Rashid v Nasrullah

The registered proprietor of land is deprived of his title in May 1989 in consequence of a series of forged documents, including a forged transfer. The new registered proprietor transfers the land to his son by way of gift later in that year. The son, who is complicit in the fraud, enters into occupation. Over 20 years later the original registered proprietor applies for rectification of the register. Can his claim be defeated by adverse possession on the part of the current registered proprietor? Both the First-tier Tribunal and the Upper Tribunal said “No”. The current registered proprietor appeals.

At the time of the forged transfer the legal regime in force was the Land Registration Act 1925. Under that regime it was possible to acquire title to registered land by adverse possession, if the registered proprietor had been dispossessed for 12 years. In that event he was deemed to hold his registered title on trust for the squatter. It is necessary to emphasise the fact that since the coming into force of the Land Registration Act 2002 the legal landscape as regards adverse possession has been radically changed. This case is concerned with events that all took place before that change.

The facts are somewhat confusing because both the fraudster and the former registered proprietor have the same name: Mohammed Rashid. I will call the fraudster MR1 and the former registered proprietor (and original Respondent to this appeal) MR2. In short, what happened was this. The land in question, 40 Henley Street in Birmingham, has been registered land throughout the relevant period. MR2 was registered at proprietor in November 1985. On 13 April 1989 a forged transfer was executed which purported to transfer the property from MR2 to MR1. It was supported by other forged documents which purported to show that MR2 held the property on trust for MR1. MR1 was registered as proprietor on 18 May 1989. MR2 was in Pakistan at the time. He returned from Pakistan in June 1989; but did not dare return to the property (for reasons that do not matter). After some months in a hostel he was rehoused by the local authority. At about this time he was told by MR1 that the property was now in his (MR1’s) name. MR2 contacted the police; but was told that it was a civil matter; and without relevant documents he was unable to persuade solicitors whom he consulted that he had a case. On 25 October 1989 MR1 transferred the property to his son, Mr Farakh Rashid (the Appellant in this appeal) by way of gift; and he was registered as proprietor on 1 November 1990. The purpose of the transfer was to facilitate the provision of the property to Mr Farakh Rashid as an investment opportunity. Mr Farakh Rashid was well aware of the circumstances of the transfer of the property to MR1 before the gift to him. 12 years from the date of that registration expired on 31 October 2002. The Land Registration Act 2002 came into force on 13 October 2003. MR2 applied to rectify the register on 12 September 2013. MR2 has sadly died since the hearing before the Upper Tribunal. His executor now resists the appeal.

Section 2 (1) of the Land Registration Act 1925 provided:

“(1)

After the commencement of this Act, estates capable of subsisting as legal estates shall be the only interests in land in respect of which a proprietor can be registered and all other interests in registered land (except overriding interests and interests entered on the register at or before such commencement) shall take effect in equity as minor interests…”

Section 3 contained a number of definitions, of which the following are relevant:

“(xi)

“Legal estates” means the estates interests and charges in or over land subsisting or created at law which are by the Law of Property Act 1925, authorised to subsist or to be created at law; and “Equitable interests” mean all the other interests and charges in or over land; an equitable interest “capable of subsisting at law” means such as could validly subsist at law if clothed with the legal estate…

(xv)

“Minor interests” mean the interests not capable of being disposed of or created by registered dispositions and capable of being overridden (whether or not a purchaser has notice thereof) by the proprietors unless protected as provided by this Act, and all rights and interests which are not registered or protected on the register and are not overriding interests, and include—

(a)

in the case of land subject to a trust of land, all interests and powers which are under the Law of Property Act 1925, capable of being overridden by the trustees, whether or not such interests and powers are so protected…

(xx)

“Proprietor” means the registered proprietor for the time being of an estate in land or of a charge”

Section 20 provided:

“(1)

In the case of a freehold estate registered with an absolute title, a disposition of the registered land or of a legal estate therein, including a lease thereof, for valuable consideration shall, when registered, confer on the transferee or grantee an estate in fee simple or the term of years absolute or other legal estate expressed to be created in the land dealt with, together with all rights, privileges, and appurtenances belonging or appurtenant thereto, including (subject to any entry to the contrary in the register) the appropriate rights and interests which would, under the Law of Property Act 1925, have been transferred if the land had not been registered, subject—

(a)

to the incumbrances and other entries, if any, appearing on the register and any charge for capital transfer tax subject to which the disposition takes effect under section 73 of this Act; and

(b)

unless the contrary is expressed on the register, to the over-riding interests, if any, affecting the estate transferred or created,

but free from all other estates and interests whatsoever, including estates and interests of His Majesty, and the disposition shall operate in like manner as if the registered transferor or grantor were (subject to any entry to the contrary in the register) entitled to the registered land in fee simple in possession for his own benefit.

…

(4)

Where any such disposition is made without valuable consideration, it shall, so far as the transferee or grantee is concerned, be subject to any minor interests subject to which the transferor or grantor held the same, but, save as aforesaid, shall, when registered, in all respects, and in particular as respects any registered dealings on the part of the transferee or grantee, have the same effect as if the disposition had been made for valuable consideration.”

Section 69 (1) provided:

“(1)

The proprietor of land (whether he was registered before or after the commencement of this Act) shall be deemed to have vested in him without any conveyance, where the registered land is freehold, the legal estate in fee simple in possession, and where the registered land is leasehold the legal term created by the registered lease, but subject to the overriding interests, if any, including any mortgage term or charge by way of legal mortgage created by or under the Law of Property Act 1925, or this Act or otherwise which has priority to the registered estate.”

It is also necessary to refer to section 114 which provided:

“Subject to the provisions in this Act contained with respect to indemnity and to registered dispositions for valuable consideration, any disposition of land or of a charge, which if unregistered would be fraudulent and void, shall, notwithstanding registration, be fraudulent and void in like manner.”

Section 75 provided:

“(1)

The Limitation Acts shall apply to registered land in the same manner and to the same extent as those Acts apply to land not registered, except that where, if the land were not registered, the estate of the person registered as proprietor would be extinguished, such estate shall not be extinguished but shall be deemed to be held by the proprietor for the time being in trust for the person who, by virtue of the said Acts, has acquired title against any proprietor, but without prejudice to the estates and interests of any other person interested in the land whose estate or interest is not extinguished by those Acts.

(2)

Any person claiming to have acquired a title under the Limitation Acts to a registered estate in the land may apply to be registered as proprietor thereof.”

As mentioned, the Land Registration Act 2002 came into force on 13 October 2003. Schedule 12 of that Act contains transitional provisions. Paragraph 18 of that Schedule provides:

“(1)

Where a registered estate in land is held in trust for a person by virtue of section 75 (1) of the Land Registration Act 1925 immediately before the coming into force of section 97, he is entitled to be registered as the proprietor of the estate.

