This is my judgment on the trial of a claim about alleged mutual wills. The testatrix whose will, made in July 2000, is in question was a lady called June Clark. She died on 8 February 2016, having made a later will dated 12 December 2014, of which probate was granted on 22 April 2016. The claimants are the two daughters of the testatrix. The three defendants are two of the grandsons of the testatrix (sons of the second claimant) and the partner of one of them. The first defendant is the executor to whom probate was granted. The second and third defendants are executors named in the will who did not prove it, but to whom power so to do was reserved.

However, the claimants in this case allege that earlier wills made by the testatrix and her husband, Bernard Clark, on the 25 July 2000 are in fact mutual wills, falling within the scope of the equitable doctrine of the same name. Mr Clark died on 16 May 2001. In the years following his death, the testatrix made more than a dozen further wills, beginning in 2004. Under the will of 2014 that was proved, the first claimant receives a legacy of £10,000, and the second claimant receives a legacy of £30,000. Under the will of the 25 July 2000, the claimants share the testatrix’s estate equally. The value of the testatrix’s estate was sworn for probate at not more than £324,000 gross, £213,000 net. It is on any view below the county court limit of £325,000. However, during the trial I formed the view that the complexity of the factual and legal issues justified transferring the case from the county court to the High Court, and I so ordered. The major asset of the estate of the testatrix was the former council house occupied by the testatrix and her husband, purchased by them under the “right to buy” scheme of the 1980s.

The claim was commenced by claim form dated 20 May 2016, seeking a declaration that the estate of the testatrix was held on trust for the claimants in equal shares. The particulars of claim accompanying the claim form pleaded as follows:

“7.

The execution of the Mutual Wills was subject to the agreement set out below.

8.

Bernard Clark died on 16 May 2001, and under the terms of his Mutual Will, and in accordance with the agreement reached between him and the Deceased and as reflected by the Mutual Wills, all of his Estate passed to the Deceased. A Grant of Probate was not necessary to achieve this.

9.

Despite the agreement reached between the Deceased and Bernard Clark, as reflected by the Mutual Wills, the Deceased executed the Later Will on 12 December 2014, by which she bequeathed the joint estate of Bernard Clark and herself to numerous Beneficiaries.

10.

It had been agreed by the Deceased and Bernard Clark at the time that they executed the Mutual Wills that neither would revoke their Wills and that neither would be entitled to change the terms of their Wills or to bequeath their joint estates to anyone other than the beneficiaries detailed within the Mutual Wills.

11.

That agreement was explained to the first claimant on the 25 July 2000 by both Bernard Clark and the Deceased. Later in the same day, the agreement was also explained to the second defendant by Bernard Clark and the Deceased in the presence of the first claimant.”

It is important to notice that the only case made is that of mutual wills. No allegation is made (for example) that, in making the 2014 will that was proved, the testatrix lacked capacity, did not know and approve the contents of the will, or was subject to undue influence or other pressure in making it. That means that the only factual matters which are important in this case are those which relate to the circumstances of the making of the will in July 2000. This further means that the greater part, in fact nearly the whole, of the evidence in the witness statements put forward by the defendants, and a considerable amount of the evidence in the witness statements put forward by the claimants, is irrelevant.

A defence was filed on 20 June 2016. In part, it says this:

“It is denied that the Deceased and Mr Clark executed mutual Wills or that any agreement to enter into mutual existed [sic] and the Claimants are put to proof on this matter. The express terms of the 2000 Will, being both contemporaneous and verified by Mr Clark, unambiguously state that the trustees should ‘pay [the] residuary estate to [the Deceased] absolutely and beneficially and without any sort of trust obligation’, clearly distinguishing the 2000 will from mutual Wills.”

At the pre-trial review in this matter on 12 July 2017, I heard an application to call the solicitor who attended on the testatrix in 2000, but decided, for reasons given at the time, that the application should be refused. I also decided that an application to exclude significant parts of the evidence as irrelevant should be adjourned to the trial judge, in the event, me.

