Erskine 1948 Trust, Re

This Part 8 claim concerns the construction of the phrase ‘statutory next of kin’ in an English settlement made in 1948, and the possible effect of the Human Rights Convention on that construction.

The basic facts are not complicated and are not in dispute. The settlement was made on 10 December 1948 by Denys Malcolm Erskine. The original assets were shares in Associated Hotels Limited and are now represented by a trust fund worth about £3.2 million. The first beneficial trust was in favour of the settlor’s daughter Margaret Lucile Erskine (known as Lucile) for life. She married three times and was lastly Mrs Saich. She died on 11 August 2010 without children or other descendants. After her death, the settlement provided a power of appointment exercisable in favour of her children and remoter issue with a trust over in favour of her children. Those provisions have failed, and the important beneficial provisions are clause 1(4) and (5), with Mrs Saich as ‘the Beneficiary’ : —

Looking first at clause 1(4), Mrs Saich’s sister Denise attained the age of 30 years, and also married, twice, becoming Mrs Pigott and later Mrs Delaney. But the opening words of clause 1(5) apply, because Mrs Delaney did die during Mrs Saich’s lifetime in the early 1980s. So the whole of the trust fund vests in Mrs Saich’s statutory next of kin on the footing that she had died a spinster. Now the state of the family at Mrs Saich’s death was that she had no children or other issue (as I have mentioned), no living parent and no living siblings. Her sister Mrs Delaney had been her only sibling, and she had left two children, Christopher Richard Erskine Pigott and Stephen Robert Erskine Pigott, both of whom are living and are the first two defendants to this action. They are not the biological children of Mrs Delaney, having been adopted by her and her then husband in 1951 and 1952.

The only question in the case is whether Christopher and Stephen Piggott are entitled to the trust fund as the persons who were Mrs Saich’s statutory next of kin at her death, or whether their claims are negated by their having been adopted. If Christopher and Stephen Pigott are not entitled to the trust fund, then it is common ground that the fund vests in a number of cousins, many of whom have not been identified. One of those who have been identified is Fiona Margaret Lady Mottram, a niece of the settlor, and she has been joined as a defendant to represent all the cousins.

All parties appeared before me by counsel, Mr Edward Hewitt for the trustees (who are neutral), the Pigott brothers by Mr David Rowell, and Lady Mottram by Mr Charles Holbech.

Mr Rowell puts his case in two ways. First, as a matter of pure construction, he says that his clients are indeed the statutory next of kin of Mrs Saich. When she died in 2010 all impediments to inheritance by adopted children had been removed by statute, and if Mrs Saich had died intestate and a spinster, Christopher and Stephen Piggott would have inherited her estate. Second, he says that, if he fails on that first basis, the Court should construe the settlement in such a way as to eliminate discrimination against adopted children, by virtue of articles 8 and 14 of the Convention for the Protection of Human Rights and Fundamental Freedoms, and should be guided in this process by the jurisprudence of the European Court of Human Rights. The High Court is itself a public authority for this purpose, and should not entertain a construction which would put the United Kingdom in breach of the Convention.

Mr Holbech submits that the phrase ‘statutory next of kin’ is given a meaning by statute and, applying that meaning, the adopted nephews of Mrs Saich do not qualify. It is indeed common ground that none of the successive Adoption Acts passed in 1950 and later have any effect on the issue before me, because they expressly do not apply to dispositions made before those Acts came into force, and they therefore do not have any effect on this settlement made in 1948. As to the Human Rights Convention, he submits that the Strasbourg jurisdiction should affect the construction of an instrument only if the disposition is ambiguous, and this one is not.

Looking at the submissions in more detail, Mr Rowell recognizes that the Adoption Acts do not help him. The Adoption of Children Act 1926 had, under section 5, no effect either to confer property or inheritance rights on adopted children or indeed to remove such rights from them. The Adoption Act 1950 did confer inheritance rights by providing in principle that the expression ‘child’ includes an adopted child, but Schedule 5 to the Act, paragraph 4, provided that this did not affect a disposition made before 1950. The Adoption Act 1976 contained a similar limitation, and that limitation has been preserved in later statutes as well. There is no doubt that the disposition in question in the present case is the 1948 settlement.

