment or transmission of a cause of action after proceedings have been commenced, no question of limitation arises. That was the point that Millett LJ made forcefully in the Yorkshire Regional Health Authority case at p 218 (see also Evans LJ at p 221). The contrary argument, rejected by the Court of Appeal, was that RSC Order 15 r 6 and Order 20 r 5 formed a comprehensive code governing amendments after the expiration of the limitation period. I have no doubt that the Court of Appeal was right in that conclusion, although I am not sure that I agree with (or indeed understand) the refinements of Millett LJ’s reasoning at p 218. Possibly the draftsman of the CPR felt some residual doubt about the position, because CPR 19.5 (3)(c) has added to the post-expiry code a new provision which covers some (but not all) cases of transmission of a cause of action (or liability). That particular oddity is not remarked on in the Law Commission’s 2001 paper Limitation of Actions (Law Com 270) which briefly discusses the topics of amending pleadings at paras 5.5 to 5.19.[105]In the Yorkshire Regional Health Authority case Millett LJ and Evans LJ referred (at pp 219 and 222 respectively) to the “impressive and penetrating” and “scholarly” judgment of Mance J in The Choko Star [1996] 1 WLR 774, to which I have already referred briefly. The Choko Star seems to be the first case in which it was necessary (because of the wrong turning taken in Toprak Enerji Sanayi AS v Sale Tilney Technology plc [1994] 1 WLR 840) to consider whether the transmission of a pending cause of action had implications under the Limitation Act. Previously it seems to have been assumed that there were no such implications. Mance J analysed the position and spelled out why that assumption was correct.[106]The case was concerned with universal succession after the merger of two companies under Article 2504 of the Italian Civil Code. After setting out RSC Order 15 r 7(2) and observing that it went back to rules in force before 1962, and indeed back to the rules in the First Schedule to the Judicature Act 1875 (38 & 39 Vict c77), Mance J said (at p 782), “The problem addressed by Ord 15, r 7 is different: during the course of the proceedings there has been some change affecting the identity of the correct claimant, which could not have been dealt with (or normally even predicted) when proceedings were originally issued.” He then explained why there was no problem under the Limitation Act. “In all such situations, of which death is only the most striking, it seems self-evident that any existing proceedings, properly constituted within the limitation period, should be allowed to continue for or against the party to whom the relevant right or obligation has been transferred in law; and that this should be permitted whether the transfer occurs before or after the expiry of the limitation period.” In the Yorkshire Regional Health Authority case the Court of Appeal approved and followed The Choko Star.[107]Reference was made in argument to the well-known definition of “cause of action” put forward by Diplock LJ in Letang v Cooper [1965] 1 QB 232, 242-243. I am conscious that this is (as Lord Collins says) a sterile topic but I venture to repeat something that I said in a dissenting judgment in Smith v Henniker-Major & Co [2002] EWCA Civ 762, [2003] Ch 182, para 95 (just before the passage quoted by Lord Collins):
“I have to say that in the context of section 35 of the Limitation Act 1980 I am uneasy about the process of lifting either of these classic definitions out of the legal lexicon, as it were, and reading them into the language of section 35(5)(a). The notion of ‘a factual situation’ which ‘arises out of the same facts or substantially the same facts’ as another set of facts is not an easy one to grasp.”