The witnesses called at the trial were: the first claimant, Ann Legg, the second claimant, Lynn Burton, the first defendant, Aaron Burton, the second defendant Victoria Brooks, and the third defendant Michael Burton, and the cousin of the testatrix Maryanne Dean. Because of the importance of the oral evidence, I give my impression of the witnesses here.

Ann Legg was softly spoken and at first cautious as a witness. But she was clear, firm, open and straightforward in the evidence that she gave. Her style of answering did not change depending on whether the answers favoured her case or not. In my judgment, she was trying to help the Court to the best of her ability, and was telling the truth so far as she understood it.

Her sister, Lynn Burton, was a fluent but more enigmatic witness. I did not get the same impression of openness as I had with Ann Legg. She clearly weighed up the possible answers to questions and knew which answers favoured her case and which the other side. Whilst I do not think that she was telling me deliberate untruths, she gave me the impression of having convinced herself that she was in the right. I regret to say that I did sometimes have the impression that she was not telling me everything she knew.

Victoria Brooks was a voluble and opinionated witness, who clearly enjoyed being in control. But she listened carefully to questions, and gave clear and straightforward answers. Although she has a financial interest in the outcome of the case, and her own view of the rights and wrongs of the situation, I thought that, apart from Maryanne Dean, she was the most disinterested of the witnesses. Again, I thought that she was doing her best to assist the court, and telling me the truth, at least from her point of view.

Maryanne Dean was a cautious and careful witness who had relatively little of relevance to say. But I am sure that she was telling me the truth so far as she knew.

Aaron and Michael Burton were slow and relatively inarticulate witnesses. To my mind they seemed to be easily led. In my judgment it would not be safe to place any significant reliance on what they said, unless it was confirmed by an objective source. In fact, however, they had almost nothing relevant to say on the important issues in the case.

I will first set out here the facts which are either undisputed or are at least clearly established by the evidence, without the need for further discussion. On 25 July 2000 the testatrix and her husband (father of the claimants) each made a will, the one mirror of the other. Essentially each will gave the property of the maker to the surviving spouse absolutely, but in the case that the spouse did not so survive the maker, then to the two claimants in equal shares absolutely. I set out the relevant clauses later.

Those wills remained unchanged at the date of the death of the husband of the testatrix on 16 May 2001. Following his death, and beginning in 2004, but mostly in the years 2011-14, the testatrix made no fewer than 13 further wills. There were three in 2004 (two five days apart), four in 2011, one in 2012, one in 2013 and another four in 2014. In broad terms, these progressively favoured the defendants at the expense of the claimants, although the last one rowed back somewhat on earlier wills.

The most disputed allegation, which I must examine and on which I must rule specifically in more detail, concerns the making by the testatrix of the will of 25 July 2000. At that date, relations between the claimants and the testatrix were good, and the grandchildren were still minors, living with their mothers, the claimants. The claimants say that the testatrix and her husband (their father) expressly agreed that the wills that they were making could not be revoked, but were “set in stone”. The first claimant was present before and also when the wills were executed, and the second claimant arrived shortly afterwards. The former gave evidence of what happened before and at the time of execution, and both gave evidence of what happened afterwards. Given their ages at that time, the grandchildren were in no position to give evidence to me now to the contrary.

After the death of her father, the claimants were involved as executrices with the administration of his estate. The first claimant did the majority of the work, because at that time the second claimant had a small child to look after. No grant was ever taken out, but there was no need, as no significant transactions had to be undertaken with third parties. An executor’s title derives from the death and not from the probate. The first claimant also assisted her mother, the testatrix, with financial matters in particular, being added to her bank accounts so that she could sign cheques and pay bills for her. However, it is clear that, later on, relations between the testatrix and her daughters, the claimants, began to deteriorate. The first claimant dates this in her case to June 2010, when her daughter Deborah became seriously ill. The first claimant was less able to spend time visiting and supporting the testatrix. In September 2010 Deborah was diagnosed with cancer, from which she died in August 2012.

At about the same time, the children of the second claimant (and in particular the first and third defendants and the partner of the third defendant, the second defendant) appear to have become more important in the life of the testatrix. As the testatrix became older, she needed more attention and more looking after, which her own daughters were less able to supply, but on the other hand her grandchildren (and their partners) were.