But Mr Rowell says that the Adoption Acts are irrelevant. The reason is that they make provision for the effect of words of relationship, such as ‘child’, ‘children’ and ‘issue’, but they do not make provision for next of kin. His clients, he says, are Mrs Saich’s next of kin (both on the hypothesis specified in the settlement and generally), and it is unnecessary to examine the process by which that status has come to pass, for instance by their being her sister’s children. He pointed out that a trust in favour of the Mayor of London depends on a simple fact of identity. One does not have to enquire how he or she became mayor. An alternative analysis, according to Mr Rowell, is that one looks not only at the relevant facts at Mrs Saich’s death but also at the law at that date. In 2010 adopted children were entitled to inherit equally with natural-born children, so they were the statutory next of kin. It would be anomalous to apply 1948 law to 2010 facts.

Mr Holbech adopts a more systematic analysis. The second part of section 50(1) of the Administration of Estates Act 1925 provides a statutory definition of the phrase ‘statutory next of kin’ : —

I should add that the 1925 Act has been amended in several ways, with effect from 1970 so as to substitute the age of 18 years as the age of majority, and most recently to include references to civil partners, but I have set out the relevant provisions as originally enacted, and as they stood at the time of the 1948 settlement.

I was referred to section 20(2) of the Interpretation Act 1978, which provides that references in a statute to another enactment are references to that other enactment as amended, but this does not apply to pre-1978 statutes. There was no similar provision in the predecessor to the 1978 Act.

The end result is that, if the word ‘issue’ includes adopted children, then Christopher and Stephen Pigott, both of whom have attained 21 years of age, would qualify as the issue living at Mrs Saich’s death of her only sibling Mrs Delaney. But Mr Holbech submits that the word ‘issue’ in the 1925 Act itself did not include adopted children, so that adopted children did not come within the intestacy rules, and the later statutes which I have mentioned do not help. Therefore, adopted nephews do not qualify within the phrase ‘statutory next of kin’.

There is a curiosity here. Subject to some special exceptions which do not apply in the present case, the question whether any given individual qualifies as the next of kin of another person must be determined at the death of that other person, not at any other time. And yet the law in question, according to Mr Holbech, is the law in force at the date of the settlement or (in the case of a will) the testator’s death. This is a mismatch which Mr Rowell’s submissions seek to avoid. But Mr Holbech points out that there are numerous instances where next of kin (or similar classes) have been held to be ascertained at one date under the law in force at an earlier date.

The closest case to the present is In re Hooper’s Settlement, Phillips v Lake [1943] Ch 116, a decision of the Court of Appeal upholding Farwell J. The case concerned a 1912 settlement containing a trust ‘for such person or persons as would have become entitled to the settled fund under the statutes for the distribution of the personal estate of intestates at the death of the settlor had he died possessed thereof intestate and without having married . . .’. It was argued that this meant the persons at the death of the settlor who were entitled under the statutes in force at the same time, but that argument was rejected. Instead, and in reliance on section 50(2) of the Administration of Estates Act 1925, it was held that it meant the persons at the death of the settlor (in 1941) who were entitled under the statutes in force immediately before the commencement of the 1925 Act.

I was also shown In re Gansloser’s Will Trusts [1952] Ch 30, which concerned a trust in a 1929 will in favour of the testator’s widow’s ‘relatives’, a word which was held to mean the persons who would have taken under the pre-1926 statutes of distribution. Evershed MR expressed the normal principle, that the beneficiaries are to be ascertained at the death of the person in question, but also found an exception to the principle in that particular case, holding that the relatives were to be ascertained at the testator’s death. The exception resulted in both the law and the facts being ascertained at the death of the testator, before the death of the widow whose relatives were beneficiaries. It also resulted in the facts being ascertained at 1929 under pre-1926 law (as it happens, a statute of King Charles II).

Also in accordance with that general principle, Falkiner v Commissioner of Stamp Duties [1973] AC 565 was a decision of the Privy Council on appeal from New South Wales. The settlor had made a settlement on 4 October 1961 and died 11 days later. The ultimate trust was in favour of ‘the next of kin of the settlor as determined by the provisions now in force of the Wills, Probate and Administration Act 1898–1954 of the State of New South Wales’. There was a claim for death duties on the settled property, and one of the issues was whether the clause which I have cited contained ‘a trust . . . to take effect after [the deceased person’s] death’ within section 102(2)(a) of the Stamp Duties Act 1920–1959. It was held that the next of kin were to be determined at the settlor’s death, not at the date of the settlement, and death duties were payable accordingly. This is therefore another example of a trust in favour of a class to be ascertained at a person’s death under a law in force at an earlier date (though in this case only 11 days earlier).

