Introduction
When in 1887 Lord Halsbury proposed a statutory provision that conferred a judicial power to alter the land register, even to the prejudice of otherwise secure registered property rights, it laid the foundation for a legislative scheme for alteration that has remained controversial ever since.Footnote 1 The provision passed through various incarnationsFootnote 2 until reaching its current expression in which the discretionary power to correct is triggered by the key event of “mistake”.Footnote 3 It is this concept which now takes first place in determining whether the land register is susceptible to correction.
As the criterion which controls the power to correct the land register, it regulates the extent of fallibility in registered titles, with recent case law even going as far as challenging the title of an innocent purchaser for value who became registered as proprietor. If used in this way, the concept of mistake would curtail the characteristics which are normally regarded as uniquely indicative of property rights: security, durability, persistence. Mistake has consequently become important in giving effect to the contemporary values which justify and constrain property rights in registered land. Furthermore, because mistake describes the extent to which the state administration is willing to cut down the durability of property rights to pursue the public interest objectives of registration, mistake represents one pivotal point in describing how land rules integrate with the constitutional scheme for the protection of possessions.Footnote 4 The loss of property following correction of a mistake is not, however, altogether without redress. When the registration of a proprietor involved a mistake and caused a former owner to lose property rights, the unsuccessful party in the correction proceedings will be entitled to statutory indemnity. But property scholarship has long recognised that some property rights will have personal significance far beyond their market value in monetary terms.Footnote 5 The concept of mistake and its relationship with indemnity is therefore the platform for addressing the extent to which land should serve social values beyond a mere abstract repository of wealth that is interchangeable with its cash equivalent.
The concept of mistake also influences the effective attainment of the policy goal of supporting the land market. To permit the correction of mistakes in the land register, albeit with indemnity, is to create an element of unreliability for proprietors which may tend to undermine faith in the register. This should be a concern to prospective purchasers, which may ultimately diminish market confidence in the register as an information source, leading to costly reliance on other sources of reassurance, altering the behaviour of conveyancers, and raising questions over the proper allocation of the costs generated by such disruption to market behaviour. The discretionary power dictated by the criterion of mistake therefore has a counterproductive effect on what is generally understood as the raison d'être of registration systems. Because of this it has not been accepted in the traditional Torrens systems of registration found in Australia and Canada;Footnote 6 there is, however, contemporary discussion over its possible introduction there as one way to ameliorate the rigours of the inviolable rule of title by registration that can lead to hard cases and questionable judicial efforts to counteract its effects.Footnote 7
Given these important matters that hinge on the meaning of mistake, this paper seeks to advance the understanding of mistake. It explores the circumstances and events which should determine whether any particular change to the register is regarded as falling within the concept of mistake. Part I isolates the specific subcategory of mistake that will be scrutinised. Part II considers whether mistake could remain an amorphous concept, capable reacting to any form of disappointed proprietary expectations, before ultimately rejecting that characterisation. Part III then proposes and reviews various contenders for a formula that could inject a degree of determinacy into mistake, discussing how each might subvert values of property law or hinder the attainment of relevant policy objectives. Part IV proposes a basis for mistake which rests on a set of rigid legal constructs about entitlement and registration. The proposed theory's interaction with various settled and controversial aspects of the land system is investigated, and its ability to handle external pressures is assessed.
I. Defining Terms: The Contextual Variety and Functional Diversity of Mistake
The power to correct depends upon the presence of a “mistake”. The same word is regrettably used elsewhere in the statute for very different purposes. In particular, it forms the basis for jurisdiction to correct the register of cautions against first registration,Footnote 8 to correct the register of title,Footnote 9 to award indemnity in an assortment of diverse enumerated instances,Footnote 10 and to enable the registrar to correct an application or accompanying document.Footnote 11 Because of the disparity in function and context, no useful definition of mistake can link together all of the occasions on which mistake is indiscriminately employed as the criterion for redress. The inquiry to be carried out by this paper requires the various statutory references to mistake to be dissected and allocated amongst separate categories according to their concept and purpose, discarding those which are irrelevant. The only usage relevant to this paper is the correction of mistakes relating to the register of titles. The species of mistake being investigated in this paper therefore covers a congruent subject matter: it defines the precondition for correction,Footnote 12 along with the first and second limbs of indemnity entitlement which are dependent on that correction jurisdiction.Footnote 13 Even within these confines, however, mistake lacks an obvious core meaning. The published preparatory materials illustrate its application rather than explain its content and it is left undefined in the statute, apart from a limited and unenlightening provision relating to compensation.Footnote 14 The following parts will offer guidance on its meaning.
II. Mistake as an Open-Textured Evaluative Standard
The correction scheme could be envisaged as allowing register entries to be upset by reference not to a fixed rule but an evaluative standard responding to an outcome which the court regards as sufficiently egregious. That is how the concepts of fraud, unconscionability or culpable neglect are often used in law, and mistake could be added to that catalogue of standards in order to refer the correction power to an individualised context-dependent value judgment. By this route the correction power could establish a broadly-based standard controlled by human or institutional expectations about title, which might include assessment of the distributive fairness resulting from amending or preserving a register entry.
Using this type of standard would achieve an open-ended concept offering substantial room for the exercise of discretion in defining and applying the values to be embodied in the correction power. It would have the capacity to respond to myriad individual factual contexts and would prevent unjust outcomes from slipping through the net. There is certainly a strong tradition in equity of emphasising responsiveness to individual circumstances and the supervision of morality through ad hoc decision making with little reliance on precedent for issues of application. It would require no great leap to extend this tradition to determining proprietary relief in correction claims by constructing suitable standards relating to expectations and fairness.
