Thomas & Ors v Kent

LORD JUSTICE CHADWICK: This is an appeal from an order made on 8 July 2005 by HHJ Weeks QC sitting as a deputy Judge of the High Court in the Bristol District Registry of the Chancery Division in proceedings brought for the purpose of determining the true construction and effect of the will dated 11 April 1944 of the late Mr John Jones, a Welsh farmer.

The testator died on 28 October 1945. By his will he appointed executors and trustees and gave to them all his real and personal estate. He directed payment of certain pecuniary legacies and he made a specific bequest of his household effects in favour of his daughter, Mary Sophia Radcliffe. By clause 3(c) of his will he gave a freehold house and adjoining land to his granddaughter, Mary Kathleen Radcliffe, for her life and after her death for her child or children, and if more than one in equal shares. By clause 5 he directed that his residuary estate should be held upon the following trusts:

upon trust to pay the income to any child of hers during his or her life and if more than one, equally between them during their lives, and after their deaths,

upon trust as to capital and income for their children absolutely, such children to take per stirpes and equally between them the share or shares of his or her parent.

The testator’s daughter, Mary Sophia Radcliffe, had only one child: the granddaughter of the testator, Mary Kathleen Radcliffe. Mary Kathleen survived her mother and died in 1992 without children or issue. The effect was that on her death, the trusts of the freehold house in clause 3(c) and the trusts of residue in clause 5 failed. The estate fell to be distributed under the ultimate trust in clause 6.

These proceedings were commenced on 18 November 2004 by the issue of a claim form on behalf of the present trustees of the estate. They sought determination by the court whether, on the true construction of the will and in the events which had happened, the funds remaining in the testator’s residuary estate (“the trust fund”) should be distributed in one, and if so which, of the three ways which are set out under paragraphs (a), (b) and (c) of clause 1(1) of the particulars of claim served with the claim form; or in such other way as the court might determine. Amongst the other relief sought by the trustees was an order that Mrs Pauline Kent be joined as a defendant to the proceedings in order to represent persons who would be interested in contending that the trust fund be distributable in accordance with paragraph (a) of clause 1(1). The trustees themselves indicated that they intended to advance arguments in favour of paragraphs (b) or (c) of that clause.

In order to understand the problem with which the trustees were faced – and the basis of the three possible solutions to that problem which they offered in paragraphs (a), (b) and (c) of clause 1(1) – it is necessary to have in mind such particulars of the testator’s extended family as they have been able to put before the court. That has not, of course, been an easy task, given that the testator died some 60 years ago. But doing the best that they can, they set out those particulars in paragraph 11 of a witness statement dated 30 September 2004 of Mr Milford Merryweather, one of the current trustees. He explained that the testator was one of the 11 children of his parents. The eldest of those children was born as long ago as 1849. At the date when the will was made (in 1944) three of the testator’s sisters – Sarah, Elizabeth and Sophia – were still living. One brother, David, was still living; but he was the brother who had been excluded in terms by clause 6 of the will. There was one brother, William, whose date of death was unknown. But, as he was born in 1850, he would have been 94 when the will was made. The other five siblings were known to have died before the will was made.

The first question which concerned the trustees was whether the gift to “my brothers and sisters in equal shares” – with the substitution of the children of any deceased brother or sister for that deceased brother or sister’s share – was to be construed so as to include all the brothers and sisters of the testator (other than the excluded brother, David) whether or not alive at the date of the will. There is, on a strict analysis, a further question: if the phrase “my brothers and sisters” is not to be construed as a reference to all the testator’s brothers and sisters, whether or not alive or dead at the date of the will, does it mean (i) those brothers and sisters who were living at the date of the will, or (ii) only those brothers and sisters who survived the testator – that is to say, those who were also living at the date of his death? But that further question did not arise on the facts. The only sibling who might have died between the date of the will and the date of the testator’s death was his brother William; and since it was not known when William died or whether he had children, the choice between those two dates was of no materiality.

The underlying structure of the proposed distribution turns on the primary question: does the phrase “my brothers and sisters” in clause 6 include all brothers and sisters of the testator (other than the excluded brother, David) whether or not they were alive at the date of the will? If the answer to that question is that the gift is only to the brothers and sisters living at the date of the will, then the estate is divisible into three, or perhaps four, shares. I put it that way because there must, in principle, be one share for the children of each of the three sisters, Sarah, Elizabeth and Sophia; or perhaps one share for each sister and a fourth share for the children of the testator’s brother William or for William’s estate. That is the distribution for which, in principle, paragraph (a) of clause 1(1) provides. More particularly, it provides for distribution in three shares with insurance provision to cover the possibility that there should be a fourth share for William’s estate or for his children.