(2)

A person has a defence to any action for the possession of land (in addition to any other defence he may have) if he is entitled under this paragraph to be registered as the proprietor of an estate in the land.

(3)

Where in an action for possession of land a court determines that a person is entitled to a defence under this paragraph, the court must order the registrar to register him as the proprietor of the estate in relation to which he is entitled under this paragraph to be registered.”

Section 65 of the 2002 Act gives effect to Schedule 4, which deals with alterations to the register. Paragraph 2 enables the court to make an order for alteration of the register for the purpose of “correcting a mistake”. Paragraph 3 provides:

“(2)

If alteration affects the title of the proprietor of a registered estate in land, no order may be made under paragraph 2 without the proprietor's consent in relation to land in his possession unless—

(a)

he has by fraud or lack of proper care caused or substantially contributed to the mistake, or

(b)

it would for any other reason be unjust for the alteration not to be made.

(3)

If in any proceedings the court has power to make an order under paragraph 2, it must do so, unless there are exceptional circumstances which justify its not doing so.”

There is some dispute whether, on the facts, the fraudster MR1 and subsequently his son, Mr Farakh Rashid, have been in possession of the property since the date of their respective registrations as proprietor. I will assume, for the moment, that they have. On that assumption, how does the Limitation Act 1980 apply, as section 75 of the 1925 Act requires?

Section 15 (1) of the Limitation Act 1980 lays down the basic rule, which is that no action may be brought to recover land more than 12 years after the right of action accrued. That gives rise to the question: when does a right of action accrue? Section 15 (6) supplies a partial answer:

“Part I of Schedule 1 to this Act contains provisions for determining the date of accrual of rights of action to recover land in the cases there mentioned.”

So the next port of call is Part I of Schedule 1. Paragraph 1 of that Schedule provides:

“Where the person bringing an action to recover land, or some person through whom he claims, has been in possession of the land, and has while entitled to the land been dispossessed or discontinued his possession, the right of action shall be treated as having accrued on the date of the dispossession or discontinuance.”

However, this is to some extent qualified by paragraph 8 (1) which states:

“No right of action to recover land shall be treated as accruing unless the land is in the possession of some person in whose favour the period of limitation can run (referred to below in this paragraph as “adverse possession”); and where under the preceding provisions of this Schedule any such right of action is treated as accruing on a certain date and no person is in adverse possession on that date, the right of action shall not be treated as accruing unless and until adverse possession is taken of the land.”

The effect of these provisions was the subject of detailed consideration by the House of Lords in JA Pye (Oxford) Ltd v Graham [2002] UKHL 30; [2003] 1 AC 419 in which Lord Browne-Wilkinson gave the leading speech, with which all the other Law Lords agreed. In view of the way in which subsequent cases have approached the question of adverse possession, it is necessary to cite a number of extracts from that speech. Although it is out of order in his speech, I think that it is helpful to begin with how Lord Browne-Wilkinson defined “possession”. He said at [40]:

“… there are two elements necessary for legal possession: (1) a sufficient degree of physical custody and control ("factual possession"); (2) an intention to exercise such custody and control on one's own behalf and for one's own benefit ("intention to possess").”

With that definition in mind, we can now examine other parts of Lord Browne-Wilkinson’s speech. At [35] he explained that before the passing of the Real Property Limitation Acts there had been a concept of “non-adverse possession” which those Acts abolished. Since then the question was whether the limitation period had elapsed since the accrual of the right whatever the nature of the possession. Thus he said:

“From 1833 onwards the only question was whether the squatter had been in possession in the ordinary sense of the word. That is still the law, as Slade J rightly said.” (Emphasis added)

At [36] he said:

“In my judgment much confusion and complication would be avoided if reference to adverse possession were to be avoided so far as possible and effect given to the clear words of the Acts. The question is simply whether the defendant squatter has dispossessed the paper owner by going into ordinary possession of the land for the requisite period without the consent of the owner.” (Emphasis added)

He continued at [37]:

“It is clearly established that the taking or continuation of possession by a squatter with the actual consent of the paper title owner does not constitute dispossession or possession by the squatter for the purposes of the Act. Beyond that, as Slade J said, the words possess and dispossess are to be given their ordinary meaning.”

The references to what Slade J had said picks up an earlier quotation in his speech from Powell v McFarlane (1977) 38 P & CR 452 in which Slade J had said:

“In the absence of authority, therefore, I would for my own part have regarded the word 'possession' in the 1939 Act as bearing the traditional sense of that degree of occupation or physical control, coupled with the requisite intention commonly referred to as animus possidendi, that would entitle a person to maintain an action of trespass in relation to the relevant land; likewise I would have regarded the word 'dispossession' in the Act as denoting simply the taking of possession in such sense from another without the other's licence or consent; likewise I would have regarded a person who has 'dispossessed' another in the sense just stated as being in 'adverse possession' for the purposes of the Act.” (Emphasis added)

With the substitution of Lord Browne-Wilkinson’s definition of “possession” that remains the law. This is made clear by what he said at [38]:

“There will be a "dispossession" of the paper owner in any case where (there being no discontinuance of possession by the paper owner) a squatter assumes possession in the ordinary sense of the word. Except in the case of joint possessors, possession is single and exclusive. Therefore if the squatter is in possession the paper owner cannot be. If the paper owner was at one stage in possession of the land but the squatter's subsequent occupation of it in law constitutes possession the squatter must have "dispossessed" the true owner for the purposes of Schedule 1, paragraph 1…”

It follows from this that, subject to there being a person in whose favour time can run, all that matters is whether (a) the squatter has possession and (b) if he has, he has it without the consent of the true owner. If those two conditions are satisfied, then the squatter has dispossessed the true owner. It is also important to understand that what paragraph 1 of Schedule 1 does is to treat a cause of action as arising on the date of dispossession. If there is dispossession, the cause of action arises. If not, not. In my judgment no other factor is relevant.

Lord Browne-Wilkinson touched briefly on the question whether there is a person in whose favour time can run at [35]:

“Paragraph 8(1) of Schedule 1 to the 1980 Act defines what is meant by adverse possession in that paragraph as being the case where land is in the possession of a person in whose favour time "can run". It is directed not to the nature of the possession but to the capacity of the squatter. Thus a trustee who is unable to acquire a title by lapse of time against the trust estate (see section 21) is not in adverse possession for the purposes of paragraph 8.”

Section 18 (1) of the Limitation Act 1980 provides:

“(1)

Subject to section 21(1) and (2) of this Act, the provisions of this Act shall apply to equitable interests in land as they apply to legal estates.