I was also shown In re Bourke’s Will Trusts [1980] 1 WLR 539, where the same principle was repeated, and without any exception such as that applied in Gansloser. In Bourke Slade J had to construe the 1938 will of a testator who died in 1943. I shall simplify the relevant trusts, excluding those which are irrelevant, in order to explain the point. There was a life interest for the testator’s widow and, on her death without issue (which happened in 1971), residue was given to the testator’s three half-siblings ‘or their heirs and surviving issue’. He then held that the heirs were to be ascertained in accordance with pre-1926 law by virtue of section 132 of the Law of Property Act 1925. And it was evidently accepted that ‘issue’ (construed to mean children in the context) were to be ascertained in accordance with the law at the date of the testator’s death in 1943 (though it is not suggested that there were any illegitimate children born after 1969 who might have made a claim). But the other finding was that the classes of heirs and issue were in each case to be ascertained at the death of each half-sibling (1958 and 1969 respectively), not at the date of the widow’s death in 1971.

I am not greatly assisted by the Falkiner and Bourke cases, but they are certainly cases where the class of beneficiaries was ascertained at one date but in accordance with the law in force at an earlier date, and to that extent they assist Mr Holbech’s case. I regard the Hooper case as more strongly in his favour. It is not conclusive of the present case, because it dealt with subsection (2) of section 50 of the 1925 Act and the pre-1926 statutes of distribution, not subsection (1) and the 1925 Act itself. But it did require the beneficiaries to be identified in accordance with previous law, by virtue of a part of the same section 50 which is invoked in the present case, even though the phrase ‘at the death of the settlor’ could grammatically apply to ascertain the relevant statutes (but was construed otherwise). I regard the present case as, if anything, clearer, because section 50(1) refers expressly and exclusively to the 1925 legislation itself.

Mr Rowell made the attractive submission which I have already summarised, namely that his clients are the next of kin of Mrs Saich and that this should be the end of the matter. There is a curiosity in this submission also, namely that it contemplates the settlor Mr Erskine establishing a trust of capital in favour of those persons who would at an uncertain future date be the next of kin of his daughter under a statutory provision in force at an uncertain future date in terms unknown to him and unpredictable by him. His answer to that is that it is no different from every other settlor, who uses words which might be construed in an unknown and unpredictable way by a later decision of the court, perhaps in an unrelated case. But, with respect, that does not seem to me to compare like with like.

The fact is that the phrase ‘statutory next of kin’ used in the 1948 settlement before me had a statutory definition. It meant the persons who qualify under the intestacy provisions in the 1925 Act itself. In short, this settlement is no different in principle from the cases where the relevant statutes were identified explicitly in the trust instrument. In my judgment it makes no difference that the settlor has identified the intestacy provisions of the 1925 Act indirectly by using a phrase having a statutory definition. If I were to hold the opposite I would be disregarding section 50(1) of the 1925 Act. In short, and disregarding the Human Rights Convention for the moment, I accept Mr Holbech’s submissions on this part of the case.

I turn to the possible effect of the Human Rights Convention, which became part of English law on 2 October 2000 by virtue of the Human Rights Act 1998. Section 2(1) provides : —

The particular rights which are invoked in the present case are those contained in articles 8 and 14 : —

Everyone has the right to respect for his private and family life, his home and his correspondence.

There shall be no interference by a public authority with the exercise of this right except such as is in accordance with the law and is necessary in a democratic society in the interests of national security, public safety or the economic well-being of the country, for the prevention of disorder or crime, for the protection of health or morals, or for the protection of the rights and freedoms of others.

It is plain from the cases mentioned above (Pla, Upton and Brauer) that the ECtHR has an emphatic aversion to discrimination against illegitimate and adopted children. But in the context of the present case this aversion is not easy to apply directly. In the first place the Convention became part of English law only in 2000. The question of retrospectivity therefore arises : how does the Convention affect the construction of a definition enacted almost 75 years before the Convention became part of English law? Second, even if the Convention affects the construction of a statute, to what extent does it affect the construction of a private-law disposition which incorporates a statutory definition? Those two questions combine into the single question before me : to what extent does the Convention affect private property rights established by a 1948 settlement, made before the Convention was even agreed by the Council of Europe (4 November 1950), and over 50 years before it became part of our law?