The other classic definition referred to was that of Brett J in Cooke v Gill (1873) LR 8 CP 107, 116.[108]As I have already noted, the most familiar provisions of the Limitation Act 1980, like their predecessors, set time limits by reference to the date on which the cause of action accrued. Its accrual is an event which occurs at a particular point of time. Moreover the Limitation Act in its amended form makes several references to particular varieties of tort which are commonly referred to as causes of action: conversion in sections 3 and 4, libel and malicious falsehood in section 4A, negligence and nuisance causing personal injury in sections 11 and 4 and negligence causing latent damage in sections 14A and 14B. It might therefore be more helpful, for practical purposes, to say that in the context of the Limitation Act “cause of action” means the factual basis of a claim for relief. Typically that factual basis falls into one or more familiar categories (such as negligence) and will consist of a sequence of essential facts (such as the facts establishing duty of care, breach of duty and damage) which must be pleaded and proved (if not admitted) in order to establish the cause of action. The cause of action accrues when the last building block of the essential facts is put into place. For reasons already mentioned the building blocks of the cause of action will not include its transmission (for instance on death or bankruptcy) after the proceedings have commenced, but they will (in the unusual case of a derivative action) include special circumstances relevant to the court’s willingness to entertain the case.[109]The unusual facts of this case are set out in the judgment of Lord Collins. They raise the issue whether, in a case where there is a properly constituted personal representative of Mrs Alice Margot Roberts, appointed on the application of Mr Mark Roberts himself, there are special circumstances justifying the commencement and conduct of derivative proceedings by Mr Mark Roberts.[110]There is ample authority, comprehensively reviewed in the judgment of Lord Collins, as to the need for special circumstances before the Court will countenance a derivative action. Such actions are now relatively common in cases concerned with mismanaged companies, and in many jurisdictions actions by or on behalf of minority shareholders are now regulated by a statutory code (for overseas examples see Oates v Consolidated Capital Services Pty Ltd [2009] NSWCA 183 and Waddington Ltd v Chan [2009] 2 BCLC 82). Derivative actions by beneficiaries under inter vivos trusts or wills are less common, Hayim v Citibank NA [1987] AC 730 (an appeal to the Privy Council from Hong Kong) and Bradstock Trustee Services Ltd v Nabarro Nathanson [1995] 1 WLR 1405 being modern examples. But in all these cases the unifying factor – what has to be special about the circumstances – is that the derivative action is needed to avoid injustice: see Goff J in In re Field, decd [1971] 1 WLR 555, 561; Browne-Wilkinson LJ in Nurcombe v Nurcombe [1985] 1 WLR 370, 378; Pill LJ in the Court of Appeal in this case, [2009] 1 WLR 531, para 59. For the reasons given by Pill LJ at paras 58 to 60, reinforced by the further reasons given by Lord Collins, special circumstances are not made out in this case.[111]In these circumstances the Court does not have to decide the issues as to amendment of pleadings which I have discussed, in general terms, in the first part of this judgment. But those issues have been fully argued, and I have therefore thought it right to comment on them. On the case as a whole my views are closer to those of Pill LJ than they are to the reasons given by Arden LJ in her longer judgment.[112]I differ from Arden LJ (with whom Patten J agreed) as to special circumstances. I also think that she was too ready to accept the analogy between a true derivative claim and a claim by an equitable assignee, or a sole beneficiary under a bare trust. A derivative claim by a residuary beneficiary interested in an unadministered estate is not, with respect, indistinguishable (as suggested in para 32 of Arden LJ’s judgment) but faces a more formidable obstacle than a claim by an equitable assignee or a beneficiary under a bare trust. As to para 34, Mr Roberts’ original claim was a personal claim as a beneficiary. If he were permitted to bring a derivative action, he would be acting in a representative capacity. He would be a beneficiary putting himself forward for the first time as a sort of self-appointed personal representative. It may be that CPR r 17.4 (4) could be satisfied, but I do not regard the point as free from difficulty.[113]For these reasons, and for the much fuller reasons given by Lord Collins, I would dismiss this appeal.[114]I agree with Lord Walker that this appeal should be dismissed on the ground that the appellant has not made out special circumstances such as to make it just for him to be permitted to proceed against the respondents by way of derivative action. I agree with Lord Collins, and with Pill LJ in the Court of Appeal, that, in reaching that conclusion, the judge at first instance, then Mr Paul Morgan QC (now Morgan J), did not err either in principle or otherwise such that it would be permissible for an appellate court to interfere. In this regard, like Lord Walker, I prefer the reasoning of Pill LJ to that of Arden LJ in the Court of Appeal.