Accordingly a right of action to recover the land shall, for the purposes of this Act but not otherwise, be treated as accruing to a person entitled in possession to such an equitable interest in the like manner and circumstances, and on the same date, as it would accrue if his interest were a legal estate in the land (and any relevant provision of Part I of Schedule 1 to this Act shall apply in any such case accordingly).”

Section 21 (1) applies to an action by a beneficiary under a trust which is either a claim in respect of fraud or a fraudulent breach of trust; or a claim to recover from the trustee trust property. In such a case there is no limitation period. However, it is now established at the highest level that this sub-section does not apply to a person accountable in equity because of his knowing receipt of trust property or his dishonest assistance in a breach of trust. It only applies to true fiduciaries. A dishonest assister is entitled to rely on the six-year limitation period in section 21 (3): Williams v Central Bank of Nigeria [2014] UKSC 10, [2014] AC 1189. On the face of it, therefore, section 21 does not apply either to MR1 or to Mr Farakh Rashid. If, therefore, MR2 retained an equitable interest in the land, his cause of action accrued on the date of his dispossession.

It seems to me to be plain that if MR2 had remained the registered proprietor and MR1 (and subsequently Mr Farakh Rashid) had simply taken possession of the property Mr Farakh Rashid would now be entitled to apply to be registered as proprietor in place of MR2; relying on section 75 of the 1925 Act and Schedule 12 paragraph 18 (1) of the 2002 Act.

Does the fact that the fraudster MR1 and latterly Mr Farakh Rashid are shown in the register as proprietors of the property make all the difference? Ms Tozer, on behalf of MR2, says that it does. She relies on the decision of this court in Parshall v Hackney [2013] EWCA Civ 240, [2013] Ch 568.

However, before dealing with that case it is necessary to consider the earlier (and controversial) decision of this court in Malory Enterprises Ltd v Cheshire Homes Ltd [2002] EWCA Civ 151, [2002] Ch 216. In that case, the true owner was a company referred to as Malory BVI. Another company was dishonestly set up with the same name, and it procured a new land certificate. That company was referred to as Malory UK. Malory BVI remained in possession (and in actual occupation) of the land. Malory UK then sold the land to Cheshire Homes. It was agreed that the register should be rectified so as to show Malory BVI as the registered proprietor. However, Malory BVI claimed damages for trespass against Cheshire Homes. The facts relating to this dispute are not entirely clear from the report; but what seems to have happened is that after the dispute arose Cheshire Homes entered the property (which until then had been in the possession of Malory BVI) and demolished a building on it. Arden LJ set out the terms of both section 20 and section 69 of the 1925 Act. The critical part of her reasoning for present purposes is at [64] and [65]:

“64 Although Malory UK had no title to convey to Cheshire, the position of Cheshire once it is registered as proprietor is governed by section 69 of the LRA. Accordingly, when it became the registered proprietor of the rear land, Cheshire was deemed to have vested in it "the legal estate in fee simple in possession".

65 However, section 69 deals only with the legal estate. Unlike section 5, which deals with first registration, that registered estate is not vested in Cheshire "together with all rights, privileges, and appurtenances". Moreover, since the transfer to Cheshire could not in law be of any effect in itself, in my judgment it cannot constitute a "disposition" of the rear land and accordingly section 20 cannot apply. In those circumstances, Cheshire's status as registered proprietor is subject to the rights of Malory BVI as beneficial owner. On this point I accept the submissions of Mr Dagnall and reject those of Mr Martin. It follows that I accept that Malory BVI has sufficient standing to sue for trespass even without seeking rectification of the register because it is the true owner and has a better right to possession: see Chowood Ltd v Lyall (No 2) [1930] 2 Ch 156, 163-164.”

It will be seen that there were two critical points. The first was that the transfer to Cheshire was not a “disposition” and hence section 20 could not apply. The second was that there was a difference between section 20 under which both legal estate and beneficial interest would pass, and section 69 under which only the legal estate would pass, leaving the beneficial interest in Malory BVI.

The decision in Malory has attracted considerable academic comment, most (but not all) of which is hostile. In Malory, and in cases which have followed it, it has generally been the case that the person who is wrongly registered as proprietor as a consequence of fraud is himself also an innocent victim of the same fraud. This has led to discussion of difficult policy questions whether the land registration system should protect true owners (sometimes called “static security”) or innocent buyers (sometimes called “dynamic security”). It is in that context that distinguished commentators such as Professor Elizabeth Cooke (who was also the judge in the UT in our case) referred rather apocalyptically to the decision in Malory as “disastrous to the integrity of the statute”: [2013] Conv 344 at 340. However, as she herself recognised in her article, the provisions of the Land Registration Acts (whether the 1925 Act or the 2002 Act):

“… cannot make a generalised assertion about the beneficial ownership of the registered land…. Nor can it affect a trusteeship imposed upon the registered proprietor as a result of his or her conduct – for example, a constructive trusteeship arising from the circumstances of acquisition.”

On the other hand Mr Simon Gardner has defended the decision in Malory “both in terms of intrinsic doctrine and of compatibility with the surrounding legislation”: [2013] Conv 530 at 537; and Emmett and Farrand on Title do so in typically trenchant terms at paragraph 9.009.1. The most recent comment is that of the Law Commission in Consultation Paper 227 in which the Commission says at paragraph 13.47:

“The Malory 1 argument is a deeply problematic approach. It is unprincipled; it nullifies the provisions of the statute for protection of the proprietor in possession; it is vulnerable to random outcomes; and it raises problems where an indemnity is claimed. The Malory 1 argument has now been held to be wrong and cannot be relied upon, but it is worth re-capping the reasons why it was so problematic.”

The last sentence of that quotation is a reference to the decision of this court in Swift 1^st Ltd v Chief Land Registrar [2015] EWCA Civ 330, [2015] Ch 602. In that case the registered proprietor remained in occupation of her property throughout. A fraudster executed what purported to be legal charges over the property, in consequence of which the lender paid money. It was agreed that the register should be rectified to expunge the charge; and the issue before the court was whether the lender was entitled to be indemnified against its loss by the Land Registry. Ultimately the court held that it was so entitled, because under Schedule 8 paragraph 1 (2) (b) of the 2002 Act it was deemed to have suffered loss by reason of the rectification. However, in the course of his judgment Patten LJ considered the decision in Malory. Before coming to what he said, I should first note that leading counsel for the Chief Land Registrar told the court that it was common ground that Malory “was wrongly decided on the beneficial ownership point, because it was per incuriam section 114 of the Land Registration Act 1925 as well as inconsistent with the reasoning in the Argyle case and for other reasons.” So the point was not in fact in issue in Swift.

Patten LJ posed the question at [1]:

“whether the much-debated decision of this court in … Malory … was decided per incuriam in so far as the court held that the innocent victim of a forged disposition acquired only the legal estate and not the beneficial ownership of the property.”

