Dissertation
Introduction
The law on certainty of subject matter (CSM) has remained unchanged since its origin in the 1841 judgement in Knight v Knight[1]. Though there were developments as seen in Re London Wine[2] and Re Goldcorp Exchange[3]; cases demonstrating that property put on trust had to be separated and segregated for the trust remain to be enforceable. Later, the judgment in Hunter v Moss[4] distinguished intangible property from tangible assets which resulted in cases of exemption from the segregation requirement. The CSM rule, in the context of trusts, is a simple one: unsegregated tangibles are invalid whilst unsegregated intangibles are valid. This applies regardless of the nature of the trust and, in this way, the law on certainty of subject matter is made certain. Nevertheless, the law can be improved upon as, indeed, there are cases where, per the strict certainty, trusts and beneficiaries have suffered inequitable and perverse outcomes. The current law lends certainty, but does it give justice?
This paper argues that the CSM rule is the unintended precedent from a misunderstood judgment in Re London Wine which has since been erroneously used as the daguerreotype for all corresponding case law. A different approach is needed, and possible, in order to protect the realisation of settlors’ wishes and to prevent failures of trusts. Drawing inspiration from the above case law reports, scholarly opinion and legislation, this paper explores the requirement of separation and segregation of assets to produce a novel view of certainty of subject matter; arguing that a departure from the current rule will promote justice above the predictability of the law. This is underpinned by two main arguments. First, distributional certainty should be formalised into CSM provided that the fungibility of assets is recognised instead of their tangibility. This would cease the discrimination of tangible items in the context of trusts and would facilitate the disposal of property. Second, the universal rule of segregation is unfit for the varying needs of different trusts. A customised taxonomy, bridging common sense and orthodox parameters, will produce fairer and more just outcomes leading to less trusts failing. A good first revision may be the Law Commission’s project, ‘Modernising Trust Law for a Global Britain’[5].
Tangibility and fungibility
There are certain legal differences between tangible and intangible assets and their fungibility in the context of trusts but the process of creating them is the same. The first requirement to create any trust is to satisfy the three certainties. There are concerns about the principles underlying these certainties[6] but perhaps the most important is, still, certainty of subject matter. The position of the courts is that subject matter of a trust, if tangible, should be separated from the settlor’s, and if intangible, it does not need this separation[7]. This paper suggests the courts were wrong to create a division and that tangibility, in the context of trusts, is immaterial. What matters is the fungibility of items subject to trusts[8].
Case law has clarified a seemingly simple answer to these conundrums that applies to any kind of trusts. The binary rule[9] applies to showcase examples of this distinction are Re London Wine and Hunter v Moss[10]. In the former case, the trust was deemed invalid because the bottles of wine, paid-for and contractually deposited by the seller, had not been separated or marked with the names of the purchasers[11]. In the latter case, the trust of shares was considered valid despite no segregation at all[12]. Regardless of the efforts of the defendant’s counsel to argue for the ruling from Re London Wine to be applied, it was rejected on the grounds that there can be no distinguishing between shares owned in the same company[13]. Though the judgment was met with consensus approval, its reasoning was criticised by many scholars[14] concerning the fungibility of tangible and intangible assets[15]. Following Lawson and Rudden, “fungibility in its proper legal sense is not confined to tangibles”[16], it is applicable to intangible property such as shares as well[17]. They explained that the essential requirement for fungibility of property is that it must be divisible into separate units that are capable of being separately owned by different individuals[18]. With tangibles, there must be at least two separable items, not one single item that cannot be divided without loss of identity. As this is tantamount to co-ownership rather than the holding of property on trust[19]. With intangibles, they have to be capable of legal division into units without loss of individual ownership[20]. Hunter v Moss’s criticism is that the courts failed to show that shares are fractions of a single unit co-owned by shareholders and not owned separately by individual people[21]. Therefore, according to Lawson and Rudden, shares cannot be segregated[22] and, thus, cannot be transferred or owned separately. This, however, for the last four centuries, is exactly how shares have worked. Further to this, if shares cannot be identified and segregated, then they cannot be sold or owned because ownership is uncertain[23]. The judgment in Hunter v Moss, then, would be inappropriate and the trust void as the certainty of subject matter could not be satisfied. It seems only logical for intangible property to be treated differently otherwise shareholding, from the legal view, is impossible. The court, having not explained that ownership of shares is like owning something as tenants in common, made their judgement on the basis that shares of the same company are equal without any physical or legal differences between them[24]. Bottles of wine, though, can be corked[25]. The question is whether tangible assets of the same batch are different from each other and whether intangible assets are the same[26].
A new authority on certainty of subject matter, after Hunter v Moss, is seen in Re Goldcorp. Upon its insolvency, the company said it did not have to give away the purchased bullions to buyers[27]. Even though the sellers provided certificates of sale to the customers the Privy Council said that, although the vendor declared themselves a trustee, there was no real intention to create a trust for the benefit of the purchasers because the bullions had not been separated[28]. Lord Mustill decided that since legal title was not passed, there had been no separation of property that could be attached to a trust, and the customers were mere unsecured creditors[29]. This judgment is frustrating in the light of Hunter v Moss. If the decision was based on fungibility, not tangibility, the result may have been different. However, to arrive at this judgment, the court must accept the interchangeability of chattels. Sarah Worthington, in critique of Re London Wine, argues it does not matter what a bottle of wine could turn out to be corked because it can only be known upon opening the bottle[30]. A valid trust of segregated and separated bottles of wine would not be held void if some of them turned out to be corked. The court may likely deem it bad luck and not dissolve the trust or provide any restitution in trusts law[31]. In this sense, all bottles of the same wine are fungible because they all bear the same risk of being undrinkable and corked. Similarly, in Re Goldcorp, bullion bars should not have been construed as fungible assets of a homogenous mass[32] as they are serialised with the manufacturer’s stamp and a unique identifier[33]. The trust of 50 shares in Hunter v Moss would not have been valid either as share certificates are also serialised. Further to this, according to London Bullion Market Association (LBMA), customers do not own specific bullions of gold but have an entitlement to an amount of bullion from a batch[34]. Benny Chung argues there is no reason to separate bullions because, on the balance, people invest in gold at times of instability and care only that the value is correct rather than which bar belongs to whom[35]. Whether the bars are divided or melted down, the purchasers are only concerned with realising its value. For the LBMA, holding bullions unsegregated for multiple customers is the better option as it is a smaller administrative burden with cheaper storage costs[36]. They also prefer running transactions by weight rather than by specific piece[37]. So, for the parties involved in a gold bullion transaction, it is best not to separate individual bars from the bulk. If certainty of subject matter is to remain a requisite for the creation of trusts, the fairer, more logical standard of separating trust property should be their fungibility, not tangibility. This solution fixes settlors’ administrative shortfalls, reducing their steps taken, and would likely decrease the failure rates of trusts. Despite the analysis of the nature of gold bullion transactions, Re Goldcorp Exchange has not been overturned and current law favours legal certainty and predictability rather than justice or, in this case, efficiency.
Do we need certainty of subject matter?
David Wilde produced a paper in 2020 pointing to a fourth certainty inherently within the CSM – the distributional certainty[38]. Wilde claims that it is a requisite in private trusts that has yet to be recognised[39]. Plainly put, it is the certainty of ‘who gets what’[40], it is an elevated certainty of subject matter determining what goes into the trust and also the beneficial share. Wilde believes that the courts, by not incorporating this as the fourth certainty, exercise more discretion enforcing or failing trusts that lack some certainties[41]. Following McPhail v Doulton[42] the threshold for certainty is low – courts only need the minimum certainty to enforce a trust and, where possible, tend to infer the intention to make up for the lack of the distributional certainty[43]. As far as it saves a proportion of trusts from failure, and their beneficiaries from a big disappointment, it should not be the judges’ burden to be aware and care for something that does not formally exist. Wilde’s point is that there should be a certainty regarding “the interest to be enjoyed by the objects” suggested already in Knight v Knight[44]. This reasoning could be used for the purpose of abandoning the requirements of CSM of separation and segregation. There has been controversy, or even misunderstanding, about the mandatory segregation of tangible assets, but not intangible ones. Wilde’s idea could be formally incorporated into CSM to make the requirements simpler for tangibles and intangibles to prevent perverse outcomes and failing trusts. Instead of the current CSM could be a distribution that required the settlor to ascertain particular beneficiaries’ beneficial interest[45]. It means that the overall trust property would be known and, further, the settlor could declare which share he wants each of the beneficiaries to have without segregation of assets. This could eliminate trusts failing as a result of unseparated assets and prevent people losing what they reasonably believed was theirs. There are risks with this, as settlors may fail to distribute their property “on paper” and judges may exercise discretion and find the requisite minimum certainty to enforce the trust. Still, it is likely that a settlor will want to make clear the “who gets what” in a way that expresses his will concerning the individual share of his beneficiaries than to rely on the trustee to physically divide his assets, whether they are tangibles, intangibles or some form of tertium quid.