Fashioning mistake out of those criteria might secure the desirable result of penalising the unmeritorious and rewarding the deserving, and in isolation it might initially appear to be a viable method for explaining mistake. But once located in the context of land registration, it can easily be perceived as failing to integrate within the structure of the registration scheme and as detracting from statutory policy objectives. It would lead to the startling result that registered land titles could be upset even more readily than unregistered land titles (although indemnity would allay the most serious concerns). This formula is also objectionable for creating an entirely indeterminate concept whose content could not be guided by reference to existing and analogous legal doctrine.Footnote 15 Because registration involves a unique statutory system, there is no obvious web of existing principles into which registration must integrate; any analogues from common law are of limited value since the registry's status as a novel bureaucratic intermediary introduces novel issues that have no counterpart in common law.
An evaluative standard would incur all the disadvantages of vagueness in property definition that generate inefficient wastefulness, and so comes under the critical eye of law and economics scholarship.Footnote 16 It would diminish the predictability of case outcomes and decrease the out-of-court settlement rate, thus raising the total costs of dispute resolution in challenging and upholding titles after acquisition. The emphasis on individualised attention under the standard would exacerbate costs by placing a premium on fact finding that tends to inhibit summary disposal at the interlocutory stage. The anticipation of such costs could have a chilling effect on land acquisition and investment behaviour, and those who are averse to disputes might be deterred from land dealings due to the dangers created by the unpredictable standard. Taking further levels of professional input in an effort to assess, detect or mitigate the risk of an adverse correction claim might add to transaction costs and indirectly lead to market drag.
The use of a vague value standard to dictate the scope of correction power would often be a poor method for resolving entitlements where the disputed allocation of title occurred fortuitously without communication or dealings between the parties. In these circumstances it is ineffective as a means to discriminate between rival claimants.Footnote 17 In disputes between immediate transferors and transferees, a behavioural standard is entirely appropriate and is the basis for much of equity's regulation through conscience, typified by the reversal of transactions when one party has taken advantage of the other's vulnerability. But when it comes to the person who has acquired property through a registry error not of one party's making, or to a person who is at one remove from a challenged transaction, the use of a value standard is inapt to differentiate between rival claimants of equal innocence and the loss would presumably have to be left where it fell. This approach would neglect the opportunity to create rules having any instrumental effect in promoting behaviour changes that could advance the policy objectives of land registration.
On the other hand, the correction power could instead be controlled by a clearly defined and hard edged rule. The quality of predictability inherent in such a rule would avert potential costs of policing and enforcing property claims, it would allow better forecasting of the occasions for correction and ensure improved information about risk, thus removing a potential deterrent to entering the land market. It is for these reasons that this paper rejects an unstructured evaluative standard and proposes that the criterion defining mistake ought to conform to typical registration rules in exhibiting a high degree of determinacy.
III. Contenders for the Determinant of Mistake
Introduction: Freestanding and Referential Methods for Correction
Rejecting a value-based standard for the correction power leaves a definitional void. One hypothesis to plug the gap is a test which asks if the register indicates that rights are located where they do not actually exist. This might be termed a “monojural” approach as the presence of a correctable mistake would be determined by the freestanding set of property rules which prescribes the location of title in registered land. It would allow a correctable mistake to be found where, for example, a notice on the register turned out to be unsupported by an underlying entitlement to the property right claimed, or where a general boundary was inaccurately recorded. Although this test displays an abstract coherence, it must fail as a candidate for the correction criterion on grounds of policy as it is unable to act upon registered titles which are fortified by the doctrine of statutory vesting through registration.Footnote 18 The validity of these titles is due to the very fact of registration, so the test would always yield a negative result for correction jurisdiction.Footnote 19 If statutory vesting were to preclude the correction power then the special provisions about rectification would become redundant,Footnote 20 and the law would effectively be restored to its unsatisfactory nineteenth century position when rectification was not possible against registered estates.Footnote 21 That cannot be an acceptable conclusion because all registered titles, apparently even those acquired by fraud,Footnote 22 would gain from its indiscriminate protection.Footnote 23
Any viable theory for regulating the correction power must sanction the alteration of all titles, even those fortified by statutory vesting. This compels the conclusion that the determinant for the correction power must be sought in a “bijural” solution which requires consideration of two sets of rules: an initial determination of the location of rights according to the current state of the register, followed by a secondary review according to another set of rules for the purpose of comparison.Footnote 24 The following sections will look at possible candidates for such a referential method.
A. Unregistered Land Rules as the Comparitor
One possible bijural solution would be to introduce the correction power wherever there is a discrepancy between the allocation of rights according to the register and the distribution of rights according to the property rules of unregistered land. Selecting the unregistered land rules of property priority as the external standard for comparison would provide a set of well-documented rules possessing internal coherence and would promote consistency between outcomes in registered and unregistered land. It would ease the transition from the old system to the new, for whenever the principles of registration would postpone an interest, that fact alone would establish mistake and lead to either correction or indemnity. Its transplantation into mistake would therefore ensure that no rightholder would be worse off by the introduction of the registration system,Footnote 25 on the assumption that compensation is an adequate substitute.
The use of the unregistered land rules for comparison was accepted by Commonwealth commentators and was found in at least one correction provision of the former English statute.Footnote 26 But when it was apparently used in English case law interpreting the former generic correction power, commentators immediately reacted with warnings that subjugating the registered land priority rules to a discretionary correction power based on rules of unregistered land would fail to give due protection to the innocent acquirer of a registered interest and would undermine confidence in the land market.Footnote 27 That traditional argument against the unregistered land comparitor stems from a desire to protect every person who acquires a title that is already tainted by mistake. But this paper contends that such a policy is over-inclusive and that the balance between an ousted owner and an acquirer who is one step removed from a mistake should be managed with greater sensitivity.Footnote 28 If the registration of the acquirer were instead classed as mistake, this would introduce the necessary responsiveness through the remaining preconditions to correction: the rectification bar (dependent on possession),Footnote 29 its provisos (addressing fraud, care, causation and unjust outcomes),Footnote 30 and finally the exercise of discretion. Although this potential for correction might have a deterrent effect on prospective purchasers, that would be substantially offset by indemnity.