Moreover the prohibition against discrimination applies to governments which have become signatories to the Convention, but the same prohibition does not apply to private individuals, even if they are citizens of a state which has adopted the Convention. The United Kingdom government is therefore prevented from introducing legislation discriminating against adopted children, but individual citizens are not. If therefore the settlor in the present case has included a discriminatory provision in his settlement, he has done nothing wrong and the provision is valid. The task of the court of construction would then be to determine and declare the true meaning and effect of the provision, discriminatory though it is, and I apprehend that in general terms the United Kingdom would not then be infringing the Convention. There will however be an infringement if the court creates a discrimination which was not effected by the settlor.

So far as retrospectivity is concerned, the leading English authority is Wilson v First County Trust Limited (No 2) [2004] 1 AC 816, which concerned a 1999 agreement with pawnbrokers, who argued that provisions in the Consumer Credit Act 1974 infringed their Convention rights. The House of Lords held, amongst other things, that it could not have been Parliament’s intention that the application of section 3(1) of the 1998 Act should have the effect of altering the existing rights and obligations of the parties to an agreement made before section 3 came into force. At the same time the presumption against the retrospective operation of legislation is not unqualified. The underlying rationale was spelt out by Staughton LJ in Secretary of State for Social Security v Tunnicliffe [1991] 2 All ER 712 at page 724 (cited in Wilson at paragraph 19 of the report) : —

In a purely statutory context the same or a similar process of balancing fairness was applied in Brauer v Germany (2010) 51 EHRR 23. The applicant was born outside marriage in 1948 in East Germany and claimed a share of the estate of her father, who lived in West Germany and who died in 1998. West German law included a 1969 statute putting illegitimate children on a equality with children of a marriage, but it expressly did not apply to a child born before 1 July 1949. This was intended to exclude, or not to include, children who had reached the age of majority (then 21 years of age) before the statute came into force. The ECtHR decided that this discriminatory provision of the statute was incompatible with the Convention. It was recognised that the aim of the legislation, distinguishing between children born before July 1949 and those born later, may have been legitimate, and the ECtHR even recognised the political and other difficulties which would have been involved if the distinction had been removed. However, as appears from paragraph 43 of the report, this was not enough : —

In paragraph 44 of the report the ECtHR went on to consider whether the means employed by the legislation were proportionate to its aims : —

At least one important general message can be derived from the Wilson and Brauer cases (above) in the present context, namely that the Convention may operate retrospectively, but only if that is fair in the individual case. Brauer also shows that evidence of the effects of the legislation and of the Convention on the applicant personally, and on other persons competing with the applicant, may be relevant and even ‘decisive’.

But Brauer was concerned with the direct effect of legislation on rights of inheritance, and the ECtHR was awarding compensation to be paid by a government which had infringed a prohibition contained in the Convention to which that government had subscribed. In that context the question was whether the discriminatory effect of the national legislation was proportionate to the acknowledged purpose of that legislation. That is not quite the same as any question before me, because I am not adjudicating on the validity of national legislation or any possible infringement of the Convention by the state. The relevant question in Wilson was whether the retrospective application of the Convention to earlier legislation was unfair to a party to an existing contract. Again the question before me is different. It is whether the Convention has any effect on the construction of a private disposition, and this question is (or at least appears to be) independent of any question of retrospectivity. As I have mentioned, a private individual is free to discriminate against adopted children if he chooses, and that remains the case even after the Convention has become part of national law. But is that decisive, or is it possible that the Convention actually has effect to change the true construction of a valid private disposition?

I now turn directly to that question. Understandably Mr Rowell placed heavy reliance on the decision of the ECtHR in Pla (or Pla and Puncernau) v Andorra (2006) 42 EHRR 25, because, by a majority, it applied the Convention, in favour of an adopted child, so as to reverse (by awarding compensation) the effect of the construction placed by the national courts on a private disposition, namely a pre-Convention will. That is a surprising decision, and in the light of what I have already written it would be surprising even in the context of a post-Convention will. As I have mentioned more than once, individuals, unlike public authorities, are free to discriminate against adopted children if they wish, and that is what the domestic court of appeal had found to have happened. The ECtHR nevertheless awarded damages to the adopted son and his mother.

Superficially it might be thought to follow, in a jurisdiction such as England and Wales where the Convention now forms part of domestic law, that the court should readily or routinely construe a private disposition in such a way as to eliminate discrimination, even where the person making the discriminatory provision was entitled to do so. That is, I may say, not my interpretation of the decision, but I shall need to describe the case at some length.