It is also clear that relations between the claimants and the defendants are non-existent. Allegations of criminal activity have been made by the claimants against the defendants. But it is no part of the function of this court to deal with those matters. What the defendants may or may not have done in the years following the making of the will of July 2000 is not relevant to the question of what happened at that time, and this court will not resolve any dispute between the parties as to these matters: cf Birch v Curtis [2002] EWHC 1158 (Ch), [52]. They are matters, if at all, for the criminal law and the criminal courts, and not for me. Accordingly, I concentrate on the events surrounding the making of the will in July 2000.

A number of authorities were cited to me. These include Re Cleaver deceased [1981] 1 WLR 939, Re Dale deceased [1994] Ch 31, Re Goodchild deceased [1997] 1 WLR 1216, CA, Lewis v Cotton [2001] 2 NZLR 21, CA of NZ, Olins v Walters [2009] Ch 212, CA, Charles v Fraser [2010] EWHC 2154 (Ch), and Fry v Densham-Smith [2010] EWCA Civ 1410. From these authorities I deduce the following propositions. In order to succeed in a claim that a will falls within the equitable doctrine of mutual wills, and is accordingly binding on the estate of the testator despite a subsequent change in that will, the claimant must prove, on the balance of probabilities, that the testator made a legally binding agreement with the other testator that both would make their wills in a particular form (not necessarily the same) and that they would not revoke them or (depending on the terms of the agreement) change them without notice to the other or others sufficient to enable that other or others to change their own wills as well, that they made their wills in that particular form and that they did not revoke them (or change them without such notice), and the first of the testators to die did so, not having revoked (or changed) his or her own will.

The use of the phrase “legally binding agreement” in the authorities demonstrates that there is a crucial difference between an obligation which is legally binding, and which will be enforced by the court, and an obligation binding in honour only. The latter may be called a moral obligation, or – as in some of the authorities – an “honourable engagement” see eg Lord Walpole v Lord Orford (1797) 3 Ves Jun 402, 419; Re Cleaver deceased [1981] 1 WLR 939, 947G.

Only one qualification needs to be made to this. This is the question whether, when reference is made to a “legally binding contract”, it will be sufficient if the otherwise binding agreement is made orally. A contract to dispose of an interest in land must be made in writing: Law of Property (Miscellaneous Provisions) Act, s 2. But under the doctrine of mutual wills the agreement imposes an equitable obligation on the second testator to die capable of taking effect by way of a constructive trust, and constructive trusts are excepted from the requirements of writing in relation to contracts for the disposal of an interest in land: see Law of Property (Miscellaneous Provisions) Act 1989, s 2(5).

In Healey v Brown [2002] EWHC 1405 (Ch), however, David Donaldson QC held that, where (as it was there) the gift in the disputed will was expressly a gift of land or an interest in land, there could be no legally binding contract unless it was in writing and complied with s 2 of the 1989 Act. The exception for constructive trusts in s 2(5) did not avail, because under the doctrine of mutual wills the constructive trust arose as a result of there being a valid contract, and of course there was no such contract unless s 2 were complied with. In Olins v Walters [2007] EWHC 3060 (Ch) Norris J held that that difficulty did not arise where the gift in question (as in that case) was a gift of residue rather than a gift of land. The decision was unsuccessfully appealed (see [2009] Ch 212), but so far as I can see this point was neither raised before nor dealt with by the Court of Appeal.

For my part, it does seem rather capricious, even unprincipled, to make the success of a claim for mutual wills depend on whether a gift of land in a will is drafted as a gift of a particular interest in land or a particular immovable (when it will fail unless the contract between the testators is in signed writing) or as a gift of residue happening to contain only immovables (when no writing is needed, and it will be valid). The form of the gift should not defeat its substance. In the present case, as will be seen below, the gift is drafted as one of residuary estate, so that strictly, on the view adopted by Norris J in Olins v Walters, the point does not arise.