Removing certainty of subject matter is drastic and perhaps not entirely effective. Incorporating the distributional certainty into the CSM as a restorative certainty, if and when CSM fails, is the best alternative. This is not some peculiar legal loophole - such as Re Rose[46] - rather an enforceable means to save valid trusts. The orthodox rule, then, would prevail but be enriched by a restorative tool to promote trusts as facilitative devices[47] and would limit unfortunate, unintended outcomes. However, this would create another universal taxonomy for different types of trusts. It may be unfair to have a cure-all rule that decides the validity of agreements and, later, this paper will examine different taxonomies. Ultimately proposing the adoption of parts of distributional certainty based on the intention of the settlor and the reasonableness of judges.
The purpose of trusts and the purpose of the law
The law can either serve its subjects and facilitate their wishes or restrict them and demand obedience[48]. Trusts belong to the former and are a privilege in the legal system that not all civil law countries have. More specifically, according to Hart, trusts ‘realis[e] their wishes, by conferring legal powers upon them to create, by certain specified procedures and subject to certain conditions, structures of rights and duties within the coercive framework of the law’[49]. Trusts are a device for people to dispose of their assets and money as they wish to. For this reason alone, should the law on certainty of subject matter be more understanding and flexible?
The rule, a product of archaic Victorian law in combination with a misunderstood judgment from 1986, is not suited to today’s needs or to people’s relationship with property. Lord Langdale formulated the requirements for the three certainties at a time when almost everything a person owned was worth its weight in gold[50]. Prized possessions were everyday things like furniture, tableware and blankets. They were tangible. Today they are both tangible and intangible[51]. What matters today is something we like to call efficiency. Moreover, humans being “inequity averse”[52] will prefer what seems just to what is certain in law. Justice and fairness of legal provisions are assessed against the outcomes they bring in real life cases, whether they are appropriate or not. The appropriateness of rules on certainty of subject matter will depend on the taxonomy of a type of trust in question.
Peter Birks’ claims that a taxonomy should promote logic, order, rationality and consistency[53]. These cannot be applied to trusts because they do not have a proper taxonomy despite the universal and orthodox rules, the poseurs, pretending so. A settlor can never be sure if his trust will fail against CSM. Why? Because there are different types of trusts held to the same rules and their validity is at the discretion, or mercy, of the courts. And, as new kinds of property arise, tested against anachronistic moulds, the old rule will fail its people and property will be lost to uncertainty.
Proposed taxonomy
The inspiration for the following taxonomy was derived by considering different situations of potentially valid trusts failing against CSM. A thought experiment could be conducted, and the following questions could be asked:
If I have 10 pairs of identical blue jeans and want 5 of them on trust for my sister, I should not be legally forced to spend 5 minutes of my busy life to separate them from the ones I intend to give to somebody else, provided I have enough left. If the jeans remain in my wardrobe until I think she is ready to ‘inherit’ them, I should not have to sew on her initials to show they are intended for her, provided that I have enough.
If my wine-storing company goes bankrupt and I owe bottles of wine to my creditors, should I have to spend my employee’s labour to tag each bottle with its new owner’s name (notwithstanding the fact that there are computer systems that record each purchase and their receipts), provided that I have enough?
If I set up an express inter vivos trust with trustees, do I need the requirement to separate the items intended for it since they are going to be ‘moved’ together with the certificate of legal ownership, deed and/or physical possession until the beneficiary receives it? If my trustee has a bulk of such items himself should he have to segregate them, provided he has enough to pass to my beneficiary? And if this trustee acts unconscionably and takes some of the trust property for himself, should there not be a procedure similar to tracing in place for him to be held liable?
If I set up a trust and do not have enough to distribute to my beneficiaries, should I make up for it, for example, in damages?
Similar questions were asked in the aforementioned cases where the law on trust property was following the path cultivating clear separation and segregation of the trust property. Though the questions can be asked by anyone with property who wants to dispose of it in the form of a trust. The following taxonomy proposes a departure from the current system and demonstrates how it may be possible to increase the rate of just outcomes.
Different kinds of trusts
For the reasons explained, the requirement of segregation does not suit all situations concerning trusts and it is not needed where it is sufficiently clear as to what the settlor intended to put on trust[54]and which property is meant for whom. The only scenario where segregation of trust property is required is where there is not enough to give out to the beneficiaries upon execution of the trust. The following taxonomies of trusts, present different scenarios with varying compositions of ‘stakeholders’, circumstances and different forming requirements. The scenarios highlight why they deserve a better adjusted set of rules to ensure the most effective disposal of property accordingly to settlors’ wishes.
Testamentary trusts
Testamentary trusts come into effect upon the settlor’s death when the legal title of property left in a will is transferred to the trustee[55]. Settlors have a large degree of discretion as too what they put on trust and the instructions they give to their trustees[56], unless the instructions cause a waste (of land or life) and prevent utility of the settlor’s assets to the degree of such irrationality that the court can interfere[57]. Requirements to create a testamentary trust are the three certainties coupled with formalities. Upon death of the testator, the legal title to the trust property is transferred to the trustee and certainty of subject matter compels him to hold that property separate to his own (in case of tangibles of course)[58]. But is this necessary? Provided that the instructions and wishes in the will are clear enough, everybody concerned with it should know who the testator intended to benefit and exactly with what beneficial share. If it can be said with certainty what the testator intended, then the requirement of separating the trustee’s assets from the trust’s is a redundant step. Such trust then should not fail on the separation requirement if it can be said who is intended to inherit what. If that trust still fails for this reason, what the court will do is decide whether they can allow creating the trust regardless for the beneficiaries if it can be worked out what was meant for whom and if not, a resulting trust will be created for the benefit of the testator’s estate[59]. So, incorporating the distributional approach to testamentary trusts would allow the court to exercise their discretion and common sense without moving the property around and changing the trusts it is on.
Distributional certainty in testamentary trusts is then the middle ground between CSM and the court’s discretion. It is very much doable in real life because equity is likely to make concessions on the rules to save trusts from failing. Proof of this is that normally, if the trust fails on any of the three certainties or the formalities, the trust property goes back on a resulting trust to the testator’s estate. But there are rules that help out in such instances such as Re Rose or Pennington v Waine[60]. These two provide exceptions to the strict requirements for valid trusts. Which shows that equity does allow alternative solutions for a trust to be saved from failure.
Self-declared trusts
Self-declared trusts where there is no requirement of proving certainty of subject matter[61]. Self-declared trusts are a peculiar type of trusts that could help pave the way to a reformed approach to CSM and the requirement of separation and segregation of property on trust. The most important requirement to form a self-declared trust is, as its name suggests, the declaration of intention to create the trust[62]. As per Scarman LJ in Paul v Constance[63] there must be sufficient intention to create trust expressed by declaration and clear evidence. It is a twofold intention to hold a trust for another’s benefit and to take on the obligations as a trustee[64]. Then, the legal title of the subject matter is not conveyed to a third party but remains with the settlor. The beneficial interest is then intended to be passed on but without completing the formalities deriving from the traditional trust formation. There is no transfer of the legal title by either physical transfer or a deed. Therefore, self-declared trusts are an exemption when it comes to certainty of subject matter. Section 53 of the Law of Property Act 1925 is disregarded and substituted by declaration as the primary requirement for creating a valid self-declared trust[65]. Certainly, there is the requirement of naming specifically the beneficial interest for the object of the trust. The “bulk of my residuary estate” from Palmer v Simmonds[66] will not suffice. But there is no obligation to separate property from the settlor’s bulk because of the strength of declaration of intention as an extra formality[67]. This proves that the strict requirement of the three certainties can be and have been overlooked due to the specific characteristics of different types of trusts. This gives hope for other trusts to also be treated individually and to avoid unfair perverse outcomes in future.