For those reasons, this paper rejects one traditional argument for dismissing the unregistered land rules as the bijural standard for the correction power. But there remain other valid arguments which justify dismissing the unregistered land priority rules. One argument comes from the costs of administering that system: delay and expense are possible as familiarity with the unregistered rules recedes into history, and there is likely to be wastage in resolving disputes over whether a particular unregistered land rule is incorporated or whether it ought to be disapplied in order to acknowledge the legal institutions of registered land.Footnote 31 A further argument against the unregistered land comparitor is that it would preserve a distinction between the binding power of legal and equitable interests. In a competition with a purchaser, unregistered priority rules usually depend on whether the interest was legal or equitable: the correction power would therefore hang on this distinction which is not easy to justify when formal documentation has been drafted, lodged for registration and publicised on the register.
The comparison with unregistered land is also undesirable because it perpetuates old habits. So long as correction can be relied on as a safety net, the registration system loses the capacity to change behaviour among the participants. This is particularly important because the unregistered land rules travel far beyond priority disputes between rightholders and purchasers.Footnote 32 It is easily overlooked that the discrepancy between registered and unregistered land rules might relate not only to priorities but any aspects of property, including the composition of the numerus clausus, the physical matter to be governed by real property law, the capacity of persons to hold and dispose interests, the circumstances of original acquisition, the events which transfer title, and so on. Many common law rules on these matters are reformed by registration systems to pursue the objectives of land registration by direct means or indirectly through simplification of titles and dealings.Footnote 33
Finally, it cannot be appropriate for correction to be dictated by a common law rule that has been pointedly extirpated from the registration system. The success of the registered land system rests on its ability to persuade participants to abide by the new registration ethic: creating only the simple permitted interests, using the stereotyped forms, protecting priority through entry on the register, taking the proper degree of care to discover others’ rights and preserve one's own rights, and so on. Harking back to the unregistered land rules at every opportunity to gain correction or indemnity would hardly stimulate the necessary change in culture and behaviour.
Collectively, these reasons militate against a correction power defined by reference to the unregistered land rules. Abandoning the unregistered property standard does, however, require one important compromise: any alternative method of determining mistake necessarily falls short of the ideal of compensating any deprivation of property rights caused by the functioning of the registration system that would not occur in unregistered land. It belies the slogan that a rightholder will be no worse off in registered land.Footnote 34 But that is of limited importance once registration has proved its social utility and taken root in participants’ consciousness. At that point the transitional comparison ceases to be such a pressing concern as operational simplicity, consistency and effectiveness.
B. Procedural Default as the Determinant
A comparison with unregistered land rules is only one of the possible bijural solutions that could define the determinant of the correction power. One putative theory of correction which escapes the problems associated with the unregistered land comparison is the test of procedural default by the registry.Footnote 35 Correction jurisdiction could be determined by asking whether, in changing in the register, the registrar breached a legal duty or exercised a discretion unlawfully. The fact that the act might have been effective to vest and divest title must be disregarded, as procedural default would be concerned only with fulfilling the necessary routines and not with the underlying rights.
Even though there are many instances when applying the concept of procedural default as the determinant for correction might happen to yield an entirely unexceptionable outcome, it must not be taken as the determinant for mistake. If procedural default were taken as the exclusive determinant, scenarios can be foreseen in which an interest could be lost without compensation. Defining mistake by procedural default alone carries the corollary that no correction is possible if the registrar has followed protocol and made an intra vires exercise of a discretionary power, even though a valid interest might thereby be expunged from the register. This result cannot be countenanced because, on the assumed absence of mistake, it would necessarily follow that indemnity would also be absent, leaving an uncompensated destruction of property rights.Footnote 36 This causes irreconcilable conflict with land registration's long-standing ethos of compensation,Footnote 37 and runs counter to the constitutional protection for property.Footnote 38 The concept of procedural default must be rejected as the basis for correction because of its potential to create this risk of uncompensated deprivation.
In Baxter v Mannion,Footnote 39 a case of utmost importance to the security of registered titles, the Court of Appeal rejected procedural default as the criterion for mistake. Baxter claimed adverse possession and applied for registration as proprietor in place of Mannion. He supported his claim with a statutory declaration which contained factual inaccuracies about the continuity and extent of his possession, yet the registry concluded that everything appeared in order and served the requisite notice on Mannion, inviting him to contest the claim. The time limit expired before Mannion took action and Baxter was accordingly registered on the strength of his statutory declaration. When Mannion sought rectification, Baxter argued that there had been no mistake as the registry had acted properly in reaching its decision on the material before it. The Court of Appeal rejected the argument, holding that the concept of correctable mistake was not to be restricted to “a mistake through some official error in the course of examination of the application.”Footnote 40 It expressly acknowledged that restricting mistake to procedural default would prevent Mannion recovering the land or compensation.
The rejection of procedural default followed from the risk of deprivation without compensation. It hinged on the relationship between correction and indemnity, both of which in the English system are activated by the criterion of a mistake whose correction would prejudicially affect the title of a registered proprietor. But, it is submitted, procedural default would remain an unsatisfactory blueprint for correction even if indemnity were made available.Footnote 41 Reliance on the registrar's procedural default would deflect attention away from the parties’ own rights, status or behaviour. The rules of common law and equity about transaction validity and property priority, which implement a sophisticated balance of moral values and utilitarian agendas, would all be discarded at a stroke. This would be especially unsatisfactory in the English registration model which readily confers title through a crude principle of statutory vesting that does nothing to pinpoint the class of acquirers who merit protection; a fraudster, for example, should not obtain unimpeachable title simply because of his skill in outwitting the registry's processes.