But I cannot help thinking that, when Leggatt LJ in Re Goodchild deceased [1997] 1 WLR 1216, 1225G, referred to the necessity of an agreement, in order to justify a constructive trust being imposed, not on the property passing from testator 1 to testator 2, but on testator’s 2’s existing beneficially owned property, he cannot have had in mind the possibility that testator 2 might make a promise, intended to be relied upon, to deal in future with her own beneficial property in a certain way, on which testator 1 relied to his detriment by making his will as (informally) agreed, and then dying, so putting it out of his power to alter his will in future. In other words, that judge was not excluding the possibility that the necessary equitable obligation to bind the conscience of testator 2, and so call into existence the constructive trust of mutual wills, might arise from a proprietary estoppel rather than from a contract. It seems to me that he must instead have been thinking either of the difference between mutual wills and secret trusts (mere communication of the gift being enough in the latter case), or alternatively of the distinction to which I have already alluded between a legally binding obligation and an “honourable engagement”. But a proprietary estoppel, if established, is legally binding.

I do not overlook the passage in the judgment of Morritt LJ in Re Goodchild at 1229H-1230C where he discusses Lloyds Bank plc v Rosset [1991] 1 AC 107, HL. Then, at 1230 C-D, Morritt LJ says:

“The doctrine of mutual wills is anomalous. The bequest of his entire estate by a husband to his wife absolutely and beneficially with a gift over of whatever was left at her death could not take effect in accordance with its terms. Either the interest taken by the wife would be limited or the gift over would be void as repugnant to the absolute and beneficial nature of the gift. Similarly the bare promise of the wife to leave her property by will in a particular manner would be unenforceable for any will she then made would be revocable under the Wills Act 1837. In my judgment, if these principles are to be excluded in the case of mutual wills it is essential that there should be a contract to that effect. In my view that is what both principle and the authorities require.”

But, with respect, whilst the decision of the court in that case is (if one at my level may say so respectfully) unexceptionable, this analysis takes no account of the doctrine of proprietary estoppel, as typified in, say, Thorner v Major [2009] 1 WLR 776, HL. That is not surprising, since Re Goodchild itself did not give rise to any such possibility. The courts both at first instance and on appeal found that there was no sufficient evidence of any binding promise not to revoke the will each party was making.

It is of the essence of proprietary estoppel in such a case that a promise, intended to be acted upon by the promisee, and in fact acted upon to the detriment of the promisee, to leave the residue of an estate in a particular way can be enforced in equity, although the will itself has subsequently been revoked or altered, or indeed never made at all. Thorner v Major itself is an example. The fact is that, in this context at least, there is no contract for the disposal of an interest in land that cannot be replicated in its effect in equity by a proprietary estoppel. So, for practical purposes, if you need a contract to achieve an object, a proprietary estoppel should equally serve your purpose. And a proprietary estoppel of land does not require writing: see eg Yaxley v Gotts [2000] Ch 162, CA. On that basis the distinction drawn in Olins v Walters would be unnecessary.

I should say a few words about the standard of proof in a mutual wills case. In Birmingham v Renfrew (1937) 57 CLR 666 at 674, Latham CJ in the High Court of Australia said that

“Those who undertake to establish such an agreement [ie of mutual wills] assume a heavy burden of proof”.

But in fact the standard of proof in a mutual wills case is the usual civil standard, that is, on the balance of probabilities: Re Cleaver deceased [1981] 1 WLR 939, 948. What Latham CJ was apparently referring to was the fact that, where a thing is inherently improbable, it takes more cogent evidence to persuade a court to find that the balance of probabilities does indeed lie in that direction.

This is exemplified by Lord Hoffmann’s famous example in Home Secretary v Rehman [2003] 1 AC 153, [55]:

“[S]ome things are inherently more likely than others. It would need more cogent evidence to satisfy one that the creature seen walking in Regent's Park was more likely than not to have been a lioness than to be satisfied to the same standard of probability that it was an Alsatian.”

A similar idea had been earlier expressed by Lord Nicholls in the context of family law, as

“the more serious the allegation the less likely it is that the event occurred and, hence, the stronger should be the evidence before the court concludes that the allegation is established on the balance of probability”: Re H and others (Minors) (Sexual Abuse: Standard of Proof) [1996] AC 563, 586; see also per Lord Hoffmann in Re B (Children) [2009] 1 AC 11, [5]-[15].