A self-declared trust is based on sufficient intention supported by declaration. What is interesting is the debate over the form of the declaration. As in other trusts, the declaration can be written or oral[68]. Most likely, it will require witnesses to attest the settlor’s intention. Lightman J said in Neptune (Vehicle Washing Equipment) Ltd v Fitzgerald[69](Neptune) that the courts are prepared to recognise declarations made in private so long as they are recorded[70]. In Neptune the question was whether a director of a company, holding a solo meeting with his secretary taking notes, complied with s. 317 of the Companies Act 1985. It was decided that because the resolutions were recorded, they were enforceable[71]. It was said further, although obiter, that the director might well have been alone, without the secretary, and the resolutions still would be valid if they were recorded[72]. This begs the question if the act of recording can go beyond taking minutes of meetings. Perhaps ‘recording’ can be taken literally, and the director could record himself, on audio or video, declaring a resolution in a company meeting, or declaring a trust only by himself.
The example of self-declared trusts shows that exceptions in terms and requirements for creating trusts exist and are allowed by court and by equity. Self-declared trusts do not match the transfer of the legal title or formalities, but they extend the certainty of intention and demand a declaration of intention[73]. The law allows valid declarations without witnesses, even though it is a standard requirement especially for establishing trusts[74]. If the law allows such a change of recording in its broad sense perhaps it could also apply to certainty of subject matter. Instead of having to separate and segregate chattels, shares or even cryptoassets[75], the settlor could record, in any of the allowed ways, what property he wants to give to whom. This would eliminate the spare requirement of separation and segregation but would increase the certainty of beneficial interest and distribution by the settlor’s declaration. This would potentially be more efficient for a settlor to record his voice saying how he wishes to divide his property on trust. Especially in those kinds of trusts that will penalise the non-segregation of items on trust, deeming them invalid.
Inter vivos trusts with trustees
In inter vivos trusts with trustees there are two instances of separation of assets. The first one occurs when the settlor transfers his property to the trustee. The second when trustee becomes the legal owner and separates the trust property from his own. The first instance of separation is not disputed in this paper. It will occur ‘naturally’ as the property will have to be transferred to the trustee legally – by deed and registration of trustee as legal owner and by physical delivery in case of chattels[76]. A failure to transfer the property from the settlor to the trustee may result in a failure of a trust but rather for the reason of non-compliance with the formalities requirements and not CSM. The second instance of separation of assets occurs once the trustee has the trust property and decides, or not, to hold it separately from his own assets. If upon execution the two are not separate the court may fail the trust overall[77]. That is in case of tangibles only, but the unfairness of that has already been discussed above. It seems that a better fit for express inter vivos trusts with trustees would be adopting the distribution certainty rather than the traditional CSM with its requirements. It, again, seems unfair that a potentially valid enforceable trust should fail simply because the trustees did not, for any reason, separate their own assets from the trust property. Especially, that it would not pose a threat in case of intangible assets such as shares, debts, and possibly cryptoassets[78]. The current rule of separation and segregation seems like a mean discrimination of tangibles put on trusts. They have to tick one extra box in order to be part of a valid trust. If distributional certainty was adopted for express trusts with trustees, it would be safer for the validity of the trust and less demanding for the trustees. The lion share of the work and effort would be on the settlor’s end as he would have to specify the beneficial share intended for each beneficiary. It seems relatively not risky as the trust is set up within the lifetime of the settlor so if the trustees consider the instruction unclear they may ask the settlor to specify his wishes. And the risk of the trustees ruining the trust by mixing the trust property with their own would be minimised.
Trusts with enough property to give to beneficiaries
In a case where the trustee becomes insolvent his assets will be taken by the authorities, so-called trustees in bankruptcy[79]. But the property he holds on trust for the benefit of someone else will not be taken away as the bankrupt remains a fiduciary and must fulfil his duties towards beneficiaries[80]. The issue here is whether the debtor must segregate the property accordingly in order to be able to pay the creditor upon insolvency and for the fiduciary duty to apply[81]. Meaning, can a rehabilitative trust be established, upon insolvency, for the benefit of the creditors even though the assets in question have not been segregated. The short answer is yes, he has to and no, it cannot. But perhaps, as in the taxonomies discussed above, this requirement may be altered. The following analysis is assuming that the trustee has enough of the assets in his mixed batch to pass on to the beneficiary in accordance with the trust instructions. Which is what could have happened in Re London Wine if there had been enough stock. It was mistakenly considered a case of unmet requirements of CSM when all it was was a case of insolvency. A relationship between creditors who paid for a product (wine) and a service (storage and distribution) and the debtor who provided them. In Re London Wine the bottles of wine were not assigned to each customer, they were not properly segregated and separated from the rest. The court reasoned the trust was not valid as CSM was not clear. And it was right. But only because there was not enough to distribute to each of the customers. To save the company from having to pay back to the customers the court decided there had been no trust. This suggests that if there had been enough bottles to distribute, there could have been a valid trust. At the time this judgment was taken as a serious precedent and just like that a solution suited for one instance became a binding rule in most of the following judgments. An exception was Re Stapylon Fletcher (in receivership)[82] where Paul Baker Q.C. noticed that Re London Wine is a specific, isolated incident that does not impose the segregation requirements on all trusts. He even suggested that Oliver J followed the Sale of Goods Act 1979 too restrictively and did not produce a good judgment. He argued that upon purchase the property should be consider customer’s and not merchant’s, especially if it was paid for and a ‘certificate of title’ was awarded[83]. He argued further that a fiduciary relationship is created between merchants and paying customers[84] Upon receiving the payment the company becomes a fiduciary and appropriate duties follow[85]. In Re London Wine, even if the bottles of wine were not segregated, there were invoices confirming who bought what and in what quantities` as well as the ‘certificates of title’. It seems terribly unjust that customers who paid for a product should lose it because of a requirement resulting from a strict application of the Sale of Goods Act. The decisive evidence of a trust is that receipt of purchase that sets the expectations of the customers and establishes the duty of the company to deliver the promise. The customers should have received that for which they had paid. If that was not possible, a share of the remaining stock proportionate to how much they had purchased, or they should have become owners in common with the merchant. By not respecting that, the company as a fiduciary was unconscionable[86]. The products purchased by the customers – held on trust by the company – were used to the benefit of the company upon their bankruptcy. So, the fiduciary duty was breached, and the company should have been held liable for that breach as an unconscionable trustee would be in any other case.
The most important takeaway from the clash between Re London Wine and Re Stapylon Fletcher is that the former should not be considered a decisive precedent dictating segregation of property bulks. It was a customised approach to the facts and circumstances in the case. Baker Q.C. recognised that and did not follow the separation requirement. Instead he interpreted the Sale of Goods 1979 for the benefit of the purchasers as owners of the property upon payment. This means that the judgments like Re Goldcorp, Hunter v Moss and Mac-Jordan Construction Ltd v Brookmount Erostin Ltd (in administrative receivership)[87] cultivated the wrong precedent and created the redundant inequality between tangibles and intangibles. Which resulted in failed trusts and counterproductive outcomes. When a trustee becomes insolvent, and he had not separated the trust property from his own, it ought not to be used against him and make him an unconscionable trustee. Instead, applying the distributional certainty, the two should be considered distinct. The settlor should ascertain the beneficial share for the beneficiary which would make it clear what property can be taken by the trustee in bankruptcy upon the trustee’s insolvency. As much as it is recognised that separating the two would make things clearer[88] it is also recognised that sometimes people do not do what they need to, they procrastinate and do not fulfil their duties to the requisite standard. This is why the distributional certainty is proposed to prevent trusts failing because of the property remaining in an unascertained batch.
Trusts without enough property to give to beneficiaries and others
The issue with the argument against CSM and above taxonomies is that they will not fully prevent trust failures in cases of insolvency when the trustee does not have enough in the mixed batch to distribute to the beneficiaries. If the trustee’s property is mixed with the trust property there is no control over the latter. It is physically not secured against insolvency of the trustee or his potential unconscionable actions. One of the main benefits of the institution of a trust is that it is supposed to provide protection against the insolvency of a trustee[89]. Even then, the property is not taken away for the benefit of the bankrupt trustee, but it continues to be held for the purpose of the trust[90]. If, however, the trust and trustee’s property are not divided, the trustee’s creditors will not know that some of the property belongs, in fact and in equity, to someone else. It will prove inefficient and even impossible to conduct checks on each and every bankrupt for any potential trust property attached to his personal one, especially in cases of secret trusts. Therefore, the requirement of separation and segregation of a trust has its uses in cases of insolvency without enough to distribute to beneficiaries. If so, the question arises why this requirement only binds tangibles, and not intangible property. Surely, there might arise trusts of shares, debts or crypto currencies where the trustee will not have enough to distribute to the beneficiaries when time comes. English courts have not yet addressed such scenario and created the dichotomy between tangible sand intangibles that was fortunate for Mr Moss, but unfortunate for nosy legal scholars. The Scottish Law Commission, taking a preventative step, proposed a requirement for all solicitors acquiring money from clients must put them in a separate account to prevent unconscionability and potential creditors from taking it away[91]. In England there is no statue requiring that, but common law demands this kind of separation in the form of certainty of subject matter.