It will be noted that the question posed was limited to the innocent victim of a forged disposition. Patten LJ set out the critical parts of the reasoning in Malory, together with a selection of academic criticism of it; and said at [40] that it was not possible to confine the reasoning in Malory to the 1925 Act. It applied to the 2002 Act as well. Accordingly, he held that it was necessary to consider whether Malory was decided per incuriam. At [41] he said:

“Absent a trust, the legal estate carries with it all rights to the property and equity has no rôle to play in separating legal from beneficial ownership. Therefore to achieve the result which obtained in Malory it is necessary to construe section 69 of the 1925 Act as creating a departure from the usual rule that legal and beneficial ownership are, without more, indissoluble and to construe the references to the legal estate as meaning a bare legal estate of the kind that would be vested in a trustee. This runs contrary to the general principle embodied in section 69 that the register is a complete record of all enforceable estates and interests except for overriding interests and that it is registration rather than the transfer which confers title.”

The key to this passage is, in my judgment, the phrase “absent a trust”. If the circumstances surrounding the acquisition and registration are such as to impose a trust on the transferee, then there is no reason to read section 69 as dealing with more than the bare legal estate. He then turned to consider whether Malory had been decided per incuriam. The first point taken was that Arden LJ:

… “makes no reference in her judgment to the earlier decision of the Court of Appeal in Argyle Building Society v Hammond (1984) 49 P & CR 148 where Slade LJ, giving the judgment of the court, refers to the statutory magic of section 69 (1) which has the effect of vesting title by registration even when the transfer has been a forgery.”

I make two observations about that. First, it is clear that Argyle was cited to the court in Malory, because in fact Arden LJ does refer to it at [45] in introducing the issue of trespass. Second, Argyle was concerned with a different point, namely whether a registered proprietor had the statutory power to create legal charges, even if his own registration had been procured by forgery. It was that question which the court in Argyle answered affirmatively. The question of beneficial interest did not arise. The second point taken was that Arden LJ did not refer to section 114 of the 1925 Act. Patten LJ said at [44]:

“An important, perhaps even critical, part of Arden LJ's reasoning in para 65 of her judgment in Malory was that section 20 of the 1925 Act had no application to a forged disposition. But section 114 makes it clear that, although a registered disposition which would be fraudulent and void if unregistered remains fraudulent and void even if registered, this takes effect subject to the provisions of the 1925 Act with respect to registered dispositions for valuable consideration one of which is, of course, section 20. It is not therefore possible to construe section 20 as having no application to a fraudulent transfer for valuable consideration that is registered.”

This criticism of Malory may be right (although Emmett and Farrand argue that it is not). However, for reasons that I will explain I do not think that this bears on the outcome of our case.

I come now to Parshall v Hackney [2013] EWCA Civ 240, [2013] Ch 568. Ms Tozer says that that case applies to this case. Mr Paget argues (a) that it is confined to cases in which there has been no fraud; or (b) that it was wrongly decided and that we are free not to follow it. The second argument came in by way of amendment to the grounds of appeal, for which we gave permission at the start of the hearing.

Parshall was a case in which the same piece of land had been registered in two titles (No 29 and No 31). The owners of No 31 had possession of the land from July 1988 when they demarcated the land as a parking space. The land was subsequently removed from the title of No 29. Both the registration of the land in the title of No 31 and its removal from the title of No 29 were mistakes by the Land Registry. When the owners of No 29 applied to rectify the register, they were met by the defence that the owners of No 31 had acquired title by adverse possession. This court rejected that defence. Mummery LJ gave the leading judgment.

At [38] he referred to Pye; but did not quote any of Lord Browne-Wilkinson’s speech, or derive any principles from it. Instead he cited Buckinghamshire County Council v Moran [1990] Ch 623 in which Slade LJ said:

“Possession is never ‘adverse’ within the meaning of the Act of 1980 if it is enjoyed under a lawful title. If, therefore, a person occupies or uses land by licence of the owner with the paper title and his licence has not been duly determined, he cannot be treated as having been in ‘adverse possession’ as against the owner with the paper title.”

But the first sentence of that citation is explained by the second. All that Slade LJ was discussing was a case in which a person has possession with the consent of the paper owner. If he was purporting to say anything more (which would have been surprising in view of his decision at first instance in Powell which I have quoted at [20] above) it cannot stand with the subsequent decision of the House of Lords in Pye (or, for that matter, with the explanation of that dictum in the earlier decision of this court in Rhondda Cynon Taff County Borough Council v Watkins [2003] EWCA Civ 129, [2003] 1 WLR 1864 at [24] to [28]).

At [43] Mummery LJ introduced Malory as follows:

“[Malory] was relied on below as authority for the general proposition that rectification of the land register was not a precondition of an equitable right to possession of land in dispute. It was argued that the owners of No 29 therefore had a right of action to recover the disputed land without obtaining a prior order for rectification.”

Having quoted the relevant part of Arden LJ’s judgment, Mummery LJ continued at [47]:

“The claimant disputes the relevance of that ruling to the circumstances of the present case, which do not involve the registration of any relevant transfer tainted by fraud or any separation of the legal and equitable interest in the disputed land. Against that, the defendant says that the absence of fraud in this case does not affect the application of the principle, which is that there was no legal basis for the registration of the defendant's predecessors in title as proprietors of the disputed land and that the owners of No 29 could have brought an action for the recovery of the disputed land without obtaining rectification of the land register before action. Those submissions are dealt with in more detail below.”

The case for the claimants, then, was not that Malory was wrong, but that it did not apply where fraud was not involved.

Having set out the rival submissions Mummery LJ said:

“[83] In my view, the key to assessing the defendant's claim to a possessory title is to be found in asking the correct questions under the 1980 Act and then answering them against the background of the concurrent registrations of the disputed land and the object, scheme and language of the relevant legislation and the overall justice of the case.

[84] The correct questions under the 1980 Act are: (1) Did the owners of No 29 have a right of action in the period 1988 to 2003 against owners of No 31 for recovery of the disputed land? If so, (2) Did time run in favour of the owners of No 31 as persons in adverse possession of the disputed land?”

I am bound to say that I cannot reconcile the first of these questions with the decision of the House of Lords in Pye. As I hope I have shown, in Pye the House decided that the question is simply whether the defendant squatter has dispossessed the paper owner by going into ordinary possession of the land for the requisite period without the consent of the owner. If he did then the effect of Schedule 1 paragraph 1 is to treat a right of action as having arisen at the moment of dispossession. It is also pertinent to observe that a right of action in trespass is not dependent on title; it is dependent on possession. Be that as it may, Mummery LJ went on to say at [87]:

“There was no dispossession in July 1988, because the taking of possession of the disputed land was not unlawful. It was lawful for the owners of No 31 to take and remain in possession of the disputed land, because they had a registered title to it. As long as they remained registered proprietors of the disputed land, that possession would be lawful and could not be adverse to the owners of No 29.”