The evidence available to me on the crucial issue is of three kinds. First, there is the evidence of the will itself. Secondly, there is the direct evidence of the two claimants. Thirdly, there is a very limited amount of indirect evidence from other witnesses or from the surrounding circumstances more generally. I deal with each in turn.

So far as the will itself is concerned, it is short, no more than one and a half pages long, and plainly professionally drafted by a solicitor, who has attested the execution, together with his clerk. It appoints the claimants to be executrices and trustees, and gives the entirety of the estate of the testatrix to them. It then proceeds:

“4.

MY TRUSTEES shall pay my residuary estate to my Husband BERNARD ROY CLARK absolutely and beneficially and without any sort of trust or obligation

5.

IF my said Husband shall have predeceased me then my Trustees shall pay my residuary estate to such of my Daughters [the claimants] as shall survive me and if more than one in equal shares absolutely Provided that if either of them shall have predeceased me but leaving a child or children living at my death my Trustees shall pay the share of my residuary estate to which they would have been entitled had they survived me for such child or children as and when they attain the age of twenty-one years and if more than one in equal shares absolutely”.

The will is made on the same day as that of the husband of the testatrix (and father of the claimants). It is attested by the same persons, and is in identical form, as her husband’s will, except that for the reference in the first will to the testatrix’s husband there is substituted a reference to the testatrix in the second. But the authorities make clear that that is not enough for the doctrine of mutual wills to apply: see Gray v Perpetual Trustee Co Ltd [1928] AC 391, PC; Renfrew v Birmingham (1937) 57 CLR 666, 674; Re Cleaver deceased [1981] 1 WLR 939; Re Goodchild [1997] 1 WLR 1216, 1224; Lewis v Cotton [2001] 2 NZLR 21; Charles v Fraser [2010] EWHC 2154 (Ch). Nevertheless this will does not go any further, eg by stating that it is a mutual will, or that any agreement has been entered into with any other person concerning its non-revocation.

Indeed, the defendants say that the gift in clause 4 of the testatrix’s will, requiring payment by the trustees to the husband of the testatrix “absolutely and beneficially and without any sort of trust or obligation”, makes clear that “the passing of the Estate from the testatrix to her husband was not impressed with any fiduciary trust obligation in respect of future disposal of the remainder”. The defendants might also have referred to the equivalent clause in the husband’s will, which required the husband’s estate to pass to the testatrix similarly “absolutely and beneficially and without any sort of trust or obligation”. After all, that was what actually happened, because the testatrix survived her husband. But I agree with the claimants when they submit that, if there is a mutual will trust, it arises outside the will, and so words such as this in the will would not affect it. In any event, the clause referred to is a standard form clause, regularly included in wills of this kind. I cannot regard its inclusion here as negativing the possibility of mutual wills.

Much more important, in my view, is the evidence of the claimants of what was said and done on the day that the wills of the testatrix and her husband were executed. The witness statement of the first claimant, which stands as her evidence in chief, emphasises the close relationship she had with her parents at that time. They invited both claimants to be present at the signing of the wills. Unfortunately only the first claimant arrived in time. She heard the solicitor explain the will to her parents.

According to her statement, her father then asked the solicitor

“5.

… whether everything was ‘set in stone’ because he said neither he nor my mother wished for anything to be changed again. Mr Reid told my father that the law cannot stop someone from changing their Will in the future. He mentioned that he was aware that my mother and father never wanted to change the terms again, and that their trust in one another not to make any future changes was enough.

6.

Happy that everything was as it should be, my mother and father signed multiple copies of their Wills. Mr Reid and his secretary then also signed them as witnesses. Mr Reid then took an original copy of each Will with him for safekeeping at his office.”

The second claimant arrived subsequently, and her parents explained the terms of the wills to her. According to the first claimant’s witness statement,

“7 … they also explained their agreement that these Wills were to be ‘set in stone’ and never changed again. They said that they had decided to make a promise to each other never to change the Wills again so that each could feel confident that their wishes would still be respected after the other had died.”