The discussion of procedural default up to this point has assumed that the register is changed to reallocate rights in a way that is not supported by an underlying entitlement. For completeness it must be recognised that a distinct analysis must be applied to a species of procedural default that occurs in another scenario: when a change in the register due to procedural default is nevertheless supported by an underlying entitlement. The typical example is a valid disposition which the registry accepts in breach of a regulation about the probative sufficiency of application documents. This is illustrated by the assumed facts of Fatemi-Ardakani v Taheri Footnote 42 in which the registry had accepted a transfer that had been executed by the donee of a valid power of attorney but had neglected to demand the power of attorney itself as required by the rules.Footnote 43 It was held that despite the procedural error there was no mistake.Footnote 44 The decision was correct in finding that the lack of supporting documentation to prove the matter to the registry does not comprise mistake. A mere omission from the applicant's bundle should not offer one party an entirely adventitious opportunity to invoke correction and resile from a valid disposition or impeach a valid transmission by operation of law. In more general terms, procedural default by the registry should never constitute a basis for correction where the expression of rights on the register is in accordance with a supporting entitlement.Footnote 45 That principle should extend beyond cases of an application's probative sufficiency to all cases of procedural default which do not alter property rights, such as the failure to serve statutory notices.Footnote 46
The preceding paragraphs have demonstrated that procedural default is neither necessary nor sufficient to introduce the correction power. There are occasions when a procedural default is integral to the event that deprives someone of an entitlement which ought to be redressed, but there may be other occasions when there is procedural default yet no deprivation of entitlement which requires redress.Footnote 47 Equally, there may be no procedural default yet a deprivation of entitlement which ought to be redressed.Footnote 48 To avoid the unnecessary multiplication of conceptual entities, it is a simple step to excise all reference to procedural default, exposing an underlying concept of substantive entitlement which accounts for the availability of the correction power. One possible form of substantive entitlement for determining mistake will be considered in the next section.
C. Departure from the Interest Preceding the Register Change
Care must be taken in identifying the type of substantive entitlement which is to rebut the accusation of an unsupported change in the register that constitutes a mistake. There is patently no basis for correcting when the register is changed to match the allocation of proprietary interests held immediately prior to the change in the register. But that is not to say that the reverse is true: correction should not be permitted merely because the register is changed in a manner that does not match the allocation of proprietary interests immediately prior to the change. The absurdity of that proposition can be observed in the simplest example of a transfer for value of registered land. When the registered proprietor of a legal estate executes a transfer, all legal proprietary effect of the instrument is sterilised pending its registration,Footnote 49 leaving the transferee with personal rights and at best an equitable interest in the land.Footnote 50 From the moment of taking delivery of the transfer, the transferee has only an equitable right, yet is entitled to convert it into a legal right by registering. It would be an intolerable impediment to the free transfer of land if the registration amounted to a correctable mistake merely because beforehand the purchaser had held only an equitable interest, or, as a purchaser under the e-conveyancing regime or a donee, had held no interest at all.Footnote 51
Despite the absurdity, there is one ambiguous case which could be read as consistent with the correction power being engaged where a register change departs from the preceding interest. Khan v Rehman Footnote 52 involved several disputed houses, including No.114 and No.75. The proprietor purported to execute a legal charge over No.114, but the apparent signature of the attesting witness was not genuine, presumed to have been forged by the chargee, and it thus failed to constitute a deed.Footnote 53 It could give the chargee equitable rights only, yet was processed by the registry according to its tenor. The court ordered rectification of the register to downgrade the entry from a legal charge to an equitable charge. The dispute in relation to No.75 concerned a transfer of the registered fee simple, which was defective for the same reason, and was likewise held to pass only an equitable fee simple. When the transferee was registered as proprietor with the legal estate, however, it was decided that rectification would not be ordered. Although the statutory underpinning for these parts of the decision was not fully explained, the tone of the judgment suggests that the correction power was accepted but that discretion was exercised in favour of preserving the entry. The brevity of analysis in the judgment leaves open various possible deductions about how the court perceived the determinant of the correction power. The finding of mistake could be attributed to the fact that the applicant held equitable interests preceding registration but was registered with legal interests. Insofar as the case invites that analysis, it must be rejected.
One exception exists to the principle that correction should not come into play merely because of a discrepancy between the quality of interests held by an applicant before and after the change in the register. The registration statute differentiates applications for first registration from applications to register a subsequent disposition: first registration extends to legal estates only.Footnote 54 In Sainsbury's Supermarkets Ltd. v Olympia Homes Ltd.,Footnote 55 it was conceded with the court's support that a correctable mistake occurs if the status of first registered proprietor is accorded to a person who previously possessed only an equitable interest.Footnote 56 It is submitted that the approved concession was rightly made and that registering Olympia as proprietor of the legal estate did engage the correction power. However, it will be argued below that the mistake arose not merely because of a discrepancy between Olympia's rights before and after registration, but because Olympia's entitlement was insufficient to authorise the register entry. The mistake lay in the unwarranted register change; the discrepancy between interests held before and after registration was not the origin of the mistake but an incidental effect of it.
Mistake cannot be characterised as a discrepancy between a disponee's interest before and after registration. Nevertheless, the correction power must somehow be connected to some form of substantive entitlement in order to avoid the conclusion that the law is unable to distinguish between the titles of a fraudster and an innocent purchaser. The nature of that entitlement will be developed in the following section.