I have considerable difficulty with this passage too. As Lord Browne-Wilkinson demonstrated in Pye a “dispossession” simply means that one person has taken possession from another without that other’s consent. Mummery LJ appears to me to have re-introduced the concept of non-adverse possession, which Lord Browne-Wilkinson was at pains to abolish. Moreover, if the owners of No 31 were in possession without the consent of the owners of No 29, I find it difficult to see why that possession was not “adverse” to the owners of No 29, even if some additional content needs to be given to that word. It is, in my judgment, clear that in stating that the dispossession was not unlawful because of the existence of the registered title Mummery LJ put an interpretation on the cited passage from Buckinghamshire which misunderstands what Slade LJ had said and is irreconcilable with Lord Browne-Wilkinson’s subsequent analysis in Pye.

Ms Tozer defended the decision, not on the basis of the way that Mummery LJ expressed himself, but on the basis that what he decided was that the registered proprietor was not a person in whose favour time could run. Hence Schedule 1 paragraph 8 precluded a right of action from arising. It seems to me to be clear from [87], which I have quoted above, that Mummery LJ held that there was no dispossession. The reason that there was no dispossession was, in his view, a function of the registered title, rather than the consent of the true owner. But that is flatly contrary to what was decided in Pye. He repeated the point that possession was lawful because of the title at [88], [90] and [94]. At [100] and [101] he summarised the decision:

“[100] In brief, the legal position is that the owners of No 29 did not have a right of action against the owners of No 31 for the recovery of possession of the disputed land for putting up the chain link fence round the disputed land in July 1988 or for parking on it. No right of action arose then or subsequently, because there was no unlawful act of taking possession by the owners of No 31; they were entitled in law, even as against other persons with a registered title to the disputed land, to go into it and to remain in possession of it. Time did not begin to run against the owners of No 29.

[101] Just as time did not begin to run against the owners of No 29, so it did not begin to run in favour of the owners of No 31, as their possession of the disputed land was referable to their registered title and was not unlawful or adverse within the meaning of the 1980 Act.” (Emphasis in original)

It is clear that [101] is concerned entirely with whether the act of taking possession was unlawful: not whether it was a taking of possession without the consent of the true owner. Thus in that paragraph Mummery LJ was not giving “dispossession” its ordinary meaning as explained in Pye. The same point is made at [102], namely that possession was not unlawful. It is true that the quoted sentence adds “or adverse” but that is too slender a peg upon which to hang Ms Tozer’s argument.

In my judgment Parshall is irreconcilable with Pye. In those circumstances we are entitled, indeed bound, to follow the decision of the House of Lords in Pye in preference to the decision of this court in Parshall: Fitzsimons v Ford Motor Co Ltd [1946] 1 All ER 429; Great Peace Shipping Ltd v Tsavliris Salvage (International) Ltd [2002] EWCA Civ 1407, [2003] Q.B. 679 at [158] to [160]; Starmark Enterprises Ltd v CPL Distribution Ltd [2001] EWCA Civ 1252, [2002] Ch. 306 at [62] to [64] (Kay LJ) and [97] (Peter Gibson LJ).

However, if I am wrong, then I consider that Parshall does not apply to the facts of our case. In Parshall Mummery LJ went on to accept the claimants’ argument that Malory was not in point. As he put it:

“[93] In my judgment, the Malory case [2002] Ch 216 is not, when properly understood, against the claimant on this point. It was not a case on the effect of first registration nor was it a case of concurrent registrations. It was a case of a purported disposition of registered land that was held not to affect the beneficial ownership of the land, because of the fraud of the transferor. It was held that only the bare legal title passed to the transferee, who was registered as proprietor. The claimant, who had been defrauded, was left with the beneficial ownership of the land and that was held to be sufficient to entitle the claimant in that case to sue for trespass.

[94] In this case there are no circumstances from which a separation of legal title and beneficial ownership of the disputed land could be inferred or implied. Subject to rectification of the register by order, both parties have a good legal and beneficial title to the disputed land conferred by registration.”

Are there circumstances in this case from which a separation of legal and beneficial titles can be inferred or implied? The three important features of this case are, in my judgment:

It is, of course, the case that the mere fact that fraud is involved somewhere does not of itself mean that there is a separation of legal and beneficial title. However, I consider that Snell’s Equity (33^rd ed) correctly distinguishes between two types of case, this case being of the first type:

“26-012 (b) Fraudulent taking.

A distinction must be drawn between fraud consisting in the outright taking of a person’s property, wholly without his consent, and a transaction induced by a fraudulent misrepresentation. In the first case, it has been said that a thief who steals the property of another holds it on constructive trust for the claimant. The thief’s possessory title is subject to the claimant’s equitable entitlement to have the property specifically restored to him so that he holds it as a constructive trustee. The consequence is that the claimant need not rely on the less advantageous common law rules of tracing to recover his property.

26-013 (c) Fraudulently induced transfer.

In the second case, where the claimant is the victim of a fraudulent misrepresentation which induces him to transfer his property to his defendant, the transaction is valid until the claimant elects to rescind it. In the meanwhile, the defendant holds his legal interest in the property as beneficial owner, though subject to the claimant’s equity to rescind, a right which is itself proprietary in character. On rescission by the claimant, the defendant holds his legal interest in the property on resulting trust. Since the trust arises only at that stage, the defendant cannot be taken to have owed duties qua trustee before then. Nor can any misapplication of money by the defendant be treated as a breach of trust until after rescission. The possibility of rescission leading to the imposition of a resulting trust will be barred if the claimant has elected to affirm the transaction.”

We were referred to the decision of Rimer J in Shalson v Russo [2003] EWHC 1637 (Ch), [2005] Ch 281. That was a case in which Mr Mimran was fraudulently induced to make a substantial loan. One question that arose was whether he retained a proprietary interest in the money. The judge held that where a person was fraudulently induced to lend money to another the money advanced did not become subject to an immediately binding constructive trust in the lender's favour but became the borrower's property both legally and beneficially. He discussed this question at some length at [106] to [119]. At [108] he made the point that the impugned contract of loan was voidable: not void. That is a distinction of importance and formed the foundation of his reasoning. As he said at [110]:

“… a thief ordinarily acquires no property in what he steals and cannot give a title to it even to a good faith purchaser: both the thief and the purchaser are vulnerable to claims by the true owner to recover his property. If the thief has no title in the property, I cannot see how he can become a trustee of it for the true owner: the owner retains the legal and beneficial title.”