Their parents allowed the claimants to read their copies of the wills. The first claimant in her statement then says

“8.

… With my parents making such a point about the Wills never being changed again, I was concerned that the documents did not make this point. I am no expert in the drafting of Wills, but I would have expected there to be some mention of this agreement given that my parents had made such a point of it. I asked my Father (as my mother had gone into the kitchen at this point) if maybe a clause should have been put in the WWills to prevent them from changing things in the future. I asked what would happen if one of them should die, and the other remarry. What would prevent the survivor from changing the terms to benefit his or her new spouse? My father stated that Mr Reid did not seem concerned about this possible complication. He reminded me that he had asked Mr Reid about the wills being ‘set in stone’ and I had heard Mr Reid’s answer. I said I didn’t think the answer was satisfactory, but my Father seemed dismissive and told me that there was no need for any further complication as both he and my Mother had discussed all scenarios and promised each other that each one would adhere to the other’s wishes. He said that they had already agreed that they would not make any further changes to their wills, and that they would not revoke them – even after the other had passed away. He said he had no intention of making further changes, and he knew that my Mother didn’t either because he had already promised her and she had already promised him.

9.

My Mother actually heard this comment, and she shouted through from the kitchen ‘No I bloody won’t change it either’…”

In her own witness statement, the second claimant also says that her relationship with her parents was very close. She similarly gives evidence of being asked to visit her parents when they executed their wills. However, as the first claimant says, when the second claimant arrived the solicitor had been and gone. Her parents explained the terms of the wills to her. Then she says:

“8.

They told us that they were certain that the terms expressed in their wills were what they wanted to collectively achieve, and that as a consequence, they had agreed that – even if their circumstances changed in the future – they would never change their Wills again. They were concerned to make sure that their intentions is to ultimately benefit Ann and I would not be frustrated, so that one or other of them could not subsequently change their minds and alter the terms of provision once the other had passed away. My Father said that they had both discussed this in great detail, and had made solemn promises to each other to make sure that this did not happen.”

Next, the second claimant refers to a point made by the first claimant in her statement:

“9.

At one point during this discussion, my sister asked my dad if there should have been a clause in the Wills confirming the binding agreement never to change the terms. He made reference to something that the solicitor had said on this subject, and re-assured Ann not to worry as they had both made promises to each other. He said that he had solemnly promised my Mother that he would never change the terms again, and that she had also made the same promise to him. He said he would not have made the will in the terms that he had if she did not promised to abide by their agreement. He said he was confident that my Mother would respect his wishes.

10.

The room we were in was the backroom to the house, which is open planned, and my Mother was stood in the kitchen area. Upon hearing this she responded quite angrily by saying ‘No I bloody won’t change it’. She took pains to remind Ann and I that they had only purchased their house in the first place so that they could provide Ann and I in the future, and so their promises that their wills were ‘set in stone’, and would never be changed, was the culmination of plans that they had started around 14 years earlier.”

What should the court’s approach be to such evidence in a mutual wills case? In Renfrew v Birmingham (1937) 57 CLR 666, 674-75, Latham CJ said:

“It is easy to allege such an agreement after the parties to it have both died, and any court should be very careful in accepting the evidence of interested parties upon such a question. Perhaps most husbands and wives make wills ‘by agreement,’ but they do not bind themselves not to revoke their wills. They do not intend to undertake or impose any kind of binding obligation. The mere fact that two persons make what may be called corresponding wills in the sense that the existence of each will is naturally explained by the existence of the other will is not sufficient to establish a binding agreement not to revoke wills so made (In re Oldham [1925] Ch 75; Gray v Perpetual Trustee Co. [1928] AC 391; and see Lord Walpole v Lord Orford (1797) 3 Ves 402, where attention is directed to many considerations which may go to show that in a particular case no binding agreement was intended).”