Pla and Puncernau was concerned with a will made by a widow in 1939, in which she left certain property to her son Francesc-Xavier, described in the report as tenant for life, with a stipulation that he was to leave this inheritance to a son or grandson of a lawful and canonical marriage, failing which the estate was to pass to the children and grandchildren of the testatrix’s daughters. The testatrix died in 1949. In 1995 Francesc-Xavier made a will in favour of his wife, but in what was described as a codicil of 3 July 1995 he left the assets inherited under his mother’s will to his wife for life and then to their adopted son Antoni, who was born in 1966 and adopted by them, in Spain, in 1969. The assets were described in the report as real estate. Francesc-Xavier died in 1996.

Two great-granddaughters of the testatrix applied to the Tribunal des Batlles to have the codicil declared void. (One might have thought that the only proper claimant for such an application would have been Antoni’s mother, who was the residuary beneficiary of Francesc-Xavier’s existing 1995 will, but evidently that would not have been the case under the law of Andorra.) The Tribunal dismissed the application on the basis that the case was governed by the testatrix’s intention determined by the terms of her will, that there was no statutory or constitutional provision relating to adopted children at that time in Andorra, that customary law was derived partly from Roman law under which adopted children ranked equally with natural-born children, and accordingly the testatrix, by making no express exclusion of adopted children, should not be taken to have excluded them by implication. I understand that legislation was introduced in 1958 putting adopted children on an equality, but that this did not affect the matter.

The great-granddaughters appealed, however, and in 2000 the High Court of Justice (which I shall call the appeal court) allowed the appeal and set aside the codicil in its entirety. The appeal court agreed that the matter was governed by the testatrix’s expressed intention when she made her will, and that Antoni could not rely on the legislation of 1958. But they regarded it as significant that in the first half of the 20th century adoption was virtually unknown in Andorra, that provisions of Roman law ‘could not easily be transposed’ to Andorran families living at that time, that in any event the relevant Spanish legislation (under which Antoni had been adopted) gave an adopted child rights of inheritance from his or her adoptive parents but not from more remote family members; so that, by not explicitly including adopted children, the testatrix should not be taken to have intended to include her son’s adopted son.

Antoni and his mother made two attempts to have this ruling annulled by the national courts in Andorra, but those attempts failed. They then brought the claim in the ECtHR. Their case was that the appeal court had acted in breach of articles 8 and 14 by allowing the appeal from the decision of the Tribunal, and they are described in paragraph 35 of the report as maintaining that the case should be governed by private law in the light of Andorran law in force in 1996, when Francesc-Xavier died, and the Convention. In the event the ECtHR took the relevant national legislation to be that in force in 1939 and 1949 (rather than 1996).

The government of Andorra took a preliminary point that no relevant ‘family life’ was affected because Antoni had been adopted 20 years after the testatrix had died, but the ECtHR ruled unanimously that inheritance by children and grandchildren does fall within the scope of article 8, relying on Marckx v Belgium (A/31) (1979–80) 2 EHRR 330 paragraphs 51–52 and Camp and Bourini v Netherlands (2002) 34 EHRR 59 paragraph 35.

The ECtHR was divided on the main issue, with a majority concluding that Antoni’s rights under articles 8 and 14 had been infringed, but with two members of the court, namely Judge Bratza and Judge Garlicki, dissenting. The majority began by pointing out that previous cases of this kind before the ECtHR had concerned the discriminatory effect of statutes in various member states, whereas the instant case related to the interpretation or construction of a testamentary disposition. They went on to record the principle that domestic law should normally be determined by the domestic courts, all the more so with a disposition such as a will. ‘Accordingly,’ they continued at paragraph 47 : —

The majority went on to find that the decision of the appeal court was blatantly inconsistent with the Convention. The main part of their reasoning was expressed in paragraphs 58 to 60 of the report : —

It is easy enough to see what the majority were deciding in those passages, namely to criticize and counteract the discrimination against the adopted grandson. It is also easy to sympathise with the end result, and even perhaps to be surprised by the decision of the Andorran appeal court. But the more surprising aspect of the ECtHR decision is that the majority explicitly disagreed with the national appeal court on the point of construction under the law of Andorra. That appears to have been an essential part of the majority opinion. In point of detail, they wrote in paragraph 56 that it was not appropriate for the ECtHR to analyse the legal theory behind the decision of the appeal court : ‘That is a sphere which, by definition, falls within the jurisdiction of the domestic courts.’ But in the later paragraphs which I have quoted they appear to have breached that principle themselves. Similarly in paragraph 59 they wrote, ‘Admittedly, the [ECtHR] is not in theory required to settle disputes of a purely private nature,’ but then proceeded to do exactly that, or at least something very similar.