Rimer J contrasted that case with one in which the disposition was no more than voidable. As he explained at [119]:

“In my view, the position immediately after the making by Mr Mimran of each of his loans to Westland was that the money advanced became Westland's property both legally and beneficially: that is what Mr Mimran intended should happen and that is what did happen. In place of his money, Mr Mimran acquired a chose in action in the nature of a right to claim repayment by Westland of the money so advanced. The money itself became Westland's to do as it liked with. I reject the argument that in equity the money remained Mr Mimran's from the moment it was paid over.”

The starting point, as it seems to me, must be to consider the effect of the registration of the fraudster, MR1. Since the transfer was a forgery, it would have been void if the land had been unregistered. Section 114 of the 1925 Act therefore applies to it. It is equally void for the purposes of registered land. This is not, therefore, merely a voidable transaction. Nevertheless, MR1 became the registered proprietor. Ms Tozer initially submitted that the effect of section 114 was that nothing passed to MR1 upon his registration as proprietor. I regard that as an impossible submission. The whole point of the register is to record who is entitled to the legal estate in any given parcel of registered land. Moreover, it is contrary to authority. In Argyle (referred to above) this court decided that even where a transfer was forged a person registered as proprietor as a result of the forgery acquired the legal estate as a result of section 69. Slade LJ explained at 155:

“The effect of the registration of Mr. and Mrs. Hammond as proprietors of the freehold interest in the house must have been to vest in them the legal fee simple, whether or not the purported signature of Mrs. Mary Steed on the transfer to them was forged. This is by virtue of the “statutory magic” effected by section 69 (1) of the 1925 Act, to which Ruoff & Roper refer.” (Emphasis added)

This court followed that ruling in Swift 1^st. It follows, in my judgment, that upon the registration of MR1 as proprietor, section 69 of the 1925 Act would have had the result that he was deemed to have vested in him the legal estate. It also follows, in my judgment, that Rimer J’s example of the thief cannot be directly applied to the facts of the case, because the effect of section 69 meant that the legal estate did become vested in MR1. After some discussion I think that Ms Tozer was inclined to accept that this was at least a possible analysis.

Would that registration also have carried the equitable interest; or was it confined to the bare legal title, leaving MR2 as the owner in equity? Since this was an outright taking of MR2’s property wholly without his consent (or even his knowledge); and there was no disposition for valuable consideration, I consider that it falls into the first of the two categories identified by Snell. In my judgment, therefore, there are good grounds for inferring a separation of the legal and beneficial interests in the property. It has the effect, as Mr Gardner proposes in the article to which I have referred, of “engineer[ing] a replica of voidness” in the context of the land registration regime. It follows, in my judgment, that upon registration MR1 acquired no more than the legal estate, leaving MR2 as beneficial owner. It is also consistent with Professor Cooke’s view in the cited article that a constructive trusteeship imposed as a result of a person’s conduct in acquisition can be accommodated within the system of registration.

Ms Tozer then submitted that whatever the position might have been in relation to the original registration of MR1, any defect was cured by the subsequent registration of Mr Farakh Rashid consequent upon his father’s gift to him. That, she said, came about because of the effect of section 20 (1), as applied to dispositions not made for valuable consideration by section 20 (4). If I am right thus far in holding that the registration of MR1 as proprietor did not carry with it MR2’s beneficial interest in the property, then I do not consider that the registration of Mr Farakh Rashid expunged it. That is because section 20 (4), unlike section 20 (1), stipulated that a gratuitous disposition would “so far as the transferee or grantee is concerned, be subject to any minor interests subject to which the transferor or grantor held the same.” A minor interest is an interest which cannot be disposed of or created by a registered disposition: section 3 (xv). But the only interests which are capable of being registered are legal estates; all other estates and interests take effect in equity as minor interests: section 2 (1). It follows that MR2’s beneficial interest was a minor interest. As such Mr Farakh Rashid’s legal estate was subject to that interest.

In her decision in the Upper Tribunal in the present case at [14] Judge Cooke took the view that Mr Farakh Rashid “must have been holding the land on constructive trust” for MR2. I agree. However, at [37] she said that she considered that it would be nonsensical to regard MR1 as having two concurrent estates in fee simple: one arising as a consequence of possession and the other arising as a consequence of registration. But once one arrives at the conclusion that the legal and beneficial interests in the property are separated as a consequence of MR1’s fraud, and that the separation is not cured by the subsequent gift, there is no conceptual difficulty involved in regarding MR1’s possession for 12 years as barring MR2’s equitable interest. That is what section 18 of the Limitation Act 1980 expressly provides. That section could only be overridden if MR2 had been entitled to rely on section 21 (1).

On the authority of Malory (which in this respect has not been doubted) MR2 could have sued for possession without waiting for the register to be rectified. But in any event his dispossession by MR1 would have had the effect under Schedule 1 to the Limitation Act 1980 of treating his right of action as accruing at that time. I do not accept the argument that MR1, let alone Mr Farakh Rashid, was not a person in whose favour time could run. Because of section 18 of that Act MR2’s right of action would apply just as much to his beneficial interest as to a legal estate. On the authority of Williams, MR2 does not have the benefit of section 21 (1) of the 1980 Act, because MR1 was not a true fiduciary. If the land had been unregistered land, at the expiry of 12 years MR2’s title would have been extinguished: Limitation Act 1980 s 17. However, since the land was registered, section 75 of the 1925 Act applied. The purpose of that section is to adapt the rules relating to unregistered land so as to accommodate the land registration system. Apart from that the Limitation Act 1980 applies to registered land in the same way as it applies to unregistered land. No doubt it was framed on the basis that the dispossessed owner would be the registered proprietor; and is unlikely to have expressly contemplated the case in which the squatter had fraudulently procured his own registration. But if, as I consider, MR2 had retained his beneficial interest in the property, I see no reason to decline to apply section 75 to that extent. There are two possible analyses, both of which lead to the same result. On the first analysis, at the expiry of 12 years from the date of dispossession MR2’s equitable “estate” would, by virtue of section 75 of the 1925 Act, have been held by him on trust for MR1. The alternative way of analysing section 75 is to say that during the running of the limitation period MR1, and subsequently Mr Farakh Rashid, held the legal estate on a constructive trust for MR2; but that after the expiry of that period MR2’s beneficial interest was extinguished; with the consequence the legal and beneficial interests in the property would have merged. Either way, that has the consequence that under Schedule 12 paragraph 18 of the 2002 Act Mr Farakh Rashid is both entitled to be registered as proprietor and also has a defence to an action for possession. If he has a defence to an action for possession under paragraph 18 (2), then the registrar must register him as proprietor in accordance with paragraph 18 (3).