In Re Cleaver [1981] 1 WLR 939, 949, Nourse J said:

“It is clear from [Birmingham v Renfrew], if from nowhere else, that an enforceable agreement to dispose of property in pursuance of mutual wills can be established only by clear and satisfactory evidence. That seems to me to be no more than a particular application of the general rule that all claims to the property of deceased persons must be scrutinised with very great care. However, that does not mean that there has to be a departure from the ordinary standard of proof required in civil proceedings. I have to be satisfied on the balance of probabilities that the alleged agreement was made, but before I can be satisfied of that I must find clear and satisfactory evidence to that effect.”

More recently, in Charles v Fraser [2010] EWHC 2154 (Ch), another case where mutual wills were alleged, Mr Jonathan Gaunt QC, sitting as a deputy High Court judge, said this:

“64.

In my judgment, a Court has to approach oral evidence of the kind that was given by and on behalf of the Claimants in this case warily and with appropriate scepticism. First, I bear in mind the inherent improbability of a testator being prepared to give up the possibility of changing his or her will in the future, whatever the change of circumstances. Secondly, I take into account that a number of the witnesses who gave evidence that the sisters said they had made an agreement and that the wills could not be changed had a financial interest in the outcome of the case. I do not mean by that that I think for a moment that anybody was being dishonest. My impression of all the ladies who gave evidence before me was that they gave their evidence honestly and scrupulously and, having taken the oath, would have been shocked at any suggestion that they might do otherwise. Nevertheless, one is aware from experience of the ability of the human mind to ‘remember’ what a person wishes to remember.

65.

Thirdly, one is only too well aware of how easy it is, when witness statements are being drafted by solicitors, for the recollections of the witnesses to be subtly improved in the direction the party calling them wishes to go. There are some kinds of case, of which perhaps this is an example, where it would make it easier for the Court to assess the evidence if it were given in chief viva voce instead of by carefully drafted witness statements.

66.

Fourthly, if the two sisters told many of their friends and relations about their agreement and if, as Mr. Last said Ethel told him they had, they told their Solicitor, it is remarkable that he did not include any recital of it in the wills he drafted for them or make any other record of it…”

As to the second point, concerned with financial interests, I do of course bear in mind that all the witnesses, on either side, with one exception (Maryanne Dean) have a financial interest in the outcome of this case, though only the claimants gave direct evidence bearing on the question of an agreement between the testators.

As to the further point there made, dealing with the fallibility of memory, I was referred to the recent decision of Leggatt J in Blue v Ashley [2017] EWHC 1928 (Comm), where the judge had to deal with the allegation of an oral agreement by a businessman to pay a sum of millions of pounds in certain circumstances to a business acquaintance with whom he was then drinking in a public house. (The judge held that the claim failed on the facts.) In doing so the judge referred to his own earlier decision in Gestmin SGPS SA v Credit Suisse (UK) Limited [2013] EWHC 3560 (Comm), [16]-[20].

In that earlier case the judge had said this:

“16.

While everyone knows that memory is fallible, I do not believe that the legal system has sufficiently absorbed the lessons of a century of psychological research into the nature of memory and the unreliability of eyewitness testimony. One of the most important lessons of such research is that in everyday life we are not aware of the extent to which our own and other people's memories are unreliable and believe our memories to be more faithful than they are. Two common (and related) errors are to suppose: (1) that the stronger and more vivid is our feeling or experience of recollection, the more likely the recollection is to be accurate; and (2) that the more confident another person is in their recollection, the more likely their recollection is to be accurate.

17.

Underlying both these errors is a faulty model of memory as a mental record which is fixed at the time of experience of an event and then fades (more or less slowly) over time. In fact, psychological research has demonstrated that memories are fluid and malleable, being constantly rewritten whenever they are retrieved. This is true even of so-called 'flashbulb' memories, that is memories of experiencing or learning of a particularly shocking or traumatic event. (The very description 'flashbulb' memory is in fact misleading, reflecting as it does the misconception that memory operates like a camera or other device that makes a fixed record of an experience.) External information can intrude into a witness's memory, as can his or her own thoughts and beliefs, and both can cause dramatic changes in recollection. Events can come to be recalled as memories which did not happen at all or which happened to someone else (referred to in the literature as a failure of source memory).

18.

Memory is especially unreliable when it comes to recalling past beliefs. Our memories of past beliefs are revised to make