IV. Mandate as the Proposed Formula for Mistake
Introduction: Mistake as the Absence of Supporting Entitlement
Having rejected an evaluative standard and various rigid legal formulae as potential determinants for mistake, it remains to describe a satisfactory, principled basis for invoking the correction power. Certain restrictions are implicit from the previous sections. Firstly, mistake should depend upon the entitlements of the parties themselves, rather than the registry operations, in order to escape the problems of procedural default. Secondly, this entitlement must refer to a system of rules other than statutory vesting to avoid the monojural fallacy. Thirdly, the requirement of promoting autonomy in transfers dictates that the mistake must not be equated to a departure from the interest preceding the change in the register.
Within these parameters, it may be postulated that a correctable mistake occurs whenever a change is made to a register but nobody was entitled to procure that change at the moment when it was made. This summary formulation captures the essence of the theory proposed in this paper but requires further clarification in order to explain how it applies to various factual situations and an assessment of the extent to which it can offer a universal solution. The following sections will examine aspects of the test in detail and how it would deal with specific problems.
A. The Mandate Supporting a Change in the Register
The entitlement should take account of all manner of rights recognised by the registered system as conferring eligibility to apply for a register change, typically an existing proprietary interest, but also extending to rights whose proprietary effect is suppressed pending registration.Footnote 57 In this paper an entitlement of this type will be termed a “mandate”. It is proposed that when the register is changed without the authority of such a mandate that a correctable mistake occurs.
The source of the mandate could originate in either of two forms: consensual disposition or transmission by operation of law. For consensual dispositions, it is necessary to prescribe the criteria for a valid disposition and the defects that would impeach an ostensible mandate. It is submitted that the relevant disposition must possess the attributes of both substantial and formal validity. Substantial validity is presupposed because the registration system embodies no general policy of altering the substantive criteria for property rights. Its primary function is the statutory re-ordering of priorities by rules that are bolted onto general property law and, except for a limited range of policy-driven reforms for the registration system,Footnote 58 the property rules of unregistered land should continue to supply all the tests for validity: whether the disposition meets the numerus clausus of property rights, whether those rights exist in a subject matter that conforms to the definition of land, whether there is a capable grantor and grantee, whether there has been true consent to dispose, and so on.Footnote 59
While it should be beyond dispute that a disposition's lack of substantial validity leads to a mistake when entered in the register, the same cannot be asserted so confidently in respect of formal validity. It is not immediately obvious that a mistake should be inferred from every deviation from formality rules. Assuming, for example, that a putative grantor expresses the will to make a grant, has the capacity to do so, and the grantee ultimately procures a corresponding entry on the register, should it really matter that the grant was not manifested in the form required by the rules? Property law has traditionally relied heavily on formalities to implement a variety of policy objectives: they alert the grantor to the instrument's legal effect, protect the grantor from external pressures, prompt the clarification of terms, evidence the grant, and so on.Footnote 60 These functions are also relevant to registered land dealings and should not be thwarted just because the grantee happens to have been entered on the register. Entries made pursuant to a putative grant lacking formal validity should therefore be liable to correction.
The link between formal validity and the correction power was seen in Khan v Rehman,Footnote 61 noted earlier, where the absence of a genuine attesting signature prevented each contested instrument from being a deed. It was held that there was a correctable mistake when the defective instrument was registered. The judgment achieved the correct result but ellipsed the intermediate logical step that explained why a deed was necessary. The judgment could be understood as relying on the formality rules for unregistered land and thus introducing the censured common law comparison into the grounds for mistake; better to have specified that the absence of a deed in each case was a failure to meet the statutory formality requirements for registrable dispositions.Footnote 62 When the ambiguity is removed in this way, the outcome can be explained exclusively by registered land concepts and supports mistake being founded on the absence of a formally valid mandate.
Where a change in the register has not been authorised by a consensual disposition, it might alternatively be sustained by a mandate arising by operation of law. That this source of entitlement precludes the correction power is already implicit from the case law. Adverse possession claimants who managed to secure registration without actually having taken adverse possession for the requisite period have been held open to correction;Footnote 63 and where a husband managed to secure a matrimonial home notice against a property which was not the matrimonial home, there was power to correct.Footnote 64 The position has been expressed succinctly: “A registration obtained by a person not entitled to apply for it would be mistaken.”Footnote 65
In principle, the concept of mandate also applies to first registration: a correctable mistake will have occurred if, on the first compilation of a register, there is an entry which was not supported by a valid mandate. In applying the concept of mandate to first registration, however, there is one significant caveat: the rules applicable to determine the validity of the entitlement are not the rules applicable to dispositions of registered land, but are instead the rules applicable to unregistered land except to the extent that they have been displaced by statutory requirements for first registration.Footnote 66 If the ultimate registered entry is out of kilter with this entitlement, as when it confers too much land, then a correctable mistake will have occurred.Footnote 67 The mistake might equally occur by the registry awarding rights of a different quality, as when an applicant holding only equitable rights is registered with legal rights. This occurred in Sainsbury's Supermarkets Ltd. v Olympia Homes Ltd. Footnote 68 where the court's recognition of a correctable mistake should have been attributed to the absence of any entitlement to procure the entry of legal rights.