It is then hard to determine what should be done in an instance of insolvency when before the trustee in bankruptcy acquires or freezes the assets, there is not enough to give to beneficiaries. Then justice ought to prevail, and creditors should be paid back as decided in Re London Wine. One possible solution is to adopt a procedure similar to tracing. As trustees owe the duty to exercise their powers in the best interest of the beneficiaries[92] so if they do not separate their assets from the trust property it should be considered unconscionable behaviour and a procedure similar to tracing should be employed to retrieve the trust property to be held for the beneficiaries again[93]. And if the whole is not covered by whatever is left, the beneficiary could acquire all of it as a form of restitution. Another option is for the trust beneficiary to become another one of the bankrupt trustee’s creditors and to become party to-be-paid by him as the rest of his creditors. However, that would certainly require more from the beneficiary than is expected from the trustee.
Conclusion
The current law on CSM promotes stability and predictability over justice and settlors’ wishes. The segregation and separation requirement first appeared in Re London Wine. It was then adopted and applied in most of the cases that followed which was not the right interpretation of the judgment. Re London Wine was only ever a solution designed specifically for an insolvency case without enough property to transfer to the creditors. Misunderstanding of this judgment has led to perverse outcomes and the avoidable loss of trust property. As a result, a settlor can never be sure if his wishes will be realised and if his trustees will exercise their fiduciary duties. Why? Because there are different types of trusts held to the same rules and their validity is at the discretion, and mercy, of the courts. This paper proposes several changes, as outlined below, that ensure that trusts, as facilitative devices, are effective and fair. First is the shift of focus from tangibility of assets to their fungibility. It would make up for the discrimination of tangible assets and facilitate disposing of them in the form of trusts. Second is to incorporate the distributional certainty into CSM and to shift the importance from segregation to intention and clear disposition of the beneficial share. It is argued that the distributional certainty already indirectly exists within the legal system and its formal incorporation would only make CSM claims more effective. Third is to treat each trust accordingly to its needs even in instances of lacking segregation of trust property. This taxonomy shows that each type of trust has its own characteristics and needs and, so, should have a tailored approach to CSM. For most of the trusts, explained in the above taxonomy, intention with ascertained beneficial share prove sufficient for their creation as valid trusts on the proviso that there is enough property to give away to beneficiaries. For self-declared trusts, this is the standard rule but it is not accepted for the other trusts. The only instance where segregation should apply is where a trustee does not have enough property to give away to beneficiaries breaching his fiduciary duties. The segregation requirement does make the trust property more certain, but it does not cure trusts. What can cure trusts is the customised view of the new taxonomy prioritising the true purpose of trusts: realising the true intention and wills of settlors as they divide their assets amongst their beneficiaries. And, where this intention is unclear, common sense must prevail.
[1] (1840) 3 Beav 148
[2] [1986] PCC 121
[3] [1995] 1 AC 74
[4] [1994] 1 W.L.R.452
[5] https://lawcom.gov.uk/project/modernising-trust-law-for-a-global-britain/
[6] Hayton, Uncertainty of subject matter of trusts, Law Quarterly Review, 1994, p.336
[7] Dillon LJ in Hunter v Moss, 458-459
[8] Chung, Challenging the orthodoxy: a critique of Re Goldcorp and the English approach to the certainty of subject matter, 2019, p.486
[9] Dilnot, Harris, Ownership of a fund, 2012, p. 272
[10] [1994] 1 W.L.R. 452
[11] Jones, A., Creating a trust over an unascertained part of homogenous whole, 1993, p. 467
[12] Hunter v Moss [1993] 1 W.L.R. 934, 940
[13] Mann LJ in Hunter v Moss, 457
[14] Chung, 2019, pp.482-483
[15] Goode, Are tangible assets fungible?, 2003, pp. 379-380
[16] Ibid, pp.382-383
[17] Ibid, pp. 382-383
[18] Ibid, p.383
[19] Ibid, p.384
[20] Ibid, p.384-385
[21] Chung, 2019, p. 487
[22] Goode, 2003, pp. 381-382
[23] Dilnot, Harris, 2012, p. 273
[24] Worthington, Sorting out ownership interests in a bulk: gifts, sales and trusts, 1999, pp.3-4
[25] Ibid, p.6
[26] Ibid, p.20
[27] Chung, 2019, pp.482
[28] Chung, 2019, pp.482
[29] Re Goldcorp [1995] 1 AC 74, 99-101
[30] Worthington, 1999, p.6
[31] Bryan J in AA v Persons Unknown [2020] 4 WLR 35, 6
[32] Chung, 2019, p.484
[33] Ibid.
[34] LBMA, Current membership,< http://www.lbma.org.uk/membership/current-membership#-> accessed 16 April 2024
[35] Chung, 2019, p.484
[36] Ibid, p.485
[37] Ibid.
[38] Wilde, The three certainties required to declare a trust – or is it four? “Distributional certainty”, 2020, p.349
[39] Ibid, p.350
[40] Ibid, pp.349-350
[41] Ibid, p.351
[42] [1971] AC 424
[43] Ibid, 431, 451
[44] (1840) 3 Beav 148, 71
[45] As already required in McPhail v Doulton.
[46] [1952] EWCA Civ 4
[47] Jaffey, Private property and intangibles, p.58
[48] Hart, Concept of Law, 2012, pp. 26-27
[49] Ibid, pp.27-28
[50] Floud, Johnson, The Cambridge Economic History of Modern Britain, 2004, pp.20-24
[51] Ibid.
[52] Fairfield, OWNED Property, privacy and the new digital serfdom, 2017, p.94
[53] Liew, Trusts: modern taxonomy and autonomy, p.27
[54] Applying the minimal threshold for certainty (McPhail v Doulton).
[55] Pawlowski, Testamentary Trusts and The Rule Against Capricious Purposes: an Underlying Rationale, 2012, p.109
[56] Ibid, p.226
[57] Ibid, p.117
[58] Coshott, To benefit another: a theory of the express trust, 2020, pp. 231-232
[59] Pawlowski, 2020, p. 225
[60] [2002] EWCA Civ 227
[61] Agnew and Douglas, ‘Self-Declarations of Trust’, p. 4
[62] Ibid, p. 1
[63] [1977] 1 W.L.R. 527
[64] Ibid, 531
[65] Agnew and Douglas, p. 7
[66] (1854) 2 Drew 221
[67] Agnew and Douglas, p. 12
[68] Agnew and Douglas, p. 17
[69] [1996] Ch 274
[70] Ibid, 284
[71] Ibid, 282
[72] Ibid, 285
[73] Agnew and Douglas, p.2
[74] Agnew and Douglas, p.17
[75] Low, Trusts of cryptoassets, 2021, paras. 192-193
[76] Coshott, 2020, p. 222
[77] Zhang, Certainty of subject matter: what China can learn from English trust law, 2018, p.1028
[78] Georgieva, Prized assets in a brave new (blockchain) world, 2020, p. 10
[79] With duties outlined in art. 305 Insolvency Act 1986.
[80] Swadling, Trusts and Ownership: a Common Law Perspective, 2016,p. 25
[81] Galoob, Intentions, compliance and fiduciary obligations, 2014, pp.115-116
[82] [1994] 1 W.L.R. 1181
[83] [1994] 1 W.L.R. 1181, 1193
[84] [1994] 1 W.L.R. 1181, 1201
[85] Galoob, 2014, p. 109
[86] Chung, 2019, p. 489
[87] (1992) 56 BLR 1 CA
[88] Norris, Uncertainty and informality: Hunter v Moss, 1999, pp. 44-45
[89] Scottish Law Commission’s Report on Trust Law 2014, p.10
[90] Ibid.
