I. Introduction
B acquires A's asset without A's consent, and sells it to C, an innocent purchaser. Can A recover the asset, or is C entitled to keep it? Choosing between A and C is notoriously difficult: both are innocent parties. Nevertheless, any legal system must confront the dilemma and choose its winner, leaving the losing party to seek indemnification – often unfruitfully – from those responsible for his loss. There is no logically correct answer, and any legal system must ultimately decide, on policy grounds, what the outcome should be, and what shape its rules should take.
This article examines one particular manifestation of this dilemma in English law – where the asset that passes down the A-B-C chain is a registered estate in land. The prevailing – and “orthodox” – understanding of the law's solution to this problem is essentially “formalist” in nature.Footnote 1 It has two key features: (i) that the solution is to be found – and found exclusively – within the four corners of the Land Registration Act 2002,Footnote 2 which operates as a predominantly self-contained legal code; and (ii) that the outcome prescribed by the Act's terms emphatically favours C – a result that contrasts dramatically with the conclusion general property law would reach, which prefers A. In essence, this orthodox view sees the registered land regime as a closed system of “title by registration”,Footnote 3 in which titles – like A's, B's and then C's – derive from the Register itself, rather than from the external rules of general property law.
This orthodox vision was formed when the LRA 2002 first came into force. In the intervening decade, A-B-C disputes have spawned a wealth of case law, as yet relatively uncharted,Footnote 4 making the time ripe for questioning whether the orthodox vision has been realised in practice. The clear picture that emerges from close analysis of these numerous decisions is that it has not. Rather than buying into the “title by registration” principle, which favours C, judges are tending towards outcomes that prefer A, and which therefore chime with general property law. This article explores this remarkable trend, and argues that it reflects both a deep-rooted judicial commitment to the fundamental values that inhere in the general law, and a corresponding scepticism for the socio-economic choices underpinning the orthodox view of the LRA 2002. These observations, in turn, raise serious doubts about the viability of self-contained formalist legal codes, not only with respect to registered land, but also across the law more generally. Scotland has recently abandoned a strict theory of “title by registration”, in favour of a land registration system that is more openly rooted in general property law principles.Footnote 5 Arguably the time has also come for England and Wales, ten years the wiser since the LRA 2002 came into force, to follow Scotland's lead. This would involve accepting that the intrinsic judicial preference for giving A the land over C means that the formalist “title by registration” principle might never be anything more than a mythical fantasy.
II. Setting the Scene for the A-B-C Problem
A full understanding of the registered land system's resolution of the A-B-C problem requires some appreciation of the typical contexts in which A-C disputes arise. In practice, such disputes arise out of circumstances that can be reduced to three core elements:
(i) A begins as the registered interest-holder;
(ii) B's registration is procured without A's consent, and is therefore mistaken; and
(iii) B, acting on the strength of his erroneously-registered title, disposes of an interest to C, whose interest is subsequently registered.
Two important variables within this core structure determine the precise factual configuration of any particular dispute.
A. Reasons Why B's Registration is Mistaken
First, there are many reasons why B might acquire a registered interest in A's land by mistake, and without A's consent. Most commonly, there is a purported disposition from A to B which, despite appearing valid on its surface, is actually void.
A key example is where A's signature on the dispositive document is forged by B or another.Footnote 6 A is most vulnerable to forgery by someone in his immediate entourage, who has easy access to his signature.Footnote 7 However, proprietors also risk forgery by strangers. Indeed, until recently, the forger's task was greatly facilitated by the Land Registry's overly-zealous commitment to principles of accessibility: proprietors' signatures regularly appeared on the publicly downloadable Register. This practice has now ceased,Footnote 8 but determined fraudsters will inevitably strive to find new and more ingenious ways of effecting forgeries.Footnote 9
Void dispositions can also occur for reasons other than forgery. The case law is regrettably rich with examples, such as purported dispositions made by deeds which are not properly attested,Footnote 10 by directors of companies in the process of being wound up,Footnote 11 or by disponors who fail to comprehend the documents they sign.Footnote 12 In each case, the crucial significance of the A-B disposition being void is that there is no legal basis for B's registration as proprietor.
Substantially equivalent to void disposition cases are “no disposition” cases, where there has never been a purported disposition from A to B of the relevant land – void or otherwise. Here too, B's registration lacks a legal basis, and is mistaken. Such cases take two main forms. First, A may have validly disposed of part of his land to B, but B is mistakenly registered as the proprietor of the whole.Footnote 13 In relation to that part which A did not convey, there was no disposition, and therefore no legal basis for B's registration. Second are situations where a squatter (B) successfully applies to take over the registered title of another's (A's) registered estate, without having satisfied the necessary statutory preconditions.Footnote 14 Here too, the underlying, statutory basis of B's registration is void.Footnote 15
Wherever there is no legal basis for B's entitlement, the Registrar is neither required, nor authorised, to register B with an estate or interest in the land. However, notwithstanding this rule, some void dispositions, or their equivalent, inevitably go undetected and result in registration. The incidence of Registry error is currently low – fewer than 1% of registrations contain errorsFootnote 16 – but it is important, for ensuring public confidence in the Register, to have clear rules to deal with such errors when they arise.Footnote 17
Different concerns necessarily arise where a disposition from A to B is voidable rather than void, as where A is induced to transfer an interest by fraud or misrepresentation. Here, the disposition is valid unless and until avoided by A, and, in the meantime, fully authorises the registration of B's interest. Being situations where B's registration has a legal basis, albeit temporarily, they are necessarily outside the scope of this article.Footnote 18
B. Nature of Interests Being Disposed
The second variable within the A-B-C pattern concerns the types of interest which B and C purportedly acquire. To keep the analysis within sensible limits, it will be assumed that A begins with a registered freehold estate.Footnote 19 The simplest configuration of the A-B-C dispute finds B, and subsequently C, registered as proprietor of the same freehold estate.Footnote 20 However, land law's rich palette of registrable estates and interests yields multiple variations on this theme.
For example, B may become the registered freeholder, and then C a registered chargee.Footnote 21 This scenario arose on the facts, as argued, of the much-litigated Barclays Bank v Guy dispute.Footnote 22 A was the registered freehold proprietor of a valuable development plot. The title was subsequently registered in B's name, pursuant to a disposition that A alleged had been forged. B subsequently registered a charge against the freehold as security for a multi-million pound loan from C, a bank. Upon discovering these events, A sought to recover his registered freehold in its previous state, unencumbered by C's charge.Footnote 23
In some cases, the fact pattern is reversed: B acquires a registered charge, and C the registered freehold. In Odogwu v Vastguide Ltd.,Footnote 24 for example, A owned the registered freehold of a house, as an investment property. Using a forged version of A's passport, a fraudster obtained a loan from B, a finance company, secured by way of a registered charge over A's freehold. Subsequently, B sold the freehold to C, pursuant to its power of sale as chargee, leaving aggrieved A to seek recovery of his registered title from the innocent purchaser.Footnote 25
Many other permutations of the A-B-C dispute are, of course, possible.Footnote 26
C. The Concern for “Mud”, not “Money”
Whatever the precise factual configuration of the A-B-C dispute, the basic legal question is the same: should A be returned to the position he enjoyed before the unfortunate catalogue of events, or should the law uphold the interest that C acquired in reliance on B's registered title? This article's concern is with which party gets to keep his interest in the land, as opposed to being indemnified by money. Whilst compensation is sometimes an adequate remedy for the infringement of rights in property, the law has long recognised that land, as opposed to other types of property, is different. Being a unique and finite resource, land is usually best protected via in specie remedies.Footnote 27 Thus, to use a metaphor commonly employed by property lawyers,Footnote 28 this article assumes that landowners would rather keep their “mud”, to which they may have formed a particular attachment, than have substitute “money”.