[115]That conclusion makes it unnecessary to reach a final conclusion on the question whether the court would have power to grant the application for permission to amend to introduce the derivative claim. However, I wish briefly to address that question because it is of potential importance in the future and because I would not go as far as Lord Collins.[116]Lord Collins has set out the relevant provisions of the Limitation Act 1980 (‘the Act’) and the CPR and has discussed the authorities in a masterful way which I could not seek to match. My concern is this. If this were a case in which there were special circumstances such as to make it just that, subject to the issue of limitation, the appellant should in principle be permitted to proceed by way of derivative action against the respondents, I would be concerned if the court had no power to give him permission to amend in circumstances in which it is common ground that the derivative claim, which I will call the ‘new claim’, arises out of the same or substantially the same facts as the appellant’s original claim, which was issued in time.[117]It is thus not in dispute that the appellant satisfies CPR 17.4(2) because he is seeking to add a new claim which arises out of the same facts or substantially the same facts as his existing claim. He is seeking to advance the new claim in a new capacity, namely a representative capacity, which he may or may not have had when the proceedings were started but, if he did not, which he ‘has since acquired’ within the meaning of CPR 17.4(4). It follows that, on the face of CPR 17.4(2) and (4), the court has power to grant an application for permission to amend to alter the capacity in which he sues.[118]The problem arises under CPR 19.5. The respondents’ case may be summarised in this way: i) The new claim is a ‘claim involving … the addition … of a new party’ within the meaning of section 35(2)(b) and (5)(b) of the Act and CPR 19.5(2)(b). ii) It follows from section 35(5)(b) that the addition of the new party must be ‘necessary for the determination of the original action’ and from section 35(6)(b) that it is not to be regarded as ‘necessary’ unless any claim already made in the original action ‘cannot be maintained ... against’ the respondent unless the new party is joined. CPR 19.5(3)(b) reflects section 35(6)(b) except that it provides that the court must be satisfied that ‘the claim cannot properly be carried on’ against the respondent unless ‘the new party is added … as claimant or defendant’. iii) The appellant cannot satisfy that test of necessity because he cannot satisfy the court that his own personal claim could not be maintained or carried on against the respondent unless the administrator was added as a defendant.[119]The first question which arises is thus whether the new claim is a ‘claim involving … the addition … of a new party’ within the meaning of section 35(2)(b) and (5)(b) of the Act and CPR 19.5(2)(b) because, if it is not, no problem arises. If the administrator (or trustee or company in a shareholder’s derivative action) must be joined in every case, this question must of course be answered in the affirmative. On the other hand, if there are circumstances in which it is not necessary to join the administrator in every case, the answer to the question will depend upon the facts of the particular case.[120]The critical questions are those addressed by Lord Collins at para 44 above. He says that the only way in which the action could proceed would be(a) if joinder of the administrator was not necessary at all or(b) if it was not necessary at the time the appellant changed the capacity in which he sued but could be effected later. He notes that the appellant’s case is that the combined effect of section 35(1)(b) and section 35(6)(b) of the Act and CPR 19.5(3)(b) is that (a) the change of capacity is deemed to take effect at the date of the original proceedings and (b) the joinder of the administrator would then be ‘necessary’ for the continuance of what would then be regarded as the claim made in the original action for the purposes of section 35(5)(b) and 35(6)(b).[121]The critical questions are thus whether joinder of the administrator is always necessary and whether such joinder must take place at the outset or can take place later.[122]I entirely accept that the jurisprudence so fully discussed by Lord Collins at paras 42 to 70 above shows that the approach of the courts has been to require the joinder of a trustee where a beneficiary is making a derivative claim in his own name but for the benefit of the trust as a whole. The same principle applies to a derivative action by a shareholder. However, as I see it, in each case it is a procedural rule. Its purpose was explained by Chitty LJ in Spokes v Grosvenor and West End Railway Terminus Hotel Co Ltd [1897] 2 QB 124 at 128 (quoted in full by Lord Collins at para 57 above) as follows:
“To such an action as this the company are necessary defendants. The reason is obvious: the wrong alleged is done to the company, and the company must be party to the suit in order to be bound by the result of the action and to receive the money recovered in the action.”