[91] Ibid, p.11
[92] Galoob, 2014, p. 109
[93] Worthington, 1999, p. 23
Other essays
Since 2003 our land registration system has been dictated exclusively by the Land Registration Act 2002 and, according to Martin Dixon, has been a fantastic success[1]. The Act changed the previous system, of unregistered land, to a more modern one. Its goals were to inform the public about the ownership of land in the country, satisfy the publicity principle[2], provide transparency concerning land registration[3], prevent fraudulent transactions of land and facilitate those transactions through electronic conveyancing[4]. These were achieved to an extent, however, there remain remnants of the feudal system that neither the Legislature nor the Law Commission have efficiently tackled, such as the law on adverse possession[5]. There are concerns about the new registration system that have not been properly dealt with, for example the accessibility of the register under the Freedom of Information Act 2000[6] or the processing times of applications made to the HMLR[7]. These shortfalls of the LRA 2002 were so concerning that the 2022 edition of The Conveyancer was subtitled “what’s wrong with land registration”[8]. The debate about land registration and the Act persists and no one appears fully satisfied with the system or the Act’s provisions[9]. However, in this essay, it is argued that the wide-ranging criticism is not an expression of discontent with the LRA and the post-2003 reality, but a strive to make the system fit for purpose in the modern world.
The first goal of land registration, per its name, is to register land[10]. This seems a reasonable success since the proportion of registered freeholds in England went from roughly 50% in 2005 to 85% in 2021[11]; though land registration was made compulsory in 1990[12]. The remaining 15% belongs to the Crown and Church[13] which will likely remain unregistered as it is not exposed to the triggers requiring registration[14]. The second goal was to comply with the publicity principle[15]. Though it is not stated in the Act, it is an implied principle underlying the system as a whole[16]. There’s a presumption, adequate or not, that land is a national resource[17]. It is certainly scarce and limited[18]. Thus, the public demands access to knowledge of the state of ownership in the country[19]. Access to the information is questionable, hidden behind a payment of £3, and some applications for information may be denied and the fee kept[20]. Guy Shrubsole believes firmly that gratis is a fair price, not only for people within the land law bubble, but for any citizen concerned with the use of their homeland. Arguing that, under the Freedom of Information Act, the information should be freely given[21]. These fees generate revenue (£318.9 million between 2021 and 2022)[22] for the HMLR and pay for indemnities. The use of land should be verified and transparent given the scale of farming subsidies[23]. Anna Powell-Smith, a technologist, is creating a map of all land in England, her own Land Registry very similar to continental European cadastres. She has been gradually filling in the map, over 10 years, with the information sporadically flowing into the register. Should this not be undertaken by HM Land Registry? This knowledge surely benefits the landowners. Clarke argues that, if the name of the owner is public, it is easier to protect the property from unwanted guests such as flower pickers. It seems unreasonable to penalise invaders for infringement of rights without making those rights known. Of course, in some instances, those invaders may seek to cause harm. They may be fraudsters, impersonating the owner of the property, trying to sell or mortgage it to a third party. The buyer becoming victim to registered title fraud[24]. The land registration system aims to reduce those incidents. The Law Commission, in 2002 and 2018, stated that this required two things. First, to have a complete register, with an accessible record of every landowner who can be verified by potential buyers. This is not possible for the reasons explained. Second, conveyancers or assigned bodies must bear additional duty of care to verify the identities of clients and root out fraudsters before they can act. Lamentably, the 2018 Law Commission Report confirmed an increase of fraudulent activity relating to registered land. The cost of indemnity payments between 2008 and 2018 was £58,000,000. For the last three years the indemnity costs approached £6,000,000 a year, including costs related to fraud. The mission of fraud prevention, then, cannot be considered successful. Lastly, e-conveyancing is absent and transactions are made on paper. This is a sad indictment of a leading country entering the third decade of the 21st century. The absence of e-conveyancing has to do with the safety and integrity concerns of conducting online transactions, despite the 2021 standard for digital identity checks[25]. Conveyancers also use a unique identifier and attest to the accuracy of information lodged[26]. Overall, land registration has made small improving steps in terms of its four missions but there has been no fantastic success.
The aspect of the 2002 Act most in need of rethinking is the law on adverse possession. Meaning both squatter’s rights and the matter of adjoining estates with general boundaries. The purpose of adverse possession is to make sure that land is used at full capacity[27]. Pre-2002, after 12 years of actual occupation, an adverse possessor could claim land as his own solely by possession[28]. The 2002 Act changed that into adverse possession being attainable only by registration. LRA 2002 says that a person can apply for a registered adverse possession after 10 years of actual occupation ending on the day of application[29]. The issue is that few people realising they’re occupying somebody else’s land will register. A squatter will not be willing to claim the land he’s occupying if it leads to opposition and litigation. A neighbour whose fence runs over onto his neighbour’s land will not want to register that change of the boundary because it may agitate and, again, lead to litigation[30]. Under sch.6, an adverse possessor has to reasonably believe that the land he’s occupying is his[31]. Then, after 10 years, he has to stop reasonably believing and make an application to register within 12 months of stopping that reasonable belief[32]. This was illustrated in Zarb v Parry[33]. There are issues with this reasoning in the Act and after the 2018 attempt to correct it. Firstly, the attainability of land in adverse possession only through registration is not effective due to adverse possessors’ unwillingness to expose themselves to the risk of legal battle and losing their occupied land. Moreover, this provision disallows utilising the land because the ability to use may be taken away if unregistered. This shows the vicious circle of procurement by registration, unwillingness to so do, and not being able to fully, and legally, use the land. The second issue concerns identifying reasonable belief, the time of identification and the reliability of anybody stating anything about reasonably believing anything[34]. Nick Hopkins, an author of the 2018 Report, says that a person adversely possessing land must reasonably believe he rightfully owns it, then he must cease to believe it and only then claim it[35]. Elizabeth Cook highlighted that it is practically impossible to achieve registration with this[36]. That is because people lie and get confused about when they realised what. It is objectively impossible for the court to say whether a claimant had a belief and if it was reasonable. Hopkins tried to place the reasonable belief under the umbrella of intention but was rebuked; the latter is objective and the former is not[37]. This rule introduces morality into the Land Registration Act which, according to Dixon, it should not do. It should be about who is using the land in the most efficient way. If it is through unregistered adverse possession, it should remain so[38]. A practical example of this approach, though from the perspective of title by registration, is Walker v Burton[39]. Mummery LJ said it is better that land is owned by proprietors in physical possession, even if they so became by mistake, then not owned and managed at all[40].
Another issue with the post-2002 registration compulsion arises when a general boundary between two neighbouring plots of land existed before registration. The 2002 Act and the 2018 Review suggest that a real boundary is better for the registry than a general boundary, which has functioned unchanged since 1875[41]. It is said to minimise arguments and litigation over any unclear separation of land[42]. However, if two neighbours register their land and set a real boundary, they may discover that they do not agree about it and that may breed disagreements and escalate to litigation[43]. Sedley LJ observed that these claims, usually involving land of little value, racked up legal fees that exceeded the value of the land being fought over[44]. On the other hand, the general boundary may be fatal for cases in commercial use of land[45]. The case of Choiceplace Properties Ltd[46] shows that the boundaries of land subject to planning permissions must be real. Otherwise, the development may be ruled unlawful. There is a clear dichotomy between the general and real boundary, and which one is better suited for the law of property. For private neighbourly disputes, the general boundary is more fit for purpose and may limit the cases flooding the courts[47]. For commercial use of land, given planning and development permissions, the real boundary is best[48]. The 2002 Act does not address this clash. The Law Commission identified a need for change and directed it to the Land Registration Rules 2003, but not much has been done. David Sawtell suggested to not ‘fix’ the existing boundaries, so not to change general boundaries into real ones, as it may create more disputes[49] but instead to implement a practise of setting precise boundaries for all newly divided plots[50]. This way the historical general boundaries between neighbours won’t be artificially fixed and real boundaries between commercial pieces of land will be fit for development.
Lastly, the issue of accessibility of the register becoming the most serious problem of HMLR. There are two kinds of entries made to HMLR: straightforward and “complex” ones[51]. The latter includes first registrations and 50% of them are completed within 13 months, the rest within 15 months[52]. Applications to divide existing titles, which are considered more complex, are completed within 21 months[53]. An outrageous waiting time, nearing two years, for an essential transaction concerning a valuable resource. With straightforward entries, one third are completed within one day. Another third takes between a month to a few months[54]. HMLR said that increasing speed is a priority. Indeed, it increased by 12% from 2022 to 2023 owing to hundreds of caseworkers joining each year[55]. An optimistic excuse may be the success of the registry in encouraging so many landowners that the system overflows. Or, perhaps, the budget shortfalls, inefficiency and staffing shortages are behind the disappointing service of HMLR[56]. This is significant and should be dealt with in order to restore the reputation of HMLR as the leading land registry amongst the common law jurisdictions[57].