The majority later expressed their conclusion in terms of the appeal court’s decision being blatantly inconsistent with the principle of article 14, but in order to reach that conclusion they had to reject the decision of the appeal court as a matter of national law. That is how I read the unnumbered paragraph lying between paragraphs 58 and 59 in the passage quoted above. In my view they decided first that Antoni was rightly entitled to inherit, as the Tribunal had decided, under Andorran law. As I see it they could not otherwise have concluded that the appeal court had effected the improper discrimination.

That is also how Judge Bratza saw the matter in his dissenting opinion. At paragraph O–I5 he wrote : —

For my part I am not entirely sure that the majority did regard the appeal court’s decision as arbitrary or manifestly unreasonable. According to paragraph 59 of the report, the circumstances in which the ECtHR should not remain passive were said to be those where the national court’s interpretation appeared unreasonable, arbitrary ‘or, as in the present case, blatantly inconsistent with the prohibition of discrimination established by Article 14 . . .’. In my view they based their decision on the appeal court having ‘blatantly’ effected the improper discrimination. But it was a necessary part of that decision that the Tribunal was right about the construction of the will, which was a question of construction under Andorran law.

The most far-reaching part of the majority opinion in Pla and Puncernau is the following paragraph. It is not apparently essential to the decision, but it does contain explicit guidance to national courts faced with the construction of discriminatory provisions in private documents : —

That is a widely expressed passage. And if ‘interpretation’ is used in the sense of ‘construction’ it requires the court to construe any disposition, whenever made, in such a way as to correspond with the Convention as well as national law, though presumably only where that is feasible. That goes beyond what is provided by section 3 of the Human Rights Act 1998 in regard to the construction and effect of prior statute law. Indeed the Convention now forms part of English law. At the same time I must bear in mind that the majority were considering two wills having effect under the law of Andorra. Francesc-Xavier was described in the report as a life tenant, but it would be wrong to assume, without more, that his mother’s will established something in the nature of a will trust conferring on him a special testamentary power of appointment. The phraseology of the report is consistent with a testamentary gift of land to Francesc-Xavier coupled with a condition that he should make a testamentary gift of the estate to a son or grandson of the required characteristic. It appears that no one other than Francesc-Xavier had any interest in the estate between 1949 and 1996 when he died. In the submissions on damages the government of Andorra is recorded at paragraph 68 of the report as submitting that Francesc-Xavier could have kept half of the estate for himself. It was also apparently possible for him to leave his widow a life interest, if only he had not given the reversion to Antoni. Paragraph 62 is therefore written in the context of Francesc-Xavier’s will having had some of the features of an independent disposition. This is consistent with the way in which Antoni and his mother are recorded as relying on 1996 law and the Convention (as I have mentioned in paragraph 36 above).

The case of an English-law settlement is different. Beneficial interests are established from the outset and are enforceable in the courts throughout the continuance of the trust relationship. Even an expectancy can be assigned for value : In re Ellenborough, Towry Law v Burne [1903] 1 Ch 697, in which the court declined to give effect to a voluntary disposition of a mere expectancy, but Buckley J nevertheless said at page 700 : —

In the present case this raises a somewhat different question of retrospectivity from the one faced in Wilson (above), and different from the question in Brauer (above). It seems to me clear that, if the present application had been before the court before 2 October 1998, the court would have been obliged to have regard only to national law when construing provisions in the settlement. The Convention could then have been invoked, but only in the ECtHR and only for the purpose of seeking compensation from the state. But now that the Convention has become part of English law paragraph 62 suggests that other considerations are relevant and might even prevail at the stage when the disposition is being construed.

This presents special problems with a settlement, or indeed a will trust, which has already come into effect before the Convention became part of English law. It is in principle essential that the beneficiaries of a trust, including potential beneficiaries, are conceptually ascertainable from time to time, even if they are not always evidentially ascertainable or ascertained. This is not limited to beneficiaries in possession. It applies to reversionary and contingent beneficiaries as well. Trustees are required to take into account the needs and characteristics of their beneficiaries, including future and potential beneficiaries, when exercising their discretions, and it would be extraordinary if they had to contend with possible future changes in the rules of construction by which the beneficiaries and the extent of their interests are identified. Similarly the court, in exercising its supervisory role over trusts, must have the means of ascertaining, conceptually at least, who the present and future beneficiaries are. Similar questions routinely arise in the context of applications under the Variation of Trusts Act 1958. More importantly, beneficiaries have interests under a continuing settlement or will trust, and it would be extraordinary (at least in most cases) if those interests were to be created, altered or destroyed by the Convention becoming part of national law.