The principle is now reflected in CPR 19.9(3) in these terms:
“The company, body corporate or trade union for the benefit of which a remedy is sought must be made a defendant to the claim.”
[123]In the note to the rule in para 19.9 of Volume 1 of Civil Procedure 2009 the editors say, by reference to para 32 of the judgment of Arden LJ in this case:
“Although the CPR contains no provision that requires it, where a derivative claim is brought by the beneficiary of an estate, at some stage in the proceedings the personal representative should be joined as a defendant.”
In para 32 Arden LJ said:
“It is sufficient that it has to be done at some stage.”
In putting the principle in that way, she made it clear that it might be appropriate to permit joinder, not at the outset, but at a later stage of the proceedings.[124]Arden LJ so stated after referring to Performing Right Society Ltd v London Theatre of Varieties Ltd [1924] AC 1, 14 per Viscount Cave LC, where he said that the person with the legal right “must in due course be made a party to the action” and where he referred to the decision in William Brandt’s Sons & Co v Dunlop Rubber Co [1905] AC 454, 462 that there may be special cases in which the rule will not be enforced. It is of course true that those were cases in which the plaintiff was an equitable assignee suing in his own name. They are among those discussed by Lord Collins at paras 62 to 67 above. However, it is to my mind of note that, at any rate for this purpose, Arden LJ regarded the requirement of joinder in the derivative class of case as the same as that in the equitable assignee class of case. It did not occur to her (or Pill LJ and Patten J, who agreed with her on this part of the case) that, so far as the time for joinder is concerned, there is a different principle in the two classes of case. Although I recognise, as Lord Walker and Lord Collins make clear, that there are differences between the two classes, I do not for my part see why there should be a difference in this particular respect. Moreover, I do not think that there has been any discussion of this point in the authorities as they stand.[125]The essential reason for the joinder of the trustee, administrator, or company as a defendant in the case of a derivative claim is in order that he or it is bound by the result of the action and entitled to receive the money recovered in the action. The purpose of the joinder of the legal owner in the case of a claim by an equitable assignee is also in order that he or it is bound: see eg EM Bowden’s Patents Syndicate Ltd v Herbert Smith & Co [1904] 2 Ch 86, 91 per Warrington J. Thus there seems no good reason why a different rule should exist in each case as to when he must be joined. In these circumstances, assuming that joinder is necessary at some stage, I prefer the view of Arden LJ that it does not necessarily have to be at the outset.[126]It is true that, as just stated, CPR 19.9(3) provides that in a derivative claim to which the rule applies, which is essentially that of a shareholder, the company must be made a defendant but there is no similar rule in a claim by a beneficiary under a trust. I entirely accept that one would expect the same principles to apply to all classes of derivative claim and, moreover, that CPR 19.9(3) reflects the general rule, but I would also accept the submission made on behalf of the appellant that CPR 19.9(3) is subject to the overriding objective and that the court must have a discretion to postpone the joinder in a particular case.[127]I would go further. As is clear from the William Brandt case [1905] AC 454, the rule in the equitable assignment case is not absolute. I recognise that not to insist upon joinder is said to be exceptional but I am inclined to think that there are in practice very many cases in the modern era in which equitable assignees proceed to recover a debt assigned to them in equity without joining the assignor. However that may be, I wonder why the rule should be absolute in the derivative action case if it is not absolute in the equitable assignment case. There may be circumstances in a particular case which make it just to dispense with the necessity of joinder.[128]For example, there may be no real point in joining the trustee, administrator or company if appropriate undertakings are given by the claimant to hold any monies recovered for his or its benefit, especially if he or it consents. It has been suggested that one reason for joinder is that, if the derivative action fails, the trustee, administrator or company might bring another action in the future. Where such an action would be time barred, again there would be no such risk. Moreover, there is now a general principle that no action should fail for non-joinder of a party: see William Brandt per Lord Macnaghten at p 462. None of these possibilities has been worked out in the cases. In these circumstances I am reluctant to decide this appeal on the basis that there is an absolute rule that the law requires the appellant to join the administrator at all or, alternatively, at the outset.