If the answer is that the gift is to include brothers and sisters who died before the date of the will, then the estate is divisible into eight, or perhaps nine, shares. Again, I put it that way, because it is not known whether William had died leaving children. That is the distribution for which, in principle, paragraphs (b) and (c) of clause 1(1) provide. It is proposed that the estate should be divided into eight shares, with insurance provision to cover the possibility of a ninth share to which William’s estate or his children would be entitled.

The answer to that primary question determines whether, in principle, distribution is by reference to three, or perhaps four, shares; or by reference to eight, or perhaps nine, shares. But there are two further questions: one of which was addressed in part by the three options offered by the particulars of claim, and the other of which was not.

The question which was addressed by the options put before the judge – in particular by paragraphs (b) and (c) of clause 1(1) of the particulars of claim – was whether the shares of the three sisters of the testator who did survive him, Sarah, Elizabeth and Sophia, were vested indefeasibly on his death – so that the substitutional gift in favour of their children under clause 6 would not take effect, whether or not they failed to survive until the date of distribution, that is to say, the date of the testator’s granddaughter, Mary Kathleen – or whether the substitutional gift could take effect if they were not living at the date of distribution. Paragraph (c) of clause 1(1) of the particulars of claim provides for distribution on the former basis: that is to say, it provides for distribution on the basis that the one-eighth shares of Sarah, Elizabeth and Sophia shall go to their personal representatives and devolve as part of their respective estates. Paragraph (b) of clause 1(1), on the other hand, provides for distribution on the basis that the shares of those three sisters did not vest indefeasibly on the death of the testator, but were liable to defeasance in favour of the substitutional gift to their children if they did not survive to the date of distribution. That is, of course, what happened: as was probably inevitable, given that the testator’s granddaughter survived him by some 47 years, thereby postponing the date of distribution for almost half a century.

I have said that that question was addressed in part by the options put before the judge: it was addressed under paragraphs (b) and (c). But it does not seem to have been appreciated that that question would arise also if distribution were made on the basis that there were three, or perhaps four, shares rather than eight, or perhaps nine, shares. On a proper analysis, the question does arise in that case also: whether there are three shares or eight shares, the question remains – do the shares vest indefeasibly in the siblings who survived the testator, notwithstanding that they die leaving children who are living at the date of distribution?

The question which was not addressed at all by the three options offered to the judge by the particulars of claim was whether the children who were entitled to take by substitution under clause 6 were to be identified at the date of the testator’s death, or at the date of distribution. It was assumed that they were to be identified at the date of the testator’s death: so that if their parent had died before the date of distribution, the distribution was to those children who were living at the date of distribution, and to the estates of those children who had been living at the date of the testator’s death but had died before the date of distribution. The only children excluded on that basis were those who had themselves pre-deceased the testator. That assumption is revealed on a reading of paragraphs (a), (b) and (c) of the Particulars of Claim in conjunction with paragraph 11 of Mr Merryweather’s witness statement.

The judge identified the primary question at page 4 in the transcript of his judgment. After reminding himself of the words which he had to construe:

The judge went on to say this:

It is important to note that, at first sight, the decision in In Re Hannam – and in particular the passage which the judge set out – answers each of the three questions which I have identified. First, the brothers and sisters who were to take were those living at the death of the testator. The phrase did not include brothers and sisters who predeceased the testator; and, a fortiori, did not include brothers and sisters who were not living at the date of the will. Second, the shares of the brothers and sisters who survived the testator did not vest indefeasibly at the date of the testator’s death; but were liable to be divested if the primary beneficiary had died before the date of the distribution, leaving children then living to take his or her share by way of substitution. Thirdly, the children who were to take by substitution were those living at the date of distribution. It was not enough that they had survived the testator. As I have said, had the judge chosen to follow In Re Hannam he would have been led to the conclusion that none of the options offered under (a), (b) and (c) of the Particulars of Claim were apposite in the present case.

On the basis of the decision in In Re Hannam the estate would fall to be distributed on the basis of three, or perhaps four, shares. Those shares would fall into the estates of Sarah, Elizabeth and Sophia, unless there were children of Sarah, Elizabeth or Sophia who were living at the date of distribution in 1992. In fact there were two children of Elizabeth – Stanley Roberts and Catherine Roberts – who died after the date of distribution. By co-incidence – or perhaps not by co-incidence – the defendant, Mrs Kent, who has been joined in a representative capacity is a partner in the firm of solicitors who act for the respective personal representatives of Stanley and Catherine Roberts. But she does not, of course participate in these proceedings in order to advance arguments on behalf of those estates.