B. Mandate and the Adjudication of Rights
Prior to the decision in Baxter v Mannion,Footnote 69 the editors of Megarry and Wade had already put forward a description of mistake along lines which rejected procedural default and instead anchored mistake in substantive entitlement.Footnote 70 Their latest formula is as follows:
It is suggested therefore that there will be a mistake whenever the registrar would have done something different had he known the true facts at the time at which he made or deleted the entry, as by:
(a) making an entry in the register that he would not have made or would not have made in the form in which it was made;
(b) deleting an entry which he would not have deleted; or
(c) failing to make an entry in the register which he would otherwise have made.Footnote 71
The crucial qualification in this extract is the opening allusion to the state of knowledge of the registrar. By assuming a hypothetical omniscient registrar who is aware of the true entitlements, the test sidesteps any question about the propriety of a particular registrar's procedures. Despite presenting the test in procedural vocabulary, it looks exclusively at the underlying substantive entitlement. The hypothetical omniscient registrar is thus neither more nor less than an anthropomorphic cypher for determining whether or not a change in the register was mandated by an entitlement to procure that change.
Condensing the idea of entitlement into the hypothetical registrar test gives a convenient rule of thumb, but its danger is that it does not explore the interaction between the entities responsible for pronouncing on property rights: registrars, adjudicators and courts. It does not prescribe how one should respond to an order from another competent authority. If, for example, the adjudicator commits an error of fact or law in reaching a decision, it is not clear whether the omniscient registrar has the capacity to detect that error and so render the entry mistaken. The issue is not far-fetched; it has been held that a register entry is liable to correction even though it had been entered by the registrar on the order of the adjudicator following a binding determination of rights. In Totton and Eling Town Council v Caunter Footnote 72 the adjudicator found that the Caunters had been in adverse possession, the registrar entered them as the owners of the land, and on appeal to court the council was held to be entitled to the land. The court ordered the registrar to de-register the Caunters, and the issue arose as to how this should be achieved. The judge commented:
I am told and I accept that procedurally the way forward is for me to make an order for the alteration of the register for the purpose of correcting it pursuant to para. 2 of schedule 4 of the 2002 Act. This is a correction in effect simply to reverse the impact of the decision which has been successfully appealed.Footnote 73
The decision can be envisaged in the terms of the hypothetical registrar test: the quality of omniscience would ensure that a hypothetical registrar would have been aware of the flaw in the adjudicator's decision, thus establishing the mistake and enabling it to be put right through correction.
The use of the mistake to undermine a judgment would require a highly artificial exercise in dealing with the consequences of finding that correction power exists. First, although correction is supposedly discretionary, it is inconceivable that a higher court's reversal of a lower court's decision would not lead to the corresponding reversal of the entry if no third parties have intervened. Secondly, the rectification bar should also be a dead letter and not permitted to stultify the higher court's judgment.Footnote 74 Thirdly, the failure to seek a stay of the lower court's order should demonstrate a lack of care that forfeits protection. Fourthly, there is little reason to provide indemnity to a party simply because of the temporary entry of a flawed judgment that is reversed, where no third party has relied on the register. By requiring that the correction jurisdiction be stripped of four distinguishing features – discretion, unavailability against proprietors in possession, conditional dependence on care, and associated indemnity – it may be argued that correction is being commandeered for processes that rightly belong to other limbs of alteration. This is strikingly evident when judgments are successively reversed and reinstated on appeal, and the entry's “mistakenness” is seen to fluctuate. This characteristic indicates that the status of the entry is being governed by new occurrences outside the register. Giving effect to new events should be implemented through a power which is dedicated to that purpose, the better candidate being the head of alteration which exists to “bring the register up to date.”Footnote 75
With the reversal of judgments on appeal removed from the domain of correction, and the decision in Totton and Eling Town Council v Caunter Footnote 76 cashiered, the potential discomfort over the hypothetical omniscient registrar test would vanish. It would afford a neat compendium of mistake principles provided that it is made clear that the hypothetical registrar's capacity to detect flaws in orders of the adjudicator and court is either suppressed entirely, or else subordinated to his duty to comply with them.Footnote 77 The clearer formula of mandate would avoid any entanglement with the issue.
C. Mandate and the Race to Register
Discussion of the mandate theory now turns to cases in which the register has been changed and the basis for challenge is not that the applicant's entry was unsupported, but rather that at the time of entry there was another outstanding claimant holding an unregistered right. These cases are important in evaluating the mandate theory as the response to that challenge determines whether correction prevails over the fundamental principle that an unregistered right is deferred to a duly registered estate acquired for value.Footnote 78
Analysis is assisted by separating out the two distinct phases occurring before and after the disputed registration, termed preliminary and postliminary respectively. The preliminary phase continues while the land remains in the hands of the person who suffered the creation of the unregistered right, or comes to an assignee against whom the right is enforceable notwithstanding its omission from the register. During this phase, the absence of an entry does not detract from the holder's ability to assert it and its entry should be allowed in order to assure its future priority over purchasers and promote the comprehensiveness of the register.Footnote 79 That is not to imply that its entry should necessarily be channelled through the correction power. There are other suitable vehicles, including the simple submission of the relevant instrument or, where this is impractical, the use of the updating head of alteration,Footnote 80 or the other miscellaneous methods for entry.Footnote 81
Once the land has passed to a person who is protected by the priority rules against enforcement of the unregistered right, the postliminary phase is entered and the omitted right should be incapable of entry by any means. As between rightholder and purchaser, the balance is already struck by the priority rules of the registration system. Immediately prior to the entry of either right, the two rival rights co-exist as relative entitlements.Footnote 82 Each, taken in isolation, would provide the mandate for entry on the register and so whichever rightholder wins the race to the registry, the registered entry will be free from mistake.Footnote 83 Even if the race were won by the second disponee, thus reversing priority, that consequence is the proper reward for the applicant's diligence and the proper penalty for the tardy rightholder's failure to warn purchasers.Footnote 84 No safety net should be provided through the correction power.Footnote 85 In this way the concept of mandate operates in conformity with the parameters set by the principle that rights must be protected on pain of subordination to subsequent purchasers.Footnote 86