[129]The basis upon which Arden LJ was against the appellant was not that there was an absolute rule that the administrator would have to be joined at the outset because of the nature of the claim. As appears from para 36 of her judgment and para 69 of that of Lord Collins, it was that the administrator would have to be joined before judgment and that it would be contrary to principle for the court to grant permission to amend a change of capacity which would not enable the appellant to proceed to judgment. However, as I see it, that is to conclude that the rule is absolute.[130]For my part, as at present advised, I would hold that the rule is not absolute, that there may be circumstances in which joinder would not be ordered and in any event that joinder does not have to be effected at the outset. Moreover, I do not think that to give the court power to give permission for a claimant to amend his capacity in these circumstances is contrary to principle. CPR 17.4 expressly confers such a power and, although there is a signpost to CPR 19.5, the power in CPR 17.4 is not made subject to it.[131]If the rule is not absolute, or if joinder does not have to be effected at the outset, it seems to me to be at least arguable that the court would have power under CPR 17.4(4) to allow the appellant to change the capacity in which he is suing and that, if he did so, the effect of section 35(1)(b), section 35(6)(b) and CPR 19.5(3)(b) would be that that change of capacity would be deemed to take effect as at the date of the original proceedings and that joinder effected thereafter would be unaffected. The court would thus have power first to permit an amendment to change capacity and to permit joinder thereafter on the basis that joinder would be necessary to allow the proceedings to continue.[132]It is important to note that the above analysis would only give the court power to allow an amendment without joinder at the same time. Whether it would exercise the power would depend upon all the circumstances of the case. It may well be inappropriate, or even (depending upon the circumstances) wrong in principle, to grant an application for permission to amend to change the capacity in which a claimant has been proceeding, but the court would have the power to do so if it appeared just in all the circumstances. I see no reason why the court should not have that power where the new claim arises out of the same or substantially the same circumstances as the existing claim. In this regard it is I think significant that the power is included in a limitation statute. The purpose of such a statute is to protect the defendant against whom a stale claim is made. On the other hand, the staleness of the claim is likely to be of less significance where the new claim arises out of the same or substantially the same facts as the original claim.[133]One of the curiosities of this type of case is that in the ordinary way one would expect it to be the new party who would complain about the addition of a new party. Thus, where a defendant is sought to be added, it is the new defendant whom one might expect to protest. In the instant case, it is most unlikely that, if the administrator were to be joined as a defendant, he would complain. In fact no application has been made by anyone to join the administrator as a defendant. The appellant has not sought to join him and there is no suggestion that he intends to do so. Nor have the respondents sought to join him. If they did, he might well say that he did not object to being joined and might, in any event, not take the point that the claim was time barred.[134]However that may be, I would expect an applicant to explain to the court why the administrator should not be joined at the outset (or perhaps at all) and what was the attitude of other interested parties, including the administrator, other beneficiaries and creditors. They might all consent to the proposal, in which case it might be just to allow the claimant to proceed with a derivative claim without joining the administrator at the outset. Indeed, it might be just even where the administrator did not consent. Such cases would no doubt be exceptional but all would depend on the circumstances. In the instant case the appellant made no attempt to explain the attitude of the administrator or of his aunt Ms Jill Roberts or of the creditors, notably HMRC. So his prospects of persuading a court to exercise its discretion in his favour would be remote.[135]In any event, as I see it, none of the interesting questions I have touched upon needs to be decided in this case because, as I said at the beginning, I agree with Lord Walker that there is no proper basis for interfering with the judge’s conclusion that special circumstances for bringing a derivative action were not established.

Cited in 35 later judgments

and 26 more of the most senior