Having set out the passage from the judgment in In Re Hannam , the judge observed that the courts had moved on since the days when the rules of construction were applied rigidly to wills. He referred to comments of Lindley LJ in In Re Muster [1889] 43 Ch 569 at 572: that little assistance in construing a will was to be gained by looking at past cases in the books in which other wills had been construed. He referred to the well-known passage in the speech of Lord Hoffmann in ICS v West Bromwich Building Society [1998] 1 WLR 896. The judge reminded himself, correctly, that in construing a will the court had to put itself, as it were, in the armchair of the testator and interpret the words which the testator had used in the light of the facts and circumstances which were known to him. That is a well-known principle of construction. It was part of the law before the Administration of Justice Act 1982 and it applies as well to wills as it does to commercial contracts. It reflects the need, when seeking to understand what a testator meant, to take account of the circumstances in which the testator used the words which he did use.

The judge went on to say this at page 7 of the transcript of his judgment:

Those considerations led the judge to reject the construction for which paragraph (a) of clause 1(1) of the particulars of claim made provision. At page 8 of his judgment he said this:

For my part, I find those two paragraphs of the judgment difficult to reconcile with a declaration in the terms of paragraph (b) of clause 1(1) of the particulars of claim. If the testator was to be taken to be looking at the point of time of the distribution, the judge might have been expected to conclude that he should follow in In Re Hannam and hold that the estate be distributed on the basis of three, or perhaps four, shares; substitution of children being confined to those cases where children of deceased siblings were living at the date of distribution. But the judge clearly intended to make the order that he did, and it is plain, at least from the earlier passages of his judgment, why he rejected the option offered by paragraph (a) of clause 1(1) of the particulars of claim.

In my view, the judge was correct to reject the option offered by paragraph (a). If I may say so, without discourtesy to the arguments that have been advanced before us, the point does not really admit of much elaboration. The factual position at the date of the will was that the testator had one, or possibly two, brothers who were living. The one who was known to be living at the date of death was his brother David, but David is expressly excluded from the class of “my brothers and sisters” by the words “(excluding and excepting my brother David Jones)”. The other brother who may have been living was William. But the reference is to “brothers” in the plural; and a reference to “brothers” in the plural cannot properly be construed as a reference to William alone – even if he were then alive. The question is therefore whether the testator made a mistake in referring to “brothers” in the plural, rather than referring to “my brother William”; or whether he intended, by using the phrase “my brothers”, to include all his brothers – not only those living at the date when he made the will, but also those who had died before he made the will.

For my part, I do not think it appropriate to attribute to the testator – or, perhaps to whoever drafted this will – a knowledge of the reasoning and analysis in In Re Hannam – logically impeccable as it may be. It seems to me much more likely that when the testator used “brothers” in the plural, he meant to refer to “brothers” in the plural, which would necessary include deceased brothers. And if he intended to include deceased brothers in that class, then it would be absurd not to conclude, also, that he intended to include deceased sisters. Given that this is an ultimate trust, in circumstances where his own descendants have died out, there is no reason, as it seems to me, to assume the testator made a mistake; and every reason to give the words a generous interpretation so that the property goes back into his family as a whole without distinction between those who have died before him and those who have survived him. So, I would uphold the judge’s view that paragraph (a) of clause 1(1) in the particulars of claim is not a construction which can be given to the words used.

That would be sufficient to dismiss the appeal; which is brought only by way of challenge to the judge’s order seeking by substituting paragraph (a) for paragraph (b). But it would not, I think, be appropriate in this case to leave the matter there. To my mind it is very hard to see how the judge could have reached the conclusion that paragraph (b) reflected the appropriate construction of this will. It seemed to me that the real choice for the judge was between holding that all the children of those siblings who died leaving children came in to share at the date of distribution (whether or not then living); and holding that only those children who were living at the date of distribution had a claim to be substituted.

In principle, it seems to me very much more likely that a substitution clause was intended to take effect in favour of the living, rather than in favour of the dead. I can see very little reason to think that the testator intended to shift, by way of substitution, assets from the estates of a class of primary beneficiaries who were dead to the estates of a class of secondary beneficiaries who were also dead. The purpose of substitution, as it is normally understood, is that it enables those who are living at the time for distribution, when the primary beneficiary is dead, to take the share that the primary beneficiary would have taken if he had survived.