Ultimately the experiment, started in 1862, developed in 1925 and 2002, though imperfect, is considered a success. Achieving its purpose of completing the register and satisfying the connected publicity principle. It introduced methods of preventing fraud and recognised the need for e-conveyancing. Indeed, the law underlying adverse possession needs revising and the four goals should be fully satisfied to create a system fit for purpose. The focus of 2025 ought to be in expediting the virtual queues to HMLR and rethinking the facilitation of adverse possession rather than getting rid of it. It should be understood that making the feudal system redundant will be a long and costly process. England will likely struggle with adapting its ancient laws related to land. However, the enormous transition from 1862 to today proves that changes can be made and that which needs fixing can be.
[1] Dixon, Updating Land Registration Act 2002,2018
[2] Clarke,2020,p.595
[3] Law Commission,2018
[4] Clarke,2020,pp.596-599
[5] Hopkins et al,2018
[6] Shrubsole,2023, p.36
[7] Dixon,2023,para.213
[8] Ibid.
[9] Dixon,2023, para.122
[10] N2
[11] Powell-Smith,2019
[12] Conveyancer,1993,para.101
[13] Ibid,para.104
[14] Cooke,2002,para.13
[15] N10
[16] Ibid,p.596
[17] Shrubsole,2023,p.22
[18] Ibid.
[19] N15
[20] Shrubsole,2023,p.41
[21] Ibid.
[22] Lennox,2022
[23] N11
[24] N3
[25] Davies,2021
[26] Ibid.
[27] Cooke, Updating Land Registration Act 2002,2018
[28] Ibid.
[29] Ibid.
[30] Ibid.
[31] Hopkins in Updating Land Registration Act 2002,2018
[32] N6
[33] [2011]EWCA Civ 1306
[34] N8
[35] N7
[36] N10
[37] N27
[38] N1
[39] [2013]EWCA Civ 1228
[40] Dixon,2013,paras.465-467
[41] Sawtell,2023,para.141
[42] Ibid.
[43] N12
[44] Sawtell,2023,para.140
[45] Ibid.,para.142
[46] [2021]EWHC 1070(Admin)
[47] N12
[48] N15
[49] N19
[50] Ibid.
[51] N35
[52] Ibid., para.214
[53] Ibid.
[54] Ibid.
[55] Ibid.
[56] Ibid.
[57] Ibid.,para.215
Gross Negligence Manslaughter (GNM), as set out in Adomako[1], consists of five requirements. The most crucial, what distinguishes GNM from other types of manslaughter, is duty of care, meaning the defendant must owe the victim a duty of care[2]. There can be no conviction of GNM without this. This is the focus of this essay. The concept of negligence appears in other areas of law, for example tort, however in criminal law it is more severe and amounts to conviction[3]. Greenberg defines criminal negligence as ‘criminal culpability’[4]. A defendant is criminally liable when they knowingly take risks that may cause harm or death and then, does nothing to avert the harm[5]. In the case of GNM, for the defendant to be convicted, there must be an established duty of care[6]. According to Crosby, there are six possible types of duty of care: statutory, public, contractual, special relationship, voluntary imposed duty, and, finally, a duty owed resulting from the creation of a dangerous situation[7]. This essay argues that courts willingly rely on the principle of ‘creation of danger’, deriving from Miller[8], to impose a duty of care in GNM cases. This essay examines how the courts establish duty of care in GNM cases and considers the adequacy of the Miller principle. The English method will be compared against other criminal codes, concluding that imposing a criminal liability in GNM cases is unjust and unclear for the courts and its citizens and possibly infringes upon the principle of legality under art.7 of the European Convention of Human Rights 1950 (ECHR).
Courts have willingly relied on the principle of creating danger, by act or by word, since the judgment in Miller in 1983. As per Lord Diplock, the principle states that when a person creates, or contributes to creating danger, there is no reason they should not be held criminally liable if they fail to take measures within their power that would counteract that danger[9]. Professor Glanville Williams outlines that the essence of the rule is to punish a result that the law forbids. Even if the result is a consequence of an accident, the mens rea conceived after that act, before the result, while the danger could have been prevented, the person will be held liable. Liable for failure to act[10]. This principle has been adopted in other areas of criminal law like assault and GNM. An example of its application is Santa-Bermudez[11]. Justice Kay concluded that the defendant should be convicted because he created a dangerous situation by his word and exposed the victim to foreseeable risk of injury by saying there was nothing in his pocket and that it was safe to inspect them[12]. Justice Kay cited Lord Diplock in Miller that grounds exist to hold someone criminally liable if they fail to attempt to counteract danger they create[13]. In Wacker[14], it was ruled that the principle remains even if the victims are aware of and agree to the danger connected with the act. Lord Kay said Wacker owed all 58 immigrants a duty of care even though they agreed to be put in a container with a limited access to fresh air[15]. Therefore, GNM convictions can be upheld even when victims in potential danger consent to that danger, as seen in the Court of Appeal (CA) judgment in Bowler[16]. Here, the conviction was upheld as the defendant and the victim were acting ‘in concert’[17]. The concept assumes that both parties create the dangerous situation together[18]. Bowler was convicted because he failed to check on his partner and failed to release him from the tape[19]. Further evidence that the CA rely on the principle of the creation of danger rather than, say, the prior special relationship[20] of two men who had known each other for years.
A more recent case showing the courts’ tendency is an unreported case of Michael Bowditch from 2017[21]. The court convicted him of GNM even though it was very difficult, if not impossible, to establish a duty of care and there was no evidence of him having created or having contributed to creating danger[22]. The court ruled to impose the duty of care based on the principle of creating danger referring to Evans and the scope of the principle there[23]. The criticism of that judgment is that it is moving the GNM convictions toward a ‘Good Samaritan Law’[24]. Crosby argues that by convicting Bowditch the courts demonstrated that a moral duty is prevalent over a substantive, legal duty[25]. This approach to a duty of care in GNM favours social, moral responsibility over personal autonomy[26]. This leads to imposing a ‘Good Samaritan Law’ on potential defendants. There would be nothing wrong with that if the law was clearer in that regard[27]. As of now, the imposition of duty of care resulting from creating danger differs from case to case[28]. In Kennedy it was not enough to convict the defendant, but in Evans it was. The law is uncertain and therefore unjust to potential defendants, as they can never be sure how their case will be considered, which is against one of the main principles of the rule of law – the principle of legality[29]. The French Criminal Code is an example of how to avoid that breach[30]. There is a statutory duty to rescue and when breached, does not amount to a homicide conviction, but a milder one[31]. That could be implemented in English criminal law to ensure compatibility with the principle of legality and execution of law, not discretion[32].
There is a special type of manslaughter cases - ‘drug cases’[33]. A leading one concerning the principle of creating danger in establishing a duty of care is Evans[34]. The duty of care was established basing on ‘undisputed’ and ‘disputed’ facts[35]. The former were that both the appellant and her mother were at the premises at the relevant time; the appellant witnessing her sister taking the drugs and feeling bad after that; the appellant knowing about the condition of the sister and her responsibility for the care of her after she had injected herself[36]. However, those were not enough to establish a duty of care between the defendant and her sister[37]. It was said that the duty cannot be established based on a special relationship, and therefore the only acceptable ground was the principle of creating danger, which was the ‘disputed’ fact[38]. The issue with applying the principle of creating danger in Evans is that the justices were reluctant to call Evans a drug dealer as in their view she was a mere supplier[39]. This judgment was met with criticism based on the fact that the defendant was not really a drug dealer, therefore she did not create danger, nevertheless the principle of creating danger remained the overriding factor in establishing the duty of care[40]. Without it, there would be no ground for conviction. Evans was the first case that began the debate concerning imposing a general duty of care on all drug dealers[41]. It would impose a standard of care expected of suppliers towards their customers[42]. The issue with this proposal was that it would be difficult to specify how a careful supplier acts and what he can do to prevent harm[43]. The idea has been debated by the Parliament, however, was never enforced[44]. Nevertheless, Evans remained a controversial and highly criticised case.