It is therefore difficult to suppose, for example, that Lady Mottram was a beneficiary on 1 October 1998, having a saleable interest under the settlement, but that her interest was overridden or downgraded on the following day by the Convention becoming part of national law. It is equally difficult to suppose that Christopher and Stephen Pigott became potential beneficiaries on that event without having been so before.

That is indeed difficult, but is it impossible? I am required by section 2 of the 1998 Act to take into account the decision of the majority in Pla and Puncernau (above), and indeed their ‘advisory opinion’ (to quote from section 2), which I think describes paragraph 62 of the report. One feature of the decision is the controversial way in which the ECtHR overrode the national court’s construction of the document; a feature criticized by Judge Bratza in the paragraphs which I have cited, and a feature to which, in the nature of things, a court of construction cannot effectively respond. In the end there are only three points of guidance which, for this purpose, I can and do derive from the decision. First, I must avoid a decision which is unreasonable, arbitrary or blatantly inconsistent with the prohibition of discrimination established by article 14 (though I do not see that anything is added by the adverb ‘blatantly’). I derive that from paragraph 59 of the report, towards the end of the passage cited in paragraph 39 above. Second, I must not put words into the settlor’s mouth, so to speak, and I should construe the disposition in a way which corresponds to national law and the Convention as developed in the ECtHR’s case law. I derive that from paragraph 62. Third, if the disposition, as worded by the settlor, did make a distinction between biological and adopted children, then I have to give effect to that distinction. I derive that, by implication, from the unnumbered paragraph which follows paragraph 60 of the judgment (the last paragraph in the passage cited in paragraph 39 above). The priorities between those points is not spelt out.

There is one way in which, superficially, the Pla and Puncernau decision might be applied very closely to the present case. In Pla and Puncernau the testatrix’s will did not mention adopted children one way or the other, and it was this which enabled the two national courts to reach opposite conclusions about what could legitimately be inferred from that omission. It also allowed the ECtHR to conclude that a will made in 1939 of a widow who died in 1949 impliedly included an adopted grandson, even though the practice of adoption was apparently unknown in Andorra at those times. (Even in 1969 Francesc-Xavier evidently had to invoke Spanish law in order to effect Antoni’s adoption.) Similarly in the present case, the phrase ‘statutory next of kin’ was defined in terms which did not themselves mention adopted children one way or the other. There was no provision in the Administration of Estates Act 1925 to state that adopted children were either included or excluded from benefit. This was not surprising. Even the Adoption of Children Act 1926 was not yet on the statute-book when the Administration of Estates Act 1925 came into force, and I infer that there was no regular legal framework for adoption at that time.

Could it therefore be suggested that Mr Erskine’s settlement was just as non-specific as the testatrix’s will in Pla and Puncernau, requiring the court to construe it in favour of the adopted nephews? After all, no one had vested rights under clause 1(5) of the settlement until Mrs Saich’s death. Lady Mottram and those whom she is asked to represent in these proceedings had only a future expectancy, contingent on surviving Mrs Saich and not carrying an entitlement to income in the meantime. That might not be enough to outweigh the ‘imperative’ of eliminating discrimination which article 14 embodies.

In my view that is not quite a sufficient answer to the question before me. By the time of the 1948 settlement the legal framework for the adoption of children had been established by the 1926 Act, and section 5 of that statute had expressly enacted that adoption had no effect to confer inheritance rights on adopted children unless a contrary intention appears. That was the state of English law when the settlement came to be made in 1948. Seeing that there was no express contrary provision in that settlement, it follows that the use of the defined phrase ‘statutory next of kin’ did not include adopted children. That was the position under English law before 2 October 1998, and in the case of most settlements and will trusts I dare say that it will be found impossible to conclude that the effect of the Convention becoming part of national law will have altered the construction of the disposition in any way.