If, as I consider, Mr Farakh Rashid is entitled to be registered as proprietor of the property, there can be no point in rectifying it so as to show the executor as proprietor. He would immediately be met with an unanswerable claim to have the register re-rectified so as to comply with Schedule 12 paragraph 18 of the 2002 Act.

I said earlier that I would assume that MR1 and Mr Farakh Rashid had been in possession. Ms Tozer submitted that although the FTT had found that Mr Farakh Rashid was in possession at the time of MR2’s application to rectify the register, it had not found that he (and his father) had been in possession for the requisite period of 12 years. In those circumstances we should remit the case to the FTT in order for it to make further findings of fact (although she said that no further evidence should be permitted). There are, apparently, no transcripts of evidence.

In the FTT Judge Verduyn recorded at [3]:

“While [MR2] does not formally accept that [Mr Farakh Rashid] has demonstrated he is in possession of the Property, the evidence is that it has been treated as an investment property in his hands and has been rented out. It follows that … [he] is to be treated as a proprietor in possession of the Property.”

That is, to my mind, a finding of fact which betrays no error of law; and is therefore unappealable. It is a general finding not merely that Mr Farakh Rashid was in possession at the moment of the application; but that the property had been treated as an investment property and rented out. It seems to have been accepted by the Upper Tribunal at [34]. It is not disputed that the receipt of rents and profits amounts to possession for the purposes of the law of adverse possession. In addition the whole thrust of MR2’s case was that he had been kept out of possession of the property since his return from Pakistan in the summer of 1989. The issue of adverse possession was one that was raised before the FTT; and if the FTT had considered that Mr Farakh Rashid had not been in possession for a period of 12 years, that would have been a very short answer to the point. I would decline the invitation to remit the case for further findings, some 20 to 30 years after the crucial events.

In my judgment the fact that Mr Farakh Rashid would be entitled to re-rectify the register so as to show him as registered proprietor amounts to “exceptional circumstances” for the purposes of paragraph 3 (3) of Schedule 4 to the 2002 Act.

The result of my analysis so far would lead to the conclusion that the right of MR2 to rectify the register would be defeated by a defence of having acquired title by adverse possession.

However, Ms Tozer had an alternative argument based on the doctrine of illegality, which was the subject of extensive consideration in Patel v Mirza [2016] UKSC 42, [2017] AC 467. That case concerned a claim for unjust enrichment when money was paid over under a contract for an illegal activity which, in the event, did not happen. But Lord Toulson, giving the judgment of the majority, conducted a wide-ranging review of the principles that a court should apply in deciding whether a claim should be defeated by a defence of illegality. The first point to make is that the case did not concern the applicability of limitation periods. But that is what this case is about. Second, the whole basis of adverse possession is that the squatter has committed a tort: namely the tort of trespass. As Nourse LJ crisply put it in Buckinghamshire:

“The essential difference between prescription and limitation is that in the former case title can be acquired only by possession as of right. That is the antithesis of what is required for limitation, which perhaps can be described as possession as of wrong.”

What the law of adverse possession is concerned with is what happened on the ground. Patel v Mirza is not, in my judgment, any warrant for the rewriting of history. Moreover, in R (Best) v Chief Land Registrar [2015] EWCA Civ 17, [2016] QB 23 this court held that even where possession itself amounted to a criminal offence that did not prevent the ordinary operation of the law of adverse possession in relation to limitation periods. The court reached that result by considering the rival policy considerations, including the coherence of the principles upon which the law of adverse possession is based. The balance struck by this court in Best was noted with apparent approval by the Supreme Court in Patel at [79].

It is right to say that in Best, the court did not rule out the possibility that some crimes might prevent time from running in a trespasser’s favour. The instances given were bribing a police officer not to take action; or murdering the true owner to prevent him from recovering possession. On the other hand, in Lambeth LBC v Blackburn [2001] EWCA Civ 912, (2001) 82 P & CR 494 the fact that the squatter broke into the property and broke the true owner’s lock did not prevent time running.

Ms Tozer argues that but for the fraud MR2 (and subsequently Mr Farakh Rashid) would not have been in possession. She also argues that it was the fraud which prevented MR2 from taking any action to recover possession, because he could not convince solicitors that he had a case without the relevant documentation, whereas if the property had remained in his name he would have little difficulty in recovering it. The relevant findings of the judge in the FTT were these:

“… a solicitor he approached said that he had insufficient evidence to support his claim.” (para [10]).

“… a solicitor he visited on a free consultation was not interested without documentary proof for his contentions. … He was told by the solicitor he visited that the possession of keys to the Property was insufficient…. As a result he was in no position to press matters or, it seems, seek Legal Aid for further investigation.” (para [14])

“[MR2’s] description of being brushed off by … a solicitor on the basis there were no documents in corroboration [is] believable and an unlikely invention.” (para [29])

Even if the doctrine of illegality were to apply, I cannot see that it could do more than invalidate the registration. It cannot undo the fact that MR1 and latterly Mr Farakh Rashid have been in possession for over 20 years. Limitation periods exist in all or most developed legal systems. They serve a number of purposes. I said in RE v GE [2015] EWCA Civ 287 at [75]:

“People arrange their affairs on the basis that stale claims cannot be pursued. Insurance cover is taken out and maintained on the basis that claims against the insured must be timeously brought. Organisations maintain document destruction policies fashioned according to limitation periods. Businesses raise finance and pay dividends on the basis that their accounts can be settled. Householders enjoy their gardens on the basis that long possession will not be disturbed. In addition the state has an interest in the principle of legal certainty. As Plumer MR said in Cholmondeley v Clinton (1820) 2 Jac & W 1, 140:

"The statute is founded upon the wisest policy, and is consonant to the municipal law of every country. It stands upon the general principle of public utility. Interest reipublicae ut sit finis litium, is a favorite and universal maxim. The public have a great interest, in having a known limit fixed by law to litigation, for the quiet of the community…"”

McHugh J explained in Brisbane South Regional Health Authority v Taylor [1996] HCA 25, (1996) 186 CLR 541

“A limitation period should not be seen therefore as an arbitrary cut off point unrelated to the demands of justice or the general welfare of society. It represents the legislature's judgment that the welfare of society is best served by causes of action being litigated within the limitation period, notwithstanding that the enactment of that period may often result in a good cause of action being defeated.”