The judgment becomes even more controversial when compared to Kennedy No 2[45] (Kennedy). The prosecution tried to prove that Kennedy created a dangerous situation by supplying a drug to his friend who later died of an overdose[46]. However, the victim self-administered the drug, breaking the chain of causation, and Kennedy was not convicted[47]. In Evans, however, the defendant was convicted of GNM even though the victim had injected herself[48]. The liability for GNM relied on the failure to get help, the defendant’s omission[49]. Some suggest that the ‘disputed fact’ argument in Evans was artificial and was simply used to punish a drug user[50]. Indeed, Williams argues that the court sought punishment for her behaviour, finding the offence and matching it with a principle rather than the other way round[51]. This is evidence of a willingness to use ‘creating danger’ in establishing a duty of care to impose a duty on the defendant; a moral duty rather that a legal or statutory one[52].
By willingly using the principle of creating danger in establishing a duty of care in GNM cases, the courts impose a moral duty on the defendants[53] who are then convicted of homicide[54]. This can be seen in Evans, Bowditch and Broughton[55]. In Broughton, the defendant, initially convicted of GNM, appealed that the judge misdirected the jury, and that no duty of care arose based on material facts[56]. The duty of care had been established based on taking drugs together, acting ‘in concert’[57]. The prosecution were unable to provide evidence that the defendant had created danger[58]. The appeal was allowed, and the conviction quashed[59]. This is demonstrable evidence that the courts have used ‘creating danger’ to unjustly convict GNM leaving defendants with a record of homicide.
As the principle of establishing the duty is unclear and uncertain, it is unfair and unjust to punish citizens for homicide without a certain law[60]. Moreover, citizens are often unaware of the law and the possibility of being liable for GNM as a result of failing to help[61]. Lord Bingham argued that criminal law should be capable of guiding people and tell them what is prohibited and what is expected of them[62]. Compared to other legal systems, English criminal law lacks certainty in difficult cases such as GNM. In contrast, the Spanish Criminal Code provides a separate offence of failing to provide assistance, where a defendant is able to help[63]. Therefore, those who fail to help, will be convicted of a milder offence, usually up to three-year imprisonment, than a homicide offence, up to five-year imprisonment[64]. As the milder offence is codified in statute, the law is clearer, and there is a smaller risk of violating art.7 ECHR[65]. Creating this type of legislation would also prevent courts from the tendency to look for a result and later find principles to support them[66].
In cases such as Wacker, Bowler, Bowditch and Evans, evidence suggests that the courts willingly apply the principle of creating danger in order to establish a duty of care despite the violation of the principle of legality under art.7 ECHR. Ultimately, then, we can say that the moral approach to criminal law is unjust. It would be fairer if there was a statutory duty to rescue that would not amount to a homicide crime.
[1] Adomako [1995] 1 A. C. 171
[2] Crosby, 2018, p.130
[3] Laird, 2017, p.7
[4] Greenberg, 2021, p.490
[5] Ibid.
[6] Ibid.
[7] Crosby, 2018, p.130
[8] [1983] 2W.L.R.539
[9] Lord Diplock in Miller para.176
[10] Baker, 2010, para.310
[11] [2003] EWHC 2908
[12] Santa-Bermudez p.7
[13] Lord Diplock in Miller para.176D
[14] [2002] EWCA Crim 1944
[15] Wacker para.12
[16] [2015] EWCA Crim 819
[17] Bowler para.307
[18] Elliot, 2006, p.988
[19] Bowler para.313
[20] Crosby, 2018, p.130
[21] Ibid., p.128
[22] Ibid. p.130
[23] Ibid. p.132-133
[24] Ibid. p.134-135
[25] Ibid. P.129
[26] Ibid.
[27] Bernal, 2017, p.195
[28] Ibid. p.196
[29] Ibid.
[30] Tomlinson, 2000, p.476
[31] Ibid. p.452
[32] Bingham, 2011, p.48
[33] Elliot, de Than, 2006, p.986
[34] [2009] EWCA Crim 650
[35] Williams, 2009, p.636
[36] Evans para.12
[37] Baker, 2010, p.318
[38] Crosby, 2018, p.134
[39] Williams, 2009, p.459
[40] Crosby, 2018, p.133
[41] Elliot, de Than, 2006, p.985
[42] Ibid.
[43] Ibid, p.986
[44] Ormerod, Fortson, 2005, p.828
[45] [2007] UKHL 38
[46] Baker, 2010, p.313
[47] Ibid.
[48] Ibid, p.310-311
[49] Ibid, p.317
[50] Williams, 2009, p.640
[51] Ibid, p.462
[52] Ibid.
[53] Bernal, 2017, p.191
[54] Baker, 2010, p.312
[55] [2020] EWCA Crim 1093
[56] Storey, 2021, p.62
[57] Lord Burnet in Broughton at 1
[58] Ibid., at 65-67
[59] Ibid., at 104
[60] Bernal, 2017, p.196
[61] Ashworth, 2015, p.577
[62] Ibid., p.565
[63] Bernal, 2017, p.195
[64] Ibid., p.195-196
[65] Ibid., p.196
[66] Ibid.
Secret trusts arise when a settlor entrusts his legacy to a trustee for his chosen beneficiary without disclosing it in his will[1]. Historically, settlors have done this in order to conceal mistresses or illegitimate children from family and friends[2]. Secret trusts also result from the indecisiveness of a settlor who is unsure about the inheritors of his property[3]. He may want to delay decisions or may wish to distribute his legacy without public knowledge[4]. Stereotypically, secret trusts are construed to be rarely enforceable, unreliable and avoidable[5]. There is debate concerning the validity and usefulness of secret trusts in this era which prioritises transparency and abhors duplicity. Despite the reasons for the existence of secret trusts, such as privacy, it is argued that they have become legally, morally and doctrinally bankrupt. This essay will consider whether secret trusts have become dated, reaching the end of any useful life that they might have had and whether they should be removed from practise. It will be argued for their removal for three reasons. First, the fraud theory and the dehors the will theory do not provide a convincing enough explanation to justify their existence[6]. Second, secret trusts mostly come to light because of the unconscionable conduct of legatees; diminishing their moral aspect[7]. Third, the desire for privacy is not granted by the doctrine of secret trusts and should be achieved in clearer ways[8], thus secret trusts are doctrinally bankrupt and redundant[9].
Firstly, two theories justifying the existence of secret trusts[10]. The fraud theory states that equity will not permit statute to be an instrument of fraud[11]. To do that, s. 9 Wills Act 1837 is overlooked and is not used as it would be in cases of testamentary dispositions[12]. This section outlines the requirements of constructing a valid will; it has to be in writing, signed and witnessed[13]. Secret trusts are enforced without satisfying these requirements[14]. If they were not enforced, against the statute, the trustee would benefit instead of the secret beneficiary which would amount to fraud as per Lord Westbury’s orthodox view of fraud[15] in Cullen v Attorney- General for Ireland[16]. This theory can only apply, if at all, to fully secret trusts as, with half-secret trusts, the existence of a trust is written in the will and there is little possibility of fraud. It is argued that fraud theory is legally unconvincing as it makes judges go against the statute because of the broad definition of fraud. Lord Buckmaster in Blackwell v Blackwell[17] explained that fraud theory is meant to prevent potential fraud that a trustee might commit by not carrying out his promises[18]. This is an overstretched definition of the orthodox view of fraud as simply a dishonest gain of a trustee[19]. This means that the courts disobey the Act of Parliament to cover the possibility of secret trustees not fulfilling their promises[20]. Promises and mere agreements do not usually amount to legally enforceable trusts[21]. They require elements that would make them legally obligatory such as the three certainties or formalities[22]. Therefore, Equity allows enforcing mere moral obligations[23] that should not be convincing enough to justify breaching the Wills Act. This can be compared to the New Zealand jurisprudence and Hammond J’s judgment in Brown v Pourau[24]. He denounced fraud as a requirement for secret trusts, established that they are express trusts and that requirements for such should be enough to recognise and enforce secret trusts[25]. Moreover, fraud theory does not serve its purpose anymore because of the threshold set by the courts to successfully enforce secret trusts[26]. Lords Westbury and Hatherley said in McCormick v Grogan[27] that such trusts may be enforced in “clear, even criminal” instances[28]. The standard of proof, although in a civil procedure, is almost beyond reasonable doubt[29]. Thus, it is hard to prove and harder still to enforce. Fraud theory, then, is unsound in two ways: it adopts a definition of fraud that is too broad to successfully justify secret trusts and it sets a threshold that is too high to later enforce those trusts.