But that is not a question which we can decide on this appeal, and I do not decide it. It is not a question which we can decide, because there is no representation before us to argue the two sides of that point. The point was not identified at the time when the representation order was sought; and it does not seem to have been identified by the court when the representation order was made. Indeed, I doubt if it was identified until it was raised by the court in the course of argument before us this morning. The point appeared to come as something of a surprise to counsel.

Nevertheless, it is a point which needs to be resolved before the executors can safely distribute this estate. For those purposes, I would propose setting aside those parts of the judge’s order which construe this will and remitting the matter back to him for further consideration in the light of our judgments; with the expectation that there will be an application for, and a direction as to, the appropriate representation to argue the point.

So I would dismiss the appeal. But I would set aside the order, with the direction I have mentioned.

LORD JUSTICE THOMAS: I agree. In the course of argument, the issue arose before us as to the extent to which principles applicable to the construction of commercial contracts and notices should be applicable to the construction of wills, given the way the learned judge had approached this case as my Lord has set out. We were referred to the judgment of Neuberger J in Cooksley v Cole 8 November 2002 (unreported), and in particular to paragraphs 29 to 31 of that judgment. There is considerable force in the propositions therein set out, but, to the extent that they are controversial, it is not necessary to consider them in this case. In my view the matter can be decided in exactly the way my Lord has set out, for the reasons he has given. It is neither necessary nor desirable to express any further view.

LORD JUSTICE LLOYD: I also agree. In particular, I would agree that the learned judge was right, for the reasons that he gave, to construe the reference in clause 6 of the will to brothers and sisters as being intended to include those siblings who had already died, because otherwise the reference to “brothers” is not fairly explicable. However, as my Lord, Lord Justice Chadwick says, the judge’s declaration does not necessarily follow from this even though it was the only option offered to him in this event. What the judge said at the end of his judgment, which my Lord has quoted, about the date of distribution points to a lack of logic in a failure to consider whether survival to the date of distribution is a necessary element in the entitlement of children under the words of substitution.

It can be said that any of the various possible outcomes in this case is likely to be somewhat arbitrary. That is perhaps not surprising in relation to a default provision, such as clause 6, which was intended to operate only if the primary gift of capital under clause 5(c), and for that matter the primary gift of the real property under clause 3, did not take effect after one life interest in respect of the real property and two successive life interests in the case of the residue; and given the fact that the subsistence of those life interests meant that distribution was deferred by some 47 years after the death of the deceased.

It seems to me that there is a substantial case to be made for the proposition that survival to the date of distribution was a necessary element in the entitlement of any child to take by way of substitution, and in that way to replace the entitlement that would otherwise, on the face of it, arise of the estate of the by then deceased sibling. As my Lord says, we cannot decide that point, and in the light of the debate between the court and counsel, and as my Lord suggests, it seems best to set aside paragraph 2 of the judge’s order, even though we agree with him that one-eighth, or possibly one-ninth, shares do appear to be correct, and to remit the matter to him for reconsideration of the point, including of the proper representation.

I would only mention in concluding that this is a case in which use was made, sensibly and valuably, of the provision, introduced into the Civil Procedure Rules in 2000, of Rule 8.2A enabling claimants to issue a Part 8 claim form initially without naming a defendant; and that was coupled with the use in a constructive fashion of Rule 19.7. We were told that there was considerable difficulty in identifying suitable representatives in the various estates which may be competing for shares of the fund held subject to the trusts of the late Mr Jones’ will. That is a situation which may arise in a number of different cases and where the appointment of a person who is not a member of the class, such as a solicitor as Mrs Kent is, is a valuable use of the court’s powers under Rule 19.7. This is a course which was advocated in a paper dated 18 July 2003 prepared by the Pensions Litigation Court Users’ Committee, entitled: Proceedings relating to trusts – Defendants representing classes of beneficiaries – some problems and a possible solution. I know not whether what was done in the present case was in any way influenced by that paper. I am aware that the paper has had some circulation within the profession. It is currently available at http://www.hrothgar.co.uk//YAWS/practice/defrep.htm . This approach can be useful in pension cases, but the present case demonstrates that it can also be valuable, perhaps less commonly, in the case of litigation about a family trust or an estate.

It will be for the judge, when the case is remitted, to consider how to use the same techniques to ensure the appropriate representation of the interests that do need to be before the court, so that all relevant parties with claims to be beneficiaries or potential beneficiaries are bound by the court’s ultimate conclusion, and so that the court can reach a satisfactory conclusion giving the necessary protection to the trustees as claimants, so that they may distribute on the correct basis. As I say, I would agree as to the disposition of the appeal with that which my Lord, Lord Justice Chadwick, proposes.