There can be no doubt that bad people are entitled to the benefit of limitation periods. Subject to the court’s discretionary power to override limitation periods, sexual abusers are entitled to rely on the limitation period in section 11 of the Limitation Act 1980. Dishonest assisters in a breach of trust are likewise entitled to rely on the limitation period in section 21 (3). Common law fraudsters are entitled to rely on the limitation period in section 2. Moreover, the Limitation Act contains its own provisions for dealing with criminality. The most familiar is section 32 which extends the limitation period in cases of fraud and deliberate concealment, so that time does not begin to run until the claimant has discovered the fraud or deliberate concealment; or could with reasonable diligence have discovered it. Theft is dealt with by the rather complicated provisions of section 4, the general effect of which is that the thief cannot rely on a limitation period; but a purchaser in good faith from the thief can. Even where civil recovery proceedings are brought under the Proceeds of Crime Act 2002 in respect of recoverable property there is a limitation period: Limitation Act 1980 sections 27A and 27B. The limitation period in such a case is 20 years. Those provisions show quite clearly, to my mind, that Parliament intended that even criminals should have the benefit of a limitation period. The Land Registration Act 2002 also has its own calibrated provisions dealing with the alteration of the register in cases of fraud.

In Patel v Mirza Lord Toulson warned at [100] against the temptation to decide cases on the basis that a wrongdoer should not get something out of his wrongdoing, rather than on the question whether allowing the illegality to have legal effect would produce inconsistency and disharmony in the law. Were we to apply the doctrine of illegality to the facts of this case, we would, in my judgment, undermine the law of adverse possession; and, indeed, of limitation more generally. In my judgment, on the facts of this case, the general principle laid down by Patel v Mirza (which as I have said had nothing to do with limitation periods) cannot be used to override the calibrated scheme in the Limitation Act 1980 and the Land Registration Act 2002. I cannot therefore accept the alternative way of putting MR2’s case. In my judgment Mr Farakh Rashid is entitled to be registered as proprietor by virtue of paragraph 18 (3) of Schedule 12 to the 2002 Act. That amounts to exceptional circumstances which preclude an alteration of the register.

Ms Tozer submitted that whether there were exceptional circumstances was a value judgment for the FTT and the Upper Tribunal with which we should not interfere unless the tribunals had gone wrong in principle. However, for the reasons I have given I consider that they did. In the FTT the judge said that it would be manifestly unjust for MR2 to be deprived of his property as a result of the forgery. In the Upper Tribunal Judge Cooke held that Mr Farakh Rashid cannot be allowed to plead his own fraud. In both cases the tribunal over-concentrated on the Land Registration Acts; and failed to appreciate the importance of limitation periods generally. They also failed to appreciate that the Limitation Act itself contains limitation periods specifically applicable to the recovery of property acquired as a result of crime. In my judgment we are entitled to interfere with those value judgments.

It may be thought to be an unattractive result to leave Mr Farakh Rashid in possession when the origin of his title depends on fraud. However, as Lord Hoffmann NPJ observed in Peconic Industrial Development Ltd v Lau Kwok Fai [2009] 5 HKC 135 (cited with approval in Williams at [28]):

“The principle is not that the limitation defence is denied to people who were dishonest. It plainly applies to claims based on ordinary common law fraud. The principle is that the limitation period is denied to fiduciaries. But dishonest assisters are not fiduciaries.”

What makes Farakh Rashid’s claim good is not his father’s fraud in procuring his registration as proprietor or his complicity in that fraud; but the fact of possession of the property for the requisite 12 years without the consent of MR2, which elapsed before the far-reaching changes made to the law of adverse possession by the 2002 Act. Although, as I have said, section 32 of the Limitation Act 1980 can extend time in cases of undiscovered fraud, it appears from the findings of fact by the FTT that MR2 knew that he had a grievance by 1990. It would not avail MR2 on the facts of this case.

I would allow the appeal.

We agree that this appeal must be allowed for the reasons given by Lewison LJ. We add this short judgment only because we are reaching a different conclusion from the tribunals and because in doing so we are departing from a decision of this court upon which they quite reasonably relied.

The central plank in the reasoning in Parshall v Hackney is that time cannot run in favour of a person in possession of land without the consent of the true owner if he also happens to be the registered owner. For the reasons explained by Lewison LJ, that plank is unsound. There are certain people in whose favour time cannot run: true trustees (because of s.21(1) of the 1980 Act) and tenants and licensees (because they have the owner’s consent). Ms Tozer relies on Parshall to expand the class to include registered owners. But, in addition to the insuperable difficulty this faces as a matter of statutory construction, there is no policy logic in this expansion. Given that time undoubtedly runs in favour of a scoundrel who seizes land, fences it off and squats on it, it is difficult to see why it should not run in favour of a scoundrel who does all that and then for good measure forges a transfer document, when the last step gives him only bare legal ownership that can readily be defeated by proof of fraud. A further difficulty is that if the submission were correct, there could be no limitation period at all in such cases. In principle, the estate of MR2 would be entitled to rectification not only now, after 30 years, but after 50 or 100 years or more.

Turning to the plea of illegality, it is impossible not to feel sympathy for the victim of a brazen fraud of this kind and to wish to deprive a scoundrel such as this appellant of the benefit. However, correctly analysed, the outcome in this case is not the result of the fraud but of the true owner’s subsequent failure to take effective action to challenge it. Had MR2 sought rectification at any time between 1989 and 2001, there is no reason why he could not have succeeded. But he did not apply until 2013 and by then it was too late. Although the result is undoubtedly a hard one, that is not because MR2 did not have full legal rights to face down the fraud but because he did not exercise them in the 12 years during which they were available to him.

We would accept Ms Tozer’s submission that a litigant faced with a plea of limitation is not as a matter of principle prevented from raising a plea of illegality and from inviting the court to apply the analysis framed by Patel v Mirza. As set out at [120] the question is whether it would be harmful to the integrity of the legal system to enforce a claim and in seeking an answer consideration must be given to (a) whether the purpose of the prohibition which has been transgressed will be enhanced by denial of the claim, (b) any other relevant public policy on which the denial of the claim may have an impact and (c) whether denial of the claim would be a proportionate response to the illegality, bearing in mind that punishment is a matter for the criminal courts. However, in a limitation case the scope for the doctrine of illegality is greatly reduced by the fact that the whole point of limitation is to limit the time for bringing claims, including good claims. As Lewison LJ says, the Limitation Act is calibrated for this purpose. As part of the calibration, section 4 provides a special time limit in the case of the theft of a chattel, sections 15-19 and Schedule 1 contain provisions concerning the recovery of land and rent, section 21 concerns trust property, s.27A concerns actions for the civil recovery of proceeds of unlawful conduct, and s.32 postpones the limitation period in cases of undiscovered fraud, concealment or mistake. This careful statutory framework, described by Sales LJ in Best at [73] as “a delicate and comprehensive balance of interests” will in almost every case incorporate all the considerations mentioned in Patel v Mirza. Like Sales LJ at [67], we would leave open whether extreme examples of illegality not encompassed within the structure of the Limitation Act might allow a plea of illegality to succeed, but we are clear that this is not such a case.