But I have come to the conclusion that in this particular settlement it is possible to contemplate and even to find a retrospective application of the principle of non-discrimination. I have mentioned that section 3 of the 1998 Act is not generally retrospective, but in Wilson (above) it was acknowledged that this is not an absolute rule. The section can be retrospective in cases (not including Wilson itself) where that is achieved without unfairness. Similarly I am prepared to accept in principle that the Convention becoming part of English law can have an effect on the construction and effect of an existing trust, if that can be achieved without unfairness. This is not fully retrospective. The new construction derived from the Convention would operate only from 2 October 1998. But it would, from that date, alter beneficial interests under the settlement. For that reason the requirement to avoid unfairness will almost certainly prevent any new or different construction being adopted for many existing settlements and will trusts. I dare say that, in broad generality, Mr Holbech was right in his submission that, if the meaning and effect of the disposition is clear, then the Convention does not affect it.

But the present case has special features. First, the settlement itself contains no explicit exclusion of adopted children. To that extent it is on a par with Pla and Puncernau. Second, the way in which the relevant statutes have effect, under English law, to exclude adopted children is itself discriminatory and now contrary to the prohibition embodied in article 14 as developed in the Strasbourg case law. (This is not to say that the United Kingdom is in breach of the Convention, because non-discriminatory legislation has now been enacted.) Third, a trust for a person’s next of kin can be regarded as unique. It does not, by definition, create vested or even contingent interests until the death of the person whose next of kin the beneficiaries eventually become. It creates an expectancy only. Fourth, although it is true that an expectancy can be assigned for valuable consideration, there is no evidence before me that any such assignment has been effected, or that any is proposed. Fifth, as in Pla and Puncernau, and indeed in Upton v United Kingdom (below), the contest is between relatively distant cousins on the one hand and the children of Mrs Saich’s only sibling on the other, and those children have been full members of the family for some 60 years. The evidence is that some of the cousins have not even been identified, and it may be inferred that those at least were not well known to Mrs Saich or Mrs Delaney. Sixth, and whichever construction is preferred, the settlement has now come to an end, and there will be no continuing problem for the trustees or the court to identify future beneficiaries.

Because of these special features, without any one of which I would have reached a different answer, I conclude that the settlement should now be construed in such a way as to eliminate the discrimination against adopted children. To the extent that this is unfair to Lady Mottram and her constituents, my view is that their own claims, though valid under English law without the operation of the Convention, are not based on an explicit exclusion of adopted children expressed by the settlor himself. It is based on his adoption of a statutory definition which is now outdated in its discriminatory effect. I bear in mind the ECtHR’s advice in paragraph 62 of the majority opinion against putting words into a settlor’s mouth and about the importance of changing social conditions. Ultimately the unfairness to them is outweighed by the corresponding unfairness to Christopher and Stephen Pigott under the opposite conclusion, and by the other considerations which I have already mentioned. I will declare that the trust fund vested in Christopher and Stephen Piggott on the death of Mrs Saich.

I have mentioned the case of Upton v United Kingdom (2008) 47 EHRR SE24, which was cited to me. The decision contained comments on Pla and Puncernau (above), suggesting that that case might have been relevant to Mr Upton’s appeal. The case concerned a will made in 1930 in favour of the testator’s children and grandchildren. The applicant was Mr Tim Upton. His father was the testator’s eldest son, and Tim’s mother was the wife of the testator’s youngest son. She died not long after Tim was born, and he was then adopted by his father in 1955.

At first instance Mr Upton’s claim was dismissed by HH Judge Behrens sitting as a Judge of the High Court. That was before the decision in Pla and Puncernau was given. There was then a long delay during which the applicant failed to launch any appeal against the decision, and on 14 November 2005 Parker LJ rejected his final application to reinstate his appeal after a gap of over a year.

The ECtHR rejected Mr Upton’s claim against the United Kingdom, on the ground that he had not exhausted his domestic remedies by pursuing the appeals process in due time. But at pages 284–285 of the report they also mentioned Pla and Puncernau : —

In that passage the ECtHR are not ruling out the national court taking account of the Convention and of Pla and Puncernau in regard to the construction of the 1930 will. I therefore find nothing in the Upton case to contradict my own decision. Needless to say, nothing in this judgment should be taken as a comment on the merits of the Upton case itself.

I shall make a declaration as described at the end of paragraph 56 above, and I shall make a representation order as asked. If asked, I shall hear submissions on costs but, subject to that, I propose to order the costs of all parties to be borne by the trust fund, including Lady Mottram’s costs to be assessed on the indemnity basis if not agreed.