III. Legal Solutions to the A-B-C Problem
When allocating the “mud”, every legal system must inevitably choose between A and C. It must also do so on policy grounds, not logic, taking account of the various socio-economic objectives that might be served by favouring one party over the other. The orthodox view of the LRA 2002, analysed later in this section, is immediately striking because its solution is diametrically opposed to that which general property law, and indeed land law prior to the LRA 2002, has offered.
A. The Position Prior to the LRA 2002: Preferring A
General property law's solution (i.e. the law “both statutory and of judicial creation, which exists independently of [land registration legislation]”Footnote 29) is straightforward, and clearly prefers A. Technically, A retains his original title: the absence of a valid disposition from A to B means that B acquires no interest in the land; and neither does C, because the validity of C's interest depends on B's title being valid. C's only hope here is to pursue a claim for compensation against B for breach of contract.
These general law rules apply to land which is unregistered.Footnote 30 For registered land, A would also have been able to recover the land prior to the LRA 2002, via the rectification provisions of the Land Registration Act 1925, which were interpreted as permitting the Register to be altered when it was out of kilter with the result that would have pertained had the land been unregistered.Footnote 31 In contrast to the unregistered land position, C could apply for an indemnity from the Registry.Footnote 32
Favouring A by restoring the status quo ante involves a preference for so-called “static security” of title over “dynamic security”Footnote 33 – the original owner is preferred to subsequent purchasers. There are many good reasons for adopting such an approach.
First, allowing C to win seems obviously objectionable from the perspective of A who, ex hypothesi, did not authorise the disposition to B. For A to be divested of his property without his consent seems anathema to democratic ideals of private ownership.Footnote 34 Indeed, A's property rights in the land (and, if applicable, his use of the land as his home) are protected under the European Convention on Human Rights,Footnote 35 such that any non-consensual interference therewith is unjustified unless proven to be in the public interest and proportionate.Footnote 36
Secondly, from a more utilitarian perspective, security of title is arguably enhanced generally, for everyone, by favouring A. As the Canadian author Thomas Mapp observed, favouring C over A would create a system of “easy come, easy go”.Footnote 37 No title would be secure, not even C's, for, whilst C would be secure against A, his security might be short-lived: his own title would be vulnerable to void dispositions in the future.Footnote 38
Thirdly, research from the Scottish Law Commission has revealed that the layperson's instinct is to unravel void dispositions, thereby preferring A over C.Footnote 39 The popular conception of a just outcome should not be dismissed lightly.
B. The Orthodox View of the LRA 2002: Preferring C (But Not Necessarily B)
Notwithstanding the good reasons that may exist for protecting A, the predominant, although not universal, interpretation of the LRA 2002, at the time of its enactment, was that it heralded a reversal of A's and C's fortunes, and preferred C.Footnote 40 Amongst those championing this “orthodox” view at this time were Elizabeth Cooke, Roger Smith and the authors of the influential practitioner guide to land registration – Ruoff and Roper's Registered Conveyancing.Footnote 41 The orthodox – essentially formalist – view regards the statute as embodying a political preference for dynamic security of title,Footnote 42 and holds that “reference back to unregistered land” is unjustified: “we should recognise registered land principles as a self-contained, considered and appropriate resolution of problems which arise”.Footnote 43 As explained in the next Part, this preference for C is achieved by deeming that B acquires a “title by registration”, which he can validly transfer to C.
There are good policy reasons why the legislature might have chosen to render C's title indefeasible vis-à-vis A. First, this choice undoubtedly simplifies and cheapens the conveyancing process for C, the purchaser. To be confident that he is acquiring a valid interest, C can take the Register at face value, and need not undergo the cumbersome process of establishing that B's title has valid historical roots. Secondly, at a macro level, this preference for C – who arguably represents the countless people contemplating buying land – might serve to create a more confident and buoyant property market. Finally, C's rights, like A's, may also be protected by the European Convention on Human Rights, to the extent that C has made the land his home, and/or his registered title counts as a relevant property right for Convention purposes.Footnote 44
Although this article's focus is on C's position, it is useful to appreciate that the policy arguments for preferring B, as against A, are considerably less compelling.Footnote 45 Reflecting this, B is treated less kindly by the LRA 2002. On an orthodox reading, B's title is inherently defeasible, unless he currently possesses the land, and therefore has a cogent reason to retain it.
Thus presented, the orthodox vision of the Act looks straightforward: C's title is secure against A, but B's might not be. However, this orthodox view has not been realised in practice in the cases. By various stratagems, both C's and B's titles are being rendered defeasible in a much wider range of situations than the orthodox vision would suggest, bringing the outcomes of cases more closely into line with those that general property law would produce.
IV. The Orthodox View of the Lra 2002
Before exposing the judicial retreat from orthodoxy in Part V, it is necessary to explain how the orthodox vision was originally thought to be realised via the LRA 2002's terms. A complete understanding of how the Act was thought to prefer C over A requires a detailed analysis of both key stages of the A-B-C transactional sequence: the mistaken registration of B, and the subsequent disposition to C. Each stage is governed by a combination of rules derived from two sources: the Act's main body and its Schedules. These two sets of rules appear to pull in different directions, and therefore warrant separate examination. In broad terms, the main body of the Act affords B and C titles upon registration; the Schedules then render registered titles potentially defeasible, via their provision for “alteration” of the Register. The orthodox reading holds that, whilst there is a broad jurisdiction to alter B's title, C's is generally immune from alteration, and therefore secure.
A. The Main Body of the Act: “Title by Registration”
The Act's main body, orthodoxly interpreted, undoubtedly creates a system of “title by registration”, in which it is the Register – and only the Register – that determines ownership of land. How exactly is this realised?
1. The mistaken registration of B
Turning first to B's position, it is clear that he has no rights in the land prior to his registration. Where there is a purported, but void, A-B disposition, B at most acquires a possibility of being mistakenly registered as proprietor.Footnote 46 However, his position is dramatically enhanced by registration because the Register is treated as “conclusive”. According to section 58:
[i]f, on the entry of a person in the [R]egister as the proprietor of a legal estate, the legal estate would not otherwise be vested in him, it shall be deemed to be vested in him as a result of the registration.
This provision is frequently described as effecting “statutory magic”.Footnote 47 Just as King Midas turned worthless objects into gold, the Registrar turns void interests into fully-fledged legal proprietary rights.Footnote 48 Furthermore, although section 58 itself is silent on the issue, the orthodox view is that the legal title vested by section 58 ordinarily carries with it beneficial ownership. The voidness of the A-B disposition is not regarded as a reason, in and of itself, to invoke equity's jurisdiction to split the beneficial from the legal title.Footnote 49 As such, B does not hold the land on trust for A unless a recognised trust-generating event occurs, such as B acting unconscionably.Footnote 50 Any equitable interest which A thereby obtains would not be retained from his days as legal owner; it would be newly acquired upon B becoming the registered legal owner.
It follows that the orthodox view gives the Registrar an extremely powerful Midas touch. He does not merely gold-plate a void interest by vesting the legal title in B; his alchemy transforms said interest into solid gold, by conferring beneficial ownership too. In this respect, section 58 is viewed as more powerful than the equivalent section in the former Land Registration Act 1925Footnote 51 which, according to the Court of Appeal in Malory Enterprises Ltd. v Cheshire Homes (UK) Ltd., vested merely the legal title in the registered proprietor, B, leaving beneficial ownership with A as a matter of course.Footnote 52
2. The disposition to C
Turning now to C's position, once section 58 deems B the full legal and beneficial owner of his registered interest, he can effect a valid disposition to C who, in turn, will become legal and beneficial owner of the disposed interest upon registration.
This outcome is reached not only via section 58 – which deems B the owner – but also by sections 23 to 26 of the LRA 2002 – which provide that registered proprietors, like B, have “owner's powers”, enabling them to convey a good title to a disponee. A valid disposition by B pursuant to those powers fully authorises C's registration. As such, C's title is inherently valid, and there is no need to rely on section 58 fictionally to deem him owner.Footnote 53 All is not lost for A, however, for he may qualify for a Registry indemnity.Footnote 54
B. The Schedules to the Act: the Vulnerability of Registered Titles to Alteration
The main body of the LRA 2002, however, tells only half of the story: crucially, the Act's Schedules allow for the Register to be “altered” on various grounds. Thus, the apparently indefeasible titles B and C acquire via section 58's statutory magic might in fact be defeasible, in the event of a Register alteration. More specifically, Schedule 4 allows for Register alterations on four grounds:
(a) correcting a mistake;
(b) bringing the [R]egister up to date;
(c) giving effect to any estate, right or interest excepted from the effect of registration; and
(d) removing a superfluous entry.Footnote 55
In theory, the possibility that the Register might be altered poses a very real threat to the viability of the “title by registration” principle. However, the orthodox view regards that principle as so integral to the Act's overall scheme, that any inroads made into it by Schedule 4's alteration provisions are to be narrowly construed. Thus, Schedule 4 is perceived as permitting A to seek alteration against B (and only in certain circumstances), but never against C, whose title remains absolutely secure.
This orthodox, narrow interpretation of Schedule 4's alteration provisions has two key features. First, it regards three out of the four alteration grounds – namely grounds (b), (c) and (d) – as permitting mere administrative, rather than substantive, changes to the Register. This is significant because A must rely on a substantive ground to seek alteration against either B or C in the A-B-C scenario. Essentially, substantive alterations bring about actual changes to the parties' rights in the land, for example replacing C with A as the registered freeholder. Conversely, administrative corrections serve the more modest function of aligning the Register with the legal state of affairs already recognised by the land registration system, and therefore pose no threat to the security of registered titles. For example, ground (d) – “removing a superfluous entry” – justifies the deletion of Register entries that are already redundant, such as an expired lease;Footnote 56 and ground (b) – “bringing the [R]egister up to date” – permits the registration of an overriding interest that, by definition, already encumbers a registered estate.Footnote 57
Note that the classification of ground (b) alterations as merely administrative is particularly narrow. It is axiomatic that ground (b) – bringing the Register up to date – is only relevant when the Register is out of date, and that to determine when this is the case, some form of up-to-date reference point is necessary. Rather than turning outwards to general property law as the benchmark – which would permit substantive Register changes – proponents of the orthodox view chose to look inwards, to the internal rules of the registered land system – thus limiting ground (b) changes to purely administrative corrections.
The second feature of the orthodox reading Schedule 4 is that the only substantive alteration ground – (a) “correcting a mistake” – is itself narrowly construed. The very existence of ground (a) means that Parliament contemplated the Register sometimes being wrong,Footnote 58 but neither the Act itself nor the joint preparatory reports of the Law Commission and Land Registry offer guidance as to where the correct legal position is to be found.Footnote 59 So what must the Register be compared against in order to ascertain a mistake?Footnote 60
Proponents of the orthodox view emphatically rejected general property law – which would see A restored as registered proprietor – as the relevant comparator. Key among them was Roger Smith who said it would be “danger[ous]” for courts to “accept a departure from unregistered law principles as being a ground for rectification”, that any “such reference back to unregistered land is certainly not the stance of the Law Commission” and that registered land principles should be recognised as a “self-contained, considered and appropriate resolution of problems which arise”.Footnote 61 Therefore, ground (a) was not regarded as an invitation to align the Register with general property law's outcomes; instead, “mistake” was interpreted more narrowly as meaning an error made by the Registry when the Register entry failed to “reflect the true effect of the purported disposition” pursuant to which the entry was made.Footnote 62 Hence, “mistakes” were to be ascertained merely by reference the validity of the preceding disposition.Footnote 63
To what extent does this narrow reading of Schedule 4 permit alterations of the Register in the A-B-C scenario?
1. The mistaken registration of B
B's registered title is prima facie insecure as against A. The invalidity of the A-B disposition means that B's registration is mistaken, and therefore vulnerable to alteration. Here, because the alteration would “prejudicially affect” B's title,Footnote 64 it is classified more specifically as “rectification” – a subspecies of alteration.Footnote 65
Whilst B's title is presumptively defeasible, he is, nevertheless, in a stronger position under the Act than he would be under the general law, for two reasons.
First, Schedule 4 affords a registered proprietor, like B, a defence (the “Schedule 4 defence”) against rectification claims if he is “in possession” of the relevant land, provided he has not: consented to the amendment; fraudulently or carelessly contributed to the mistake; or “it would for any other reason be unjust for the alteration not to be made”.Footnote 66
Secondly, even if B ultimately loses his registered title – the “mud” – he is entitled to monetary indemnification by the Registry for loss caused by the rectification, provided it did not follow from his own fraud or carelessness.Footnote 67 Note that, conversely, in the event that B successfully defends A's rectification claim and keeps the “mud”, A will be indemnified.
2. The disposition to C
Whilst B's title is potentially defeasible, the orthodox view regards C's registered title as absolutely secure, owing to its narrow interpretation of “mistake”, which sees C's registration as mistaken only where there is some flaw in the immediately preceding disposition. That is clearly not the case in the A-B-C scenario, where B, as the deemed owner of his interest, is fully empowered to effect a valid disposition to C.Footnote 68
Note that it is only when the Register is compared to the outcomes reached by general property law that C's registration looks mistaken. This type of comparison is expressly ruled out by proponents of the orthodox view.
C. The Orthodox Vision's Broad (But Non-Absolute) Commitment to “Title by Registration”
To what extent does this picture of the LRA 2002 realise the Law Commission's and Land Registry's objective of creating a system of “title by registration”, whereby the Register, and not general property law, is the source of titles in land?
From C's perspective, the “title by registration” principle rings absolutely true: the title C acquires by registration is fully secure, such that he cannot, in any circumstances, be divested of his “mud”. The flaw in the A-B disposition is an historic irrelevance for C, whose title derives from the Register – and only the Register. Crucially, it is immaterial that general property law would regard his title as invalid.
Even though B's position is not the primary focus of this article, it is instructive to note that the “title by registration” principle also operates in A-B disputes, although to a lesser degree. To understand why, it is necessary to distinguish between the specific protection of a registered title as “mud”, and compensatory protection via “money”.
Because B's title is potentially defeasible as against A, the LRA 2002 makes no promise that B can keep his “mud”. Thus, the ‘title by registration’ principle which is assumed to be enshrined in the LRA 2002 should be understood as being designed, as a matter of policy, not to be absolute – in the sense of allowing each and every registered proprietor to keep their mud. Rather, it aims to achieve a more subtle and commercially significant objective of producing a Register that potential purchasers Footnote 69 can rely on as being correct when deciding to make their purchase. C can absolutely rely on the fact of B's registration at the time of purchase. Similarly, in A-B disputes, B, as a potential purchaser, can absolutely rely on the validity of A's registered title: the fact that B's title is potentially defeasible is not caused by the Register being unreliable at the time of purchase, but by the invalidity of the A-B disposition – a problem which the LRA 2002 was not primarily designed to cure.Footnote 70 The fact that B might keep the “mud” if in possession should therefore be regarded as anomalous, and justified by a wholly different policy concern of land law – namely, that those in occupation should not generally be uprooted.Footnote 71
B's position looks rather different when it comes to monetary protection. Here, although B might lose the “mud”, the fact that he acquires a real – and therefore valuable – “title by registration” affords him the opportunity to seek an indemnity from the Registry, for the “loss” caused when that title is removed from the Register.Footnote 72
D. The “Formalist” Nature of the Orthodox Vision
It remains to observe that the orthodox view involves an essentially formalist reading of the statute. “Formalism” as a concept is ill-defined in the academic literature,Footnote 73 but essentially has two related facets.Footnote 74 First, it tends towards viewing the law (or any given part of the law) as a closed system of logic, comprising an autonomous and self-contained system of rules.Footnote 75 Secondly, formalism requires that judges follow these rules rigidly and mechanically, and screen off from their decision any other substantive factors they might otherwise consider relevant to the outcome, such as political, moral or social considerations.Footnote 76 Although this formalist approach to the law might be criticised as unduly inflexible, this rigidity serves some important purposes: it means that outcomes in cases are predictable and consistent; and also – because judges must apply the law mechanically, without evaluating the competing policy arguments themselves – that decisions are reached efficiently.Footnote 77
How does the orthodox interpretation of the LRA 2002 conform with the formalist model? In broad terms, it interprets the land registration regime as a sovereign and internally coherent system,Footnote 78 and minimises the extent to which judges might rely on external factors when reaching their decision. For example, judges are required to take a leap of faith and believe in the statutory magic that makes C the owner, even if their better instinct (whether informed by their own morals, political preferences, or by the tenets of general property law) might tell them to prefer A. Furthermore, where the statute itself refers to external principles,Footnote 79 the orthodox view interprets these references as restrictively as possible, so as to minimise the disruption they might otherwise cause to the regime's autonomy. Take, for example, the alteration ground of “bringing the [R]egister up to date”. The orthodox view sees this ground as limited to correcting administrative errors in the Register, and emphatically not as a general licence for judges to synchronise the Register with the rules of general property law – an interpretation which would destroy the registration system's credibility as an autonomous system. A further example is found in the alteration ground of “correcting a mistake”, which logically requires a comparison of the incorrect Register with an external benchmark. Here, too, the orthodox view rejects general property law as the relevant comparator, preferring to interpret the alteration ground more narrowly (i.e. by reference to the validity of the disposition immediately preceding the disputed registration), so as to minimise the disruption caused to the otherwise self-sufficient registration regime.
Thus interpreted, the statute is regarded as embodying a policy choice, made by Parliament, that C should be favoured over A in all circumstances, so as to ensure that potential purchasers can securely rely on the Register as being correct. If this is indeed Parliament's political preference, the formalist interpretation is self-evidently valuable, because it forces judges to implement that chosen policy even if, on the facts, they may have an instinctive preference for A. For example, in Stewart v Lancashire Mortgage Corporation Ltd., the formalist approach compelled the Adjudicator to HM Land Registry to overcome his “reluctance” to find for C as against A in an A-B-C dispute, on the basis that the LRA 2002 represented “a balance struck by Parliament which preferred certainty of title over the property rights of those who had been the victims of fraud”.Footnote 80
V. Unearthing a Judicial Retreat from the Orthodox View
The orthodox, formalist reading of the LRA 2002 – which prefers C over A – seems beguilingly correct, and aligns with the Law Commission's and Land Registry's reform objective of creating a system of “title by registration”. However, it is becoming increasingly clear that formalist cases such as Stewart, discussed above, represent the high-point for the orthodox approach. As this Part reveals, judges are increasingly retreating from the formalist reading of the LRA 2002, in favour of outcomes that chime with general property law. This trend is of fundamental significance, for it strikes at the heart of the “title by registration” principle, and undermines the internal coherence of the LRA 2002.
Many cases decided under the LRA 2002 are well-reported; however, some decisions of the Adjudicator to HM Land Registry are buried in the unwieldy archived Tribunals Service website,Footnote 81 and are therefore under-examined. This has led to strikingly little cross-fertilisation between cases and, consequently, several independent strands of reasoning have developed. Indeed, there are at least eight different manifestations of the judicial retreat from the orthodox view, which themselves cluster around three broad categories. The first three involve undermining C's registered title by interpreting the LRA 2002's alteration provisions expansively; the next three involve diluting section 58's statutory magic – a direct attack on the “title by registration” principle; and the final two belong in a miscellaneous category.
A. Expansively Interpreting the Statute's Provisions for Altering the Register
The first sign of the judicial retreat from orthodoxy is a tendency to give the alteration ground of “correcting a mistake” a much broader spin than orthodoxy would permit, resulting in C's registered title being destabilised. Many cases have seen C deregistered in this way, but the precise manner in which the meaning of “correcting a mistake” is enlarged differs from case to case. Properly understood, the judges' expansion techniques fall into three distinct categories, depending on whether they: (1) broaden the circumstances in which A might make an alteration claim; (2) weaken the current registered owner's ability to defend A's alteration claim; or (3) permit A's alteration claim against B to be enforced against C.
1. Broadly interpreting “correcting a mistake”
Within the first category, the courts have found no fewer than six different ways of potentially manipulating the meaning of “correcting a mistake” within Schedule 4 to provide A with an alteration claim against C.Footnote 82
(i) Alteration requiring correction of the consequences of the mistaken registration of B
First, some cases have accepted the orthodox view of “mistake” – that B's, but not C's, registration is mistaken – but decide that “correcting” the mistake requires eradicating not only the original mistake, but also the consequences of that mistake. Thus, in Ajibade v Bank of Scotland Plc,Footnote 83 A's registered freehold was disposed via a forged power of attorney to B, who registered his supposed estate, and thereafter charged it to C. It was undisputed that A could recover her freehold from B, as B's registration was mistaken, but the Deputy Adjudicator also ordered that C's charge be expunged. Having been registered in consequence of the original mistake, the charge was the “fruit of a poisoned tree” which it would be “perverse” to leave on the Register.Footnote 84 It is noteworthy that, when making this decision, the Adjudicator was reassured by the fact that general property law would have reached the same outcome.
This interpretation of “correcting a mistake” is likely to produce arbitrary outcomes. The courts have decided that the “consequences” of a mistake may only be unravelled if the original mistake still persists on the Register;Footnote 85 and, if so, rectification will be denied in at least two important factual configurations of the A-B-C dispute. For example, where B acquires A's registered freehold via forgery, and then sells it to C, the original mistake (B's registration) is no longer on the Register, and so C would keep the title. The same would be true in cases like Odogwu v Vastguide, discussed above, where B is a mistakenly registered chargee who exercises his power of sale to transfer A's registered freehold to C.Footnote 86 There is no principled reason why in these cases C retains the registered title, whereas in cases configured like Ajibade, A can obtain full rectification of the Register.Footnote 87
(ii) Geographical mistakes persisting in the Property Register
Secondly, some cases arguably adopt a more generous view of “mistake” where A validly disposes of part of his land to B, but the Land Registry includes too much of A's land in B's registered title. Here, there is a mistake in the “Property Register” – that part of the Land Register which defines the geographical co-ordinates of any given title – and arguably this mistake persists once C's interest is registered over the erroneously large plot. Because that original mistake still taints the Property Register, A may apply for rectification directly against C.Footnote 88
The finding of a persisting mistake here is intuitively appealing; after all, the Property Register was undoubtedly wrong when B registered his title, and is seemingly never corrected. However, such reasoning amounts to an abandonment of the “title by registration” principle. If section 58 is taken seriously and the orthodox view followed, B becomes the legal owner of the extra slice of A's land upon registration. Section 58's statutory magic should apply equally to all three constituent parts of the Land Register – the Property Register, the Proprietorship Register (which denotes the current owner), and the Charges Register (which details any encumbrances).Footnote 89 Accordingly, the mistake in the Property Register made at the time B was registered should be cured by registration, thereby empowering B to convey the extra slice of land to C, whose registered title is error-free.
Although a geographical property error (as opposed to a proprietorship error) in the Register somehow feels as though it continues through successive owners, the logic of section 58 means that it does not. Any signs of a judicial instinct for finding a mistake in such circumstances once again betray a preference for general property law, and a rejection of “title by registration” and the orthodox view of the Act.Footnote 90
(iii) Registration of C being “part and parcel” of B's mistaken registration
Thirdly, in Barclays Bank plc v Guy (No 2),Footnote 91 Lord Neuberger MR suggested, in obiter dicta, that the registration of C might be regarded as “part and parcel” of B's mistaken registration.Footnote 92 Thus conceived, there is one big mistake, starting with B's erroneous registration, and encompassing all subsequent Register entries.
This interpretation of the “mistake” ground of alteration is similar to the first: both seem to regard registrations subsequent to B's as the “fruit of a poisoned tree”. However, the current interpretation is arguably broader. The first confines the mistake to the registration of B, but permits the unravelling of its consequences, whereas the “part and parcel” interpretation regards C's registration as part of the mistake itself. This difference could be significant: whereas the first interpretation seemingly permits rectification against C only where the original mistaken registration of B remains on the Register, the “part and parcel” interpretation may not be so limited – and rectification may be available whether or not B retains a registered title. For the reasons explained above, the latter interpretation is to be preferred as producing fewer arbitrary distinctions.
The “part and parcel” interpretation displays, yet again, a potential judicial willingness to hark back to general property law principles: it is difficult to regard C's registered title as tainted by a mistake without at least some reference to the result that would have pertained under general property law.
(iv) Removal of A's name from the Proprietorship Register being a mistake
Fourthly, Lord Neuberger MR in Barclays Bank v Guy (No 2) also suggested, in obiter dicta, that the relevant mistake in these cases might not be the mistaken registration of B or C, but rather the mistaken “removal of [A's] name” from the Register,Footnote 93 and that “in order to correct that mistake”, C's interest might should be removed. Thus, in Guy itself – where B mistakenly became the registered freeholder, and charged the property to C – his Lordship suggested that C's charge, as well as B's freehold, might in theory be wiped from the Register. Once again, such a result would mirror that reached by general property law principles.
At first glance, this characterisation of “mistake” seems a neat way of interpreting Schedule 4 to ensure all parties are taken back to their starting positions, without having to rely explicitly on general property law principles. However, on closer analysis, this is not necessarily the case. In Guy itself, if the relevant mistake was simply the removal of A's freehold from the Register, correction thereof would merely require that A is reinstated as the registered freeholder, which could be done without upsetting C's registered charge. Of course, the removal of C's charge could be justified as the correction of the consequences of the mistaken removal of A from the Register, but the fact that Lord Neuberger MR saw no need to explain such a result in these terms suggests he regarded it as self-evident. Again, this reasoning indicates an intuitive preference for general property law principles.
(v) Registration of C being a mistake if C knew B's registration was mistaken
Fifthly, in an earlier round of the Guy litigation, Lloyd L.J. opined that the registration of C might not itself be a mistake, “unless … [C] had either actual notice or, what amounts to the same, what is referred to as ‘Nelsonian’ or ‘blind-eye’ notice of the defect in [B's] title”.Footnote 94
This interpretation of “mistake” seems to lack any obvious rationale. On the one hand, the basic premise – that C's registration is not normally mistaken – is consistent with the orthodox view of “title by registration”. On the other hand, the idea that C is affected by “notice” is redolent of the unregistered land (i.e. general property law) principles, which ascertain whether, in equity, a disponee is bound by an earlier interest. The doctrine of notice plays no part in registered land,Footnote 95 so why introduce it here? Further, if the law is to hark back to general property law to ascertain a mistake, why resort to something resembling the equitable doctrine of notice? General property law would regard C's title as void, regardless of notice.Footnote 96 Being a peculiar mishmash of unregistered and registered land principles, this formulation of “mistake” is best forgotten, except insofar as it provides further evidence of the judges' determination to find reasons to deregister C.Footnote 97
(vi) Registration of C being itself a mistake
Finally, a number of cases have suggested, without deciding the issue, that C's registration might itself be mistaken.Footnote 98 If accepted, this argument deals a near-fatal blow to the “title by registration” principle, in favour of general property law. C's registration cannot be mistaken according to the “title by registration” logic – for B was the lawful owner when he disposed of an interest to C. It is only by re-running the events according to general property law, and by giving primacy to general property law, that there is any mismatch between what the Register reveals (that C has an interest) and the supposedly “correct” position (that C's interest is void). If accepted, this argument means that the Register cannot be believed, and the true legal position is to be found exclusively within the rules of general property law.
(vii) Indemnity
To summarise, the courts' various expansive interpretations of “correcting a mistake” mean there is usually a prima facie case for alteration against B and C – an unequivocal departure from the orthodox view, and a serious erosion of the “title by registration” principle. Although this article is concerned primarily with “mud”, not “money”, it is nonetheless useful, for a rounded view of the outcomes, to consider whether B and C would qualify for a Registry indemnity.
As discussed above, indemnity is available for the correction of a mistake, where the alteration prejudicially affects the claimant's registered title, thereby causing the claimant loss.Footnote 99 This requires asking whether the alteration makes the claimant worse off. The answer depends on whether section 58's statutory magic is real or illusory, and therefore whether B or C acquire something of value when their names appear on the Register, which is lost upon deregistration.
Most cases have chosen to believe in the statutory magic for the purposes of indemnity so that, if B/C cannot keep their “mud”, they at least get some “money”.Footnote 100 Indeed, in Knights Construction v Roberto Mac, the Adjudicator opined that denying indemnity to the losing party might breach that party's ECHR right to respect for his property.Footnote 101 However, the indemnity matter cannot be regarded as settled because, as explained below, some cases have refused to believe that registration of a void disposition confers anything beneficially of value at all.Footnote 102
2. Rendering illusory the defence against alteration
The second sign that the judges are rejecting the orthodox, formalist view concerns the availability of the Schedule 4 defence to rectification claims. It will be remembered that this defence is generally afforded to registered proprietors in possession, unless it would “be unjust for the alteration not to be made”. The orthodox view was that the defence would only fail in exceptional circumstances.Footnote 103 Therefore, even if “mistake” were interpreted generously to give A a prima facie claim against C, the defence would – more often than not – allow C to resist to rectification.
However, the defence is showing serious signs of erosion. This occurred most prominently in Baxter v Mannion Footnote 104 where a squatter, B, successfully applied to become the registered proprietor of A's land, according to the LRA 2002's adverse possession rules.Footnote 105 As it transpired, B's registration was mistaken, for he had not been in adverse possession for the requisite time period. When A sought to recover his registered title, B, despite being in possession, was denied the protection of the Schedule 4 defence. The judge, with whom the Court of Appeal agreed, said that it would be “unjust not to alter the Register” because B “was never entitled to be registered as proprietor of the field” and “simple justice require[d] that, in the absence of strong countervailing factors, [A] should now be able to regain title to his property”.Footnote 106
Crucially, this reasoning comes close to concluding that it will always be unjust not to rectify where there is a mistaken registration; and yet the statute is specifically designed to protect those who are mistakenly registered and in possession! So strong seems the judge's instinct to favour A that he effectively emptied the Schedule 4 defence of any real content.
If Baxter is followed,Footnote 107 there is no reason why its reasoning should be restricted to claims against B; it could equally deny C a defence against A's rectification claim. Thus the combination of interpreting “mistake” expansively and the defence restrictively significantly threatens C's position.
3. Holding that A's “right to alter the Register” against B may bind C
It is trite law – even for the staunchest supporters of the orthodox view – that a registered purchaser, like C, takes his title subject to overriding interests belonging to third parties.Footnote 108 The third judicial attack on the orthodox view regards A's statutory right to alter the Register against B as not merely a personal right enforceable against B, but a proprietary right, capable of overriding the disposition to C, if A is in discoverable “actual occupation” of the land.Footnote 109
Thus, in Crawley v Gudupati,Footnote 110 A, a long-suffering alcoholic, purported to transfer her leasehold flat to her local shopkeeper, B, who subsequently charged it to C, a bank. In fact, the A-B transfer was void for want of attestation. After A's death, the Deputy Adjudicator allowed her son to obtain alteration of the Register not only against B, but also against C, on the basis that A, having lived in the flat throughout, was in discoverable “actual occupation” thereof, and that her right to alter the Register against B therefore overrode C's registered charge.
This case sees the formalist view of the statute yielding to the instinctive judicial preference for A over C: not only was the nature of A's Schedule 4 alteration right elevated into a proprietary interest without any direct requirement to do so in the LRA 2002, but also the content of the alteration right morphed, without discussion, from a right merely allowing A to recover her registered leasehold from B (which would not necessitate removing C's charge from the Register), into a more expansive entitlement to recover an unencumbered registered leasehold (thereby requiring deregistration of C's charge).
To adherents of the orthodox, formalist view, such reasoning is heretical, for if the “title by registration” principle is taken seriously, there is no reason why A should retain a proprietary interest in the land once B becomes the registered owner.Footnote 111 Furthermore, such reasoning may deny the losing party – C – the right to a Registry indemnity: because C's registered title is inherently bound from the outset by A's overriding interest, C may not meet the preconditions for indemnity – that the alteration must “prejudicially affect” his title and cause him “loss”.Footnote 112 Accordingly, C, an unfortunate innocent party, may end up with neither “mud” nor “money” – a result which is anathema to the orthodox, coherent interpretation of the Act, which guarantees innocent registered proprietors protection in one form or other.
B. Diluting the Registrar's Statutory Magic
Up to this point, the judicial stratagems for favouring A over C have indirectly attacked the title C acquires by registration, by rendering it vulnerable to alteration. The next three stratagems involve a direct attack on the “title by registration” principle, by diluting the power of the Registrar, given to him via section 58, to turn non-owners into owners by registration (his so-called “Midas touch”).
1. Re-characterising cases as boundary disputes
A full appreciation of the first form of direct attack requires a preliminary appreciation that the LRA 2002 itself contains an exception to section 58's “title by registration” principle: there is no promise of Register accuracy in respect of “general boundaries”, i.e. boundaries which have not yet been precisely determined by the Registry.Footnote 113 Thus, in respect of land on such boundaries, section 58 is never engaged, and the “title by registration” promise is inapplicable.Footnote 114
Some judges have seemingly exploited – and enlarged – this exception to section 58 by giving a broader meaning to “boundary” land than might be warranted by a natural reading of the term. The cases reveal a remarkable propensity to characterise disputes between neighbouring registered landowners concerning land abutting a boundary-line as mere “boundary disputes” (thereby not engaging section 58), even if the land-mass concerned covers a fairly significant geographical area, and the disposition might more naturally be classified as a normal “property dispute” (which would engage section 58).Footnote 115 This was the case, for example, in Derbyshire County Council v Fallon, where the disputed land was a sizeable strip on which a garage had been built,Footnote 116 in Drake v Fripp, where the disputed land extended to one and a half acres,Footnote 117 and in Barwell v Skinner, where the so-called “boundary” land measured as much as 15–20% of the total area covered by the two neighbouring plots.Footnote 118 This generous approach to the meaning of “boundary” land enables judges to favour A in the A-B-C scenario in cases where: A disposes of part of his land to B; B is mistakenly registered with too much of A's land; and B subsequently disposes of his title to C. Here, if the extra land included in B's/C's registered titles is classified as “boundary”-related, it is never owned by B/C because section 58's statutory magic does not apply. Hence, because A has remained owner of the extra land throughout, the registrations of both B and C are undoubtedly mistaken – even on the orthodox interpretation of “mistake”. Consequently, A can seek alteration of the Register against C.Footnote 119 Regrettably for C, he will not qualify for an indemnity if the Register is altered because, having never owned the disputed boundary land via section 58's statutory magic, he suffers no loss when divested of his registered title.
Outwardly, this judicial method for favouring A seems to conform to the orthodox, formalist view by fitting squarely within the four corners of the LRA 2002: the Act itself concedes that, in some cases, section 58's statutory magic does not operate. However, in reality, these cases represent a significant retreat from orthodoxy. By adopting an unnaturally wide meaning of “boundary” land, they arguably go beyond the plain meaning of the statutory language, and therefore carve out a wider exception to the “title by registration” principle than a formalist interpretation would seem to permit.
2. Treating A as beneficial owner, whose right may bind C
The second judicial attack on section 58's magic is stronger and more direct. Proponents of the orthodox view saw the Registrar as a powerful King Midas who, via the act of registration, could fashion full legal and beneficial title out of void dispositions. This view has, however, been seriously doubted by recent cases, which have decided that registration of B's defective “title” only ever confers a formal, legal title on B via section 58, leaving A as the beneficial owner under a trust. Thus, whilst the Register may give the impression that B is the owner, the real value remains with A, and the principle of “title by registration” is compromised – in substance, if not in form. As such, the Registrar performs a limited gold-plating service on B's defective title, rather than full alchemy.
The most striking example of this trend is Fitzwilliam v Richall Holdings Services Ltd.Footnote 120 There, a fraudster, acting via a forged power of attorney, purported to convey A's registered freehold to B, an innocent and diligent purchaser, which subsequently registered its title. On discovering the fraud, A sought recovery of his registered title, either: (i) within Schedule 4, on the ground of a mistake; or (ii) via the non-statutory route of asserting an absolute beneficial interest in the house which would allow him – under the Saunders v Vautier ruleFootnote 121 – to call for a retransfer of the legal title into his own name.
Although argument (i) would have yielded a claim, because B's registration was clearly mistaken, it was believed that there was a possibility that it might be defeated by the Schedule 4 defence.Footnote 122 Accordingly, Newey J. found for A via argument (ii) instead, against which the defence was unavailable, holding that “[A] remained the beneficial owner of [the house] notwithstanding [B's] registration”.Footnote 123 Despite acknowledging the strength of the orthodox view militating against this finding, he deemed himself bound by the decision in Malory v Cheshire Homes,Footnote 124 where the Court of Appeal found a trust in similarFootnote 125 circumstances, under the corresponding section of the former LRA 1925 – section 69.
However, in reality, and with respect, the judge was not hemmed in by precedent: section 69, LRA 1925 is materially different in function to section 58, LRA 2002,Footnote 126 hence a departure from Malory was justifiable. The fact that Newey J. could have rejected Malory, but chose not to, is indicative of his strong propensity to favour the original owner.
The ramifications of this finding are significant. First, at a practical level, when transposed to the A-B-C scenario, A's trust interest undoubtedly poses a real threat to C's title as well as to B's. Provided A is in discoverable actual occupation of the land when B disposes of his interest to C, A's equitable interest will override C's legal title, rendering C trustee for A.Footnote 127 As such, A may seek a retransfer of the legal title into his own name.Footnote 128 Here, A's right to recall the property exists independently of Schedule 4,Footnote 129 but A may use the procedural machinery of Schedule 4 to vindicate his right if he chooses, on the ground of “bringing the [R]egister up to date” with the location of his beneficial ownership. Either way, C would not qualify for any indemnity: he suffers no loss because his title is flawed from the outset.
Secondly, from a theoretical standpoint, this trend threatens to destroy the orthodox vision of the Act. Even though section 58 itself only explicitly provides for the vesting of legal title in a registered proprietor, the orthodox view sees this legal title as carrying with it beneficial ownership, unless and until there is a trust-generating event, recognised by the general law, that justifies splitting legal from beneficial ownership. It is trite law that trusts only arise in certain prescribed circumstances. Unfortunately, there is no examination in Fitzwilliam of the nature of the trust generated, and the answer is far from clear.
There was evidently no express trust. It is not obvious that a constructive trust could be found either: according to conventional reasoning, constructive trusts typically respond to the trustee's unconscionable conduct,Footnote 130 and, on the facts, B was innocent.Footnote 131 A resulting trust explanation is similarly improbable: these trusts are normally generated in circumstances where it is the former legal owner (A), rather than a third party fraudster, who effects the transfer of legal title to the purported trustee (B), which was not the case in Fitzwilliam. Robert Chambers, however, has argued – with some support from the cases – that resulting trusts also arise where A is wholly ignorant of the transfer.Footnote 132 Thus, only if Chambers' view of resulting trusts were accepted – which is doubtful – would proponents of the orthodox view be satisfied with the finding of the trust in Fitzwilliam.
The fact that Newey J. was prepared to overlook the difficult issue as to the nature of the trust suggests that the decision was not reached via a formal, orthodox reading of the statute; it was instead driven by the judge's instinct that the original owner should recover his land.
3. Treating A as legal and beneficial owner throughout
Finally, and more remarkably, there are very tentative signs that some judges are wholly ignoring section 58's statutory magic, and regarding A as the continuing owner of the property in every respect – both legally and beneficially. Such signs are evident in those cases, discussed above,Footnote 133 which assume, without necessarily discussing the point, that C's registration is itself mistaken. This conclusion presupposes a mismatch between what the Register says – that C is legal owner – and the true legal position. The assumption may be that legal ownership remains in A.Footnote 134
How technically might A's retention of a legal title be explained? On one view, A has the only legal title to the land, and B/C acquire nothing upon registration. Here, section 58 is stripped of all its magic, and the “title by registration” principle is wholly abandoned. Another view relies on the doctrine of relativity of title: arguably B acquires a legal title via section 58's statutory magic, which he subsequently transfers to C, but A retains throughout a superior legal title. This latter view is equally damaging to the “title by registration” principle, in substance, if not in form. Whilst B/C seem to acquire a “title by registration”, that title, being secondary to A's, is effectively worthless.
Both views represent a serious departure from the “title by registration” principle. On the first view, this is because registration gives B/C no rights whatsoever; on the second, registration confers a formal legal title on B/C, but this title is rendered substantively valueless by A's persisting superior rights. Furthermore, in the event of A recovering his registered title from B/C, no indemnity would be payable, because their titles are inherently flawed from the outset.
C. Miscellaneous
Finally, there are two further indications of a judicial preference for general property rules at the expense of the orthodox preference for upholding the ‘title by registration’ principle. These belong in a miscellaneous category.
1. Being content for the Register to remain out of kilter with actual rights
First, in Derbyshire C.C. v Fallon,Footnote 135 the court recognised that the Register was indeed out of line with the true legal position of a boundary, but was nonetheless content for the Register to remain incorrect. The court's refusal to correct the Register, when given the opportunity to do so, reveals a striking disregard for the principle, inherent in the orthodox view, that the Register should be an accurate source of rights for potential purchasers.
2. Insights from another area of land registration
Finally, useful insights are obtained by examining judicial attitudes to another register kept by the Land Registry – the “Day List”. By way of necessary background, the Day List records the date on which applications for registration are made, and is important because interests that are ultimately registered are dated by reference to this application date.Footnote 136
In Chief Land Registrar v Franks,Footnote 137 the Court of Appeal faced a Day List dispute which broadly corresponds to the A-B-C configuration. A had successfully adversely possessed part of B's land (the disputed land), and applied to be registered as proprietor. A's application was duly noted on the Day List, but subsequently was mistakenly removed. Thereafter, B charged its registered title (including the disputed land) to C, which would have assumed from the Day List that no registrations were pending over the property. A then sought to restore his title application to the Day List, as if it had never been cancelled, with the possible effect of binding C.
Neither the LRA 2002 nor the associated Land Registration Rules deal with Day List errors. Therefore, the Court of Appeal had to decide the dispute from first principles. The Chief Land Registrar argued for B/C on familiar policy grounds – that the retrospective reinstatement of A's interest would “undermine the registration of title to land and the protection it gives to third parties who acquire their interests in reliance on what their search discloses”.Footnote 138 Crucially, however, the court decided, by a majority, to restore A's Day List record as against B, leaving open whether A's interest would also bind C, because C had chosen not to oppose the alteration.Footnote 139
Here, without the guidance of a statutory framework, the judges had to appeal to their bare judicial instinct. It is telling that, faced with a choice between A and B, they favoured A, even though that might compromise the reliability of the Day List for third parties like C.
VI. Conclusions
It is clear from the preceding analysis that the judiciary is steadily derailing the orthodox, formalist interpretation of the statute, and the notion that the LRA 2002 operates a system of “title by registration”, in favour of outcomes that align with general property law. The unearthing of this judicial trend gives rise to four crucial insights concerning the present state of the law which, in turn, beg the question as to what path the land registration system should follow in the future.
A. The Current – Unsustainable – Position
First, from A's and C's perspectives, the present state of the law means that the outcome of A-B-C disputes is regrettably unpredictable. When allocating the “mud”, a few rare cases still cling to the orthodox view that favours C, but many others find for A, via one or other of the judicial stratagems revealed above. This level of uncertainty as to the allocation of property rights is unacceptable, and unsustainable. To compound this problem of uncertainty, it is further unclear whether, in the event that C loses the “mud”, he might qualify for a Registry indemnity. It is axiomatic that an indemnity is only available where deregistration would cause C “loss”, which is potentially the case if stratagem A.1. above is followed, but not if the judge opts for stratagems A.3., B.1., B.2. or B.3., where C's title is regarded as inherently burdened from the outset by A's superior claim.Footnote 140 Furthermore, to the extent that C loses both the “mud” and the “money”, he may argue that his rights under the ECHR to have his property (and, if applicable, his home) protected have been violated.
Secondly, and closely related to the first point, the judicial rejection of the orthodox view dramatically changes the function and reliability of the Land Register itself. No longer can the Register properly be regarded as a source of rights in land; arguably, it is becoming akin to a mere bureaucratic record of rights and – as seen in stratagem C.1. above – in some cases not even an accurate record, rendering the Register (which is, of course, expensive to run) of increasingly questionable legal value.
Thirdly, why are so many judges rejecting the orthodox, formalist view of the LRA 2002 in practice? There are two possible – and interrelated – explanations for their instinctive preference for returning the “mud” to A. The first is that, whilst the “title by registration” principle might be attractive to policy-makers at an abstract level, owing to the benefits it would bring to the property market, these wide-scale benefits might be insufficiently manifest to the judges hearing concrete A-C disputes, for whom justice inter partes is more important. Secondly, the judges' impulsive preference for A inter partes suggest a strong, if unarticulated, commitment to the notion – inherent in the rules of general property law – that owners, like A, should not normally be deprived of their titles without their consent. This idea that proprietors, especially those occupying the land as their home, should be protected against non-consensual dealings is deeply rooted in the legal order, and strongly bolstered by human rights jurisprudence.Footnote 141
Fourthly, the judicial rejection of the LRA 2002 as a self-contained, formal legal code raises fundamental questions concerning the viability of formal codes more generally, especially when they seek to implement policies and achieve outcomes that run counter to deep-rooted legal principles and judicial instincts. The cases analysed above suggest that it will be at least extremely difficult to impose an artificial, formalist legal order in such circumstances: it is perhaps inevitable that judges find means for escaping closed structures, in favour of results that align with their instincts. The LRA 2002, however, perhaps makes this escape too easy. The legislature's failure both to define the meaning of “mistakes” that require correction in the Register, and explicitly to provide that both legal and beneficial ownership should vest in registered proprietors, offers ready-made escape-routes for judges intent on finding for A. The crucial lesson to be learned is that in order for closed, artificial legislative codes to succeed in practice, they must be framed in wholly unambiguous, watertight terms. Even the most gifted draftsman would struggle to achieve such perfection.
B. What is the Way Forward for the Land Registration System?
Turning back to the specific concerns of land registration, it is absolutely clear that the path that proponents of the orthodox view thought the law would and should take – which adheres to the “title by registration” principle – is not in fact always being followed by the judges, many of whom are following their judicial instinct instead. As the Act enters its second decade, this aimless meandering cannot be allowed to continue. Indeed, this view now seems to be shared by the Law Commission, which suggested in July 2013 that a review of the provisions on fraud and the title guarantee might be included in its next programme of law reform.Footnote 142 So which direction should the law take in the future? The chosen path must not only be clearly signed, and workable in practice; it must also – most fundamentally – be designed to achieve sound policy objectives. The briefest of glances across to other jurisdictions' registration systems reveals a myriad of options, but all ultimately converge into three broad routes forward.
1. Retaining and fortifying the “title by registration” principle
First, English law might choose to retain and reinforce the LRA 2002's “title by registration” principle, by amending the statute to close the current judicial escape-routes. One obstacle to this route, as the last decade's experience has revealed, is that, however carefully the gaps are plugged, there remains a real danger that judges, driven by their instinct for justice in individual cases, will drill new holes into the legislation, and thwart its scheme. Even if judicial adherence to the “title by registration” principle could be ensured, the principle arguably puts the pursuit of abstract logic – producing a perfect register – ahead of more urgent policy concerns. It is by no means clear, for example, that the results achieved by the principle – viz. retention of the land by C as against A in all A-B-C disputes – are always socially desirable, or indeed consistent with the tenor of human rights jurisprudence that has developed over the last decade: the latter may require a more subtle degree of balancing between the interests of A and C in the land. Arguably, the time has come, now that “title by registration” has been tried, tested and seemingly failed, for English land law to abandon the principle.
2. Reverting to general property law principles
A second option is for English law to determine the outcome in A-B-C cases exclusively by reference to general property law, which would see the land restored to A in every case. Whilst this option might be more aligned with judicial instincts and less artificial than the “title by registration” principle, it too is riddled with problems, and should be rejected. First, in such a scheme the Register would become a mere record of – or gloss on – the underlying general property law scheme.Footnote 143 Why maintain an expensive land Register if purchasers like C – who are, in part, funding the Registry via their registration fees – cannot rely on it to acquire the land they have chosen to buy?Footnote 144 Secondly, this option, by always favouring A, suffers from the same level of inflexibility as the “title by registration” system. Further, the constant preference for A would ultimately have an adverse impact on buyers' confidence in the property market.
3. A middle way: integrating registration principles with general property law
The third – and preferable – option is to devise a system that integrates the best parts, and avoids the pitfalls, of both the “title by registration” and general property law schemes. This might be achieved by (i) abandoning the artificial, formalist notion that title stems from the Register, and instead rooting title openly in the rules of general property law, whilst also (ii) recognising that the rules of general property law might themselves be modified for registered land cases, to respond to the unique policy concerns raised by the registration system.
Following this route, the rules of general property law would, by default, give title to A in the A-B-C scenario, but be tweaked for registered land cases, to give C the title in certain exceptional circumstances. Neither A nor C would then be favoured in any blanket, abstract fashion; rather the legislature would have the opportunity to define precisely when C should win based on carefully worked-through and openly-discussed policy reasons.
Scotland has recently chosen this path. Legislation there in force now provides that C – exceptionally – should acquire title after he and/or B has possessed the land for a combined period of one year.Footnote 145 English land law should follow this lead by abandoning ‘title by registration’, and having an open discussion as to how, instead, general property law might itself be moulded to realise the particular policy ambitions of English land law.