The dehors the will theory says that secret trusts do not conflict with s.9 Wills Act 1837 because they operate outside of the will[30]. They could not do that if they were only testamentary trusts as then they would fall under the will and Wills Act[31]. But they are not; they are also inter vivos[32]. All secret trusts come into effect after the death of the trustor[33], however the matter of communicating the trust by the settlor to the legatee is different for fully secret trusts and half-secret trusts[34]. With secret trusts, communication and acceptance between settlor and legatee must take place during the lifetime of the trustor[35]. This suggests that fully secret trusts are inter vivos; they would not exist but for that communication before the settlor’s death[36]. With half-secret trusts, communication and acceptance, as per Blackwell v Blackwell, takes place during or before signing the will[37]. They are more testamentary than inter vivos. However, the settlor must intend to create the trust during his lifetime and must make preparations for the communication prior to his death[38]. Otherwise, the trust will never be enforced because it simply would not be created. Secret trusts, then, are both inter vivos and testamentary because of this. This means that they are not fully outside the will and that ‘dehors the will’ is factually untrue [39]. A more accurate name would be ‘dehors the Wills Act’ as there is no mention about secret trusts in the statute[40]; secret trusts operate outside of it. The formalities required by s.9 are, then, not applied[41]. However, some argue that the status quo is widely accepted by the Legislature[42]. The Statute of Frauds 1677, highlighting the formalities requirements for testamentary dispositions, was amended by the 1837 Act. This change in legislation was an opportunity for Parliament to incorporate secret trusts into statute and to dispel uncertainty. They chose not to[43] and therefore, while passing the amended legislation, accepted the doctrine of secret trusts again. Lord Nottingham summarised the practise of enforcing secret trusts as a “constant course of this court”[44]. It seems that Parliament, informally, does not perceive it as a breach of statute. Even so, it is legally unpalatable for the courts to go against legislation openly and repeatedly[45]. To bring this to terms and, for the sake of the certainty of the law, a section should be added to the Wills Act 1837 that the requirements of s.9 are not applicable to, or expected from, fully and half-secret trusts. Without this, enforcing secret trusts against the Act is against the Parliamentary Supremacy rule and the dehors the will and Wills Act theories do not constitute a good justification.
Secondly, judges, scholars and law students learn about secret trusts because secret trustees have acted unconscionably and immorally[46]. Secret trusts are usually discovered and verified because trustees have been caught attempting to commit fraud. They might otherwise never be known. Fraud theory, then, is reactionary. It does not prevent fraud[47]. A secret trustee may act unconscionably, but if no one else - like a secret beneficiary – knows about the existence of the trust, it cannot be verified. There may be thousands of secret trusts or none[48]. This suggests that fraud theory is ineffective[49] and any further research of secret trusts, ultimately, is dependent on moral misconduct[50]. So, the doctrine, resolution and application of secret trusts relies on unconscionability and immorality. In other words, the law is dependent on immorality. If secret trusts were somehow tracked, this may at least soothe the confusion and debate of their validity and usefulness.
Lastly, secret trusts have become doctrinally bankrupt. Secret trusts grant privacy in relation to wills but, in this era of transparency, all wills are public domain and there is an assumption of a right to know their full details and reasoning[51]. Some testators find this uncomfortable[52]. Secret trusts, with their low requirements, may have seemed like an adequate tool to give trustors some privacy in disposing their property[53]. However, there are issues, especially with half-secret trusts, in their ability to grant such privacy under the scrutiny of the court[54]. With half-secret trusts, they are typically drafted with the assistance of a solicitor[55]. Thus, the solicitor knows about the agreement and holds documents concerning it. It was decided in Larke v Nugus[56] that solicitors must give “full and frank” information to anyone having an interest in contesting or scrutinising a will[57]. This means that the information, meant for the solicitor, can be surrendered. Moreover, Buckley J said in the same case that a solicitor, being a material witness in a dispute regarding a will, is obliged to reveal all relevant information concerning that will, including circumstances leading to its creation[58]. If a testator, A, for example, used a solicitor in creating a half-secret trust for the benefit of his illegitimate children, and that was contested by his legitimate children after his (A’s) death, the solicitor would surrender the documents and details about the trust. This defeats the purpose of the secret trust and acts against the wishes of the testator. With fully secret trusts, the “Larke v Nugus letters” would not apply as the solicitor drafting a will might not be aware of a fully secret trust existing[59]. However, through litigation, the court may order disclosure of confidential communications between settlors and trustees[60]. Secrecy is not always preserved in cases of fully and half-secret trusts. For the purposes of transparency, secret trusts are, then, doctrinally bankrupt and ineffective because they do deliver a guarantee of privacy.
Even in this era of transparency, everyone is entitled to some degree of privacy and there is no harm keeping or having secrets so long as they remain lawful and do not confuse the interpretation and execution of a will. In 2003, only 29% of surveyed solicitors stated that their clients would choose secret trusts as a preferred type of property disposition; 52% declared they would choose other devices granting them confidentiality[61]. Indeed, privacy can be achieved through other settlements[62]. For instance, an inter vivos settlement is constituted when the owner of the property wishes to settle it during his lifetime by creating a trust which he retains only a life estate; the remainder left to his decided beneficiary[63]. Such settlement confers an immediate proprietary interest onto beneficiaries that can be altered either in accordance with the terms of the trust or with the consent of all beneficiaries[64]. Using inter vivos settlements would work well to replace fully secret trusts, however, it would not address half-secret trusts used by testators because of their indecisiveness in dividing their property. Meaning, trustors should either be more resolute in their wills or some alternative, allowing indecisiveness but not uncertainty, should be invented.
To conclude, secret trusts are legally, morally and doctrinally bankrupt for three reasons. Two theories used to justify them are unsound and unconvincing. The fraud theory adopts too broad a definition of fraud and demands an almost criminal standard of proof in proving the existence of the trust. The dehors the will theory cannot apply to secret trusts because they are not fully outside the will. Its possible substitution, dehors the Wills Act theory, could be effective however unachievable until Parliament acknowledges secret trusts in the statute. Secret trusts are morally bankrupt because they make the law depend on immorality and unconscionability of secret trustees. Lastly, they are doctrinally bankrupt because they cannot provide full confidentiality of testamentary dispositions. All in all, leaving this world, we cannot take anything out of it and must dispose our property. For the reasons discussed above it would be best not to do so using secret trusts.
[1] Stubbins,2015,p.819
[2] Kincaid,2000,p.420
[3] Meager,2003,p.207
[4] Evans,2014,p.229
[5] Meager,2003,p.209
[6] Griffin,2017,p.373
[7] Allan,2015,p.342
[8] Meager,2003,p.206
[9] Griffin,2017,p.373
[10] Allan,2011,p.312
[11] Ibid.,p.328
[12] Ibid.,p.312
[13] Ibid.
[14] N10
[15] Siu,2005,p.79
[16] (1866)L.R.1 H.L.190
[17] [1929]UKHL 1
[18] Wilde,2020,p166
[19] Ibid.
[20] Wilde,1995,p.369
[21] Griffin,2017,p.376
[22] Ibid.
[23] Rickett,1996,p.308
[24] [1995]1N.Z.L.R.35
[25] N23
[26] Allan,2011,p.325
[27] [1869]LR.4.H.R.82
[28] Ibid.,pp.317-318
[29] Rickett,1996,p.307
[30] Wilde,2020,p.170
[31] Ibid.,p.164
[32] Kincaid,2000,p.422
[33] N10
[34] Ibid.
[35] Ibid.
[36] Critchley,1999,p.633
[37] N10
[38] Critchley,1999,p.632
[39] Wilde,2020,p.167
[40] Ibid.
[41] Ibid.,p.173
[42] Ibid.,p.170
[43] Ibid.,p.171
[44] Allan,2011,p.329
[45] Griffin,2017,p.374
[46] Critchley,1999,p.647
[47] Chung,2018,p.163
[48] Davies,2016,p.127
[49] Chung,2018,p.167
[50] N36
[51] Evans,2014,p.229
[52] Ibid.
[53] N1
[54] Evans,2014,p.234
[55] N51
[56] [2000]W.T.L.R.1033
[57] Ibid.p.1044
[58] Evans,2014,p.231
[59] Ibid.,p.235
[60] Ibid.,p.236
[61] Meager,2004,p.206
[62] Pawlowski,2004,p.391
[63] Pearce,2010,p.245
[64] Ibid.