Very different policy considerations are relevant where one of two rival claimants gets on the register at first registration. The mandate principle would ensure that a correctable mistake occurs if at first registration a rightholder were awarded lesser rights than those previously held, or if the registry were to award no rights at all by vesting an absolute title to the estate in another and thereby destroying the omitted right unless saved as a protected interest.Footnote 87 This may be surprising given that the first proprietor, by definition, could not have relied on the register when purchasing and does not have a particularly compelling call to security in the title so conferred, while the unregistered rightholder may have been quite unaware of first registration and the need to protect his position. The balance between the sanctity of property rights and the aspirations of the first proprietor would therefore be poorly struck if it destroyed the omitted right without hope of its resurrection through the alteration power. It invites the inference that the purpose of first registration, as emphasised by the inquisitorial processes undertaken, should be seen as the assembling of a comprehensive record which, within the resource constraints of registry operations, discovers all existing interests affecting the title. Under the English model, any omitted rights ought to be capable of being reinstated at a later date.Footnote 88
The mandate theory proposed by this paper, however, is unsuited to that task. It seeks a supporting entitlement which justifies the entry made, but it does not require proof that the entitlement is free from adverse claims, which would be a necessity for achieving the total capture of rights at first registration. An applicant for first registration might be able to show title to a legal fee simple, yet if an equitable easement,Footnote 89 for example, were omitted from the initial compilation, that alone would not establish a mistake under the mandate theory. If omitted rights at first registration could not be processed through other mechanisms and were forced into the correction regime, the mandate theory would be unable to handle them appropriately. A special subsidiary formula for mistake would be required to constitute the necessary triggering event for correction jurisdiction. That can be seen not as a flaw of the mandate theory but a confirmation of its integrity and an indication of the contextual differentiation required to pursue a plurality of registration policies.
D. Mandate and Successors
The greatest challenge for any theory of correction is its treatment of the successor when a registered proprietor (RP1), who is mistakenly entered without valid mandate, then makes a registered disposition in favour of a successor (RP2). In preparing the registration statute, the artful drafter avoided prejudging whether the title of RP2 might be corrected and there are no direct indicators in the alteration or indemnity provisions of the parliamentary intent. One possible interpretation of the provision is that the entry of RP2 should not be regarded as a mistake.Footnote 90 This argument rests on the foundation that, although the entry of RP1 amounted to a mistake, by virtue of his registration he nevertheless possessed the legal estate and owner's powers of disposition, so that the subsequent disposition, being sanctioned by the statutory empowerment, enjoys full validity and its registration cannot be a mistake.Footnote 91
This paper has already criticised the use of any bijural comparitor which would give successors universal immunity against correction claims, and those criticisms apply equally to attaining the same result via the statutory empowerment solution. In particular, it is submitted that a rule as blunt as the statutory empowerment solution, which can expropriate a registered rightholder in favour of, say, a fraudster's donee, will not strike the right balance. The sacrifice extracted from the dispossessed rightholder may, on any measurement, be grossly disproportionate to the weak justifications for protecting particular classes of recipient. The statutory empowerment solution does have the merit of simplicity, but it comes with crass indifference to the many varieties of transferee. The adjustment between rightholders and remote successors should be more finely tuned to their respective needs and merits. This could be achieved by admitting the existence of mistake and thereby resolving the dispute through the subtle and discriminating rules about possession, consideration, fraud, care, causation, and justness which are engaged even before reaching the exercise of discretion to correct.Footnote 92 That approach would also escape a serious deficiency associated with the statutory empowerment solution which leaves the owner with an impractically short-lived opportunity to discover the fraud by RP1 and take protective or remedial steps before RP1 hastily sells on or mortgages the property.
The prospect of correction against remote successors would collide with grand axioms attributed to registration which encapsulate its mission to protect registered transferees from unprotected interests.Footnote 93 But it is not obvious that transferee protection should operate in an unqualified fashion, indeed it has always been the position that different levels of protection are dispensed to different participants by different doctrines: statutory vesting applies to all registered proprietors, freedom from unprotected incumbrances applies to registered disponees for value, presumptive immunity from correction applies to those in possession. In the face of this diversity, it would take the inductive process a step too far to elevate the specific priority rule for never-registered interests into a universal precept that should regulate correction claims against all transferees. At most, protection should be limited to transferees who pay value, but that limitation is not feasible under the statutory empowerment solution.Footnote 94
Whatever views may be held on the optimal balance between owners and remote successors, there are insidious problems that would accompany the statutory empowerment solution. These problems arise due to the statutory indemnity being linked to mistake.Footnote 95 If RP2's entry is to be held free from mistake, the unpalatable result appears to be that the dispossessed owner obtains neither reinstatement nor compensation.Footnote 96 The absence of compensation would breach the constitutional guarantee of possessionsFootnote 97 and an owner would be even worse off than in unregistered land, where legal title is protected against all comers. It is, however, possible to envision an interpretation of the registration statute which confers indemnity on the ousted owner without having to treat the entry of RP2 as a mistake: this is achieved by construing the historic entry of RP1 as the mistake which introduces indemnity.Footnote 98 That approach would be compatible with the statutory empowerment solution and would yield compensation, but still suffers the objection that due process in expropriating the rightholder is not satisfied by compensation alone but demands a balanced assessment of the rule's purpose and proportionality. Compliance with that model of due process must be doubtful given that the statutory empowerment solution confers indiscriminate protection on all registered disponees.
Two other routes might be proposed which allow correction against the remote registered successor while supplying indemnity to the loser.Footnote 99 The first route proposes that registering the remote successor constitutes a mistake in itself. By establishing the mistake, this analysis has the advantage of ensuring that indemnity will be forthcoming. It would, however, require mistake to encompass an entry even though it had been made pursuant to a valid mandate derived from the statutory empowerment solution. That approach would demand a reconceptualisation of the basis for the correction power. It would have to rely on something other than a supporting mandate as the determinant for mistake. It would leave options that have already been discarded, such as the use of procedural default or an evaluative standard. The first route therefore does not offer an acceptable account of the correction power.
The second route to rectification of a successor's title with indemnity proposes that mistake is curtailed by the statutory empowerment solution, but takes an expansive view of the consequential relief. It holds that the mistaken entry of RP1 introduces correction jurisdiction, and that the entry of RP2, although itself cleansed of mistake, may nevertheless be corrected to provide relief from the earlier mistake.Footnote 100 This route might be described as the “long-arm” jurisdiction to correct as its reach extends beyond the original mistaken entry to all derivative dispositions. Long-arm correction does not of course resolve the underlying policy tension between the relative protection of owners and acquirers but simply deflects the debate away from the triggering event of “mistake” and onto the remedial scope of “correction”. But in doing so, it brings a quadruple benefit: it preserves the universal and coherent framework for mistake offered by the mandate theory, it admits correction against registered successors, it guarantees indemnity for the loser, and it avoids classifying the correction claim as a traditional property right whose priority might be assured as an overriding interest.Footnote 101
V. Conclusion
This paper set out to penetrate the mystery of mistake. It has put forward a principled basis for understanding a concept whose content has been obscured by the apparently open-ended statutory term in which it is expressed. The proposed structured formula for mistake effectively amounts to an algorithm for ordering the inquiries to be made when mistake is in issue: Has there been a register change? Did that register change appear to confer greater rights than were contemporaneously justified by any supporting entitlement? Would that entitlement have been eligible for entry in the register if it had been lodged by the rightholder in a formal application? Did the entitlement pass the tests for substantive and formal validity necessary for a right of the intended type? Positive answers to all of these would establish mistake.
By identifying a set of rigid and exhaustive components for determining mistake, the proposed formula furnishes the predictability that tends to promote confidence in the property market. Comprehensive attainment of the ideal of predictability in the concept of mistake is yet to be confirmed, however, as this paper is restricted in its scope to mistakes by changing the register, and has not answered the remaining controversial question of the potential for mistake by bare omission. Predictability alone is of course not the sole criterion for an acceptable correction power. The jurisdiction to correct mistake must not jeopardise those principles of property which ought to apply to unregistered and registered land alike. The mandate theory of mistake pays respect to various such principles. In particular, because the mandate theory depends on substantial and formal validity, it cannot stray outside the closed list of property rights, it cannot nullify the protective functions of formality rules, and it cannot alter the rules for dispositive intent which ensure autonomy over transactions. Most importantly, this account of mistake ensures that the classic attribute of durability in property rights is preserved: any proprietary entitlement which is conferred by a disposition recognised by the general law, and which is protected as an overriding interest or as a register entry, is assured of perpetual security (whether through correction or indemnity). It confirms that, in the absence of fault, registered rights are guaranteed against loss caused through a change in the register which involves either improper deletion or the entry of incompatible rights.
Mistake should also be assessed by the extent to which it is antagonistic towards the policy goals of registration. The mandate theory encounters a couple of complications in harmonising with potential goals. One such complication relates to title at first registration: assuming, somewhat controversially, that first registration should not preclude the insertion of interests omitted by oversight, it appears that the mandate theory does not allow mistake to be employed for this purpose – although there may be alternative mechanisms for securing entry. The second complication relates to remote successors. The primary goal of English registration was to improve the purchaser's lot, and this would be weakened by the mandate theory with long-arm correction that would permit correction against successors. Nevertheless, this paper has argued for that result. In the contest between a purchaser and an earlier rightholder who never registered nor received the protection of overriding status, the mandate theory accepts that the purchaser always prevails; but when the contest is a successor versus an earlier rightholder who did register, yet was removed without mandate, this paper argued that any response based on a rigid rule of universal priority or subordination is undesirable. Enlisting the correction power to resolve the dispute would allow a preferable legal solution that is tailored to the relative positions of the disputing parties themselves.
Mistake must not be analysed in the abstract. It is pervasively conditioned by its statutory context. First, the event of mistake dictates not only the availability of correction but also the indemnity which is crucial in securing human rights compatibility. The concept of mistake must therefore flex to accommodate the needs of the indemnity clause. It is the impetus for compensation which should be instrumental in guiding the interrelationship of mistake, procedural default, long-arm correction, and the status of remote successors. Secondly, mistake should also respond to the existence of the other mechanisms which serve related or overlapping policies. It is only one of several conditions for alteration of the register, and the territory occupied by the other heads of alteration must be acknowledged in order to regulate their respective boundaries. Thirdly, mistake might also be susceptible to pressures arising from judicial decisions which find alternative mechanisms to detract from the register, such as finding broad generic powers to intervene, imposing in personam liabilities, or reinterpreting the scope of registered rights.Footnote 102 Fourthly, the concept of mistake interacts with the rule of statutory vesting which confers protection on all registered proprietors without differentiation. The inability of the vesting rule to distinguish the meritorious from unmeritorious indicates the desirability of a liberal regime for correction which possesses the sophistication to undertake a more sensitive assessment of competing property claims. It is mistake that introduces this element of contextual balance and can be seen as imparting a limited moral dimension to registered property. Blandishments about the discretionary nature of correction and its capacity to achieve individualised justice must not, however, obscure the point that they should prevail only within rigid jurisdictional constraints. It is the purpose of the paper to show that mistake should be and can be given fixed boundaries in order to maintain registered property's core characteristic of predictable durability and, at the same time, support the purchaser protection programme that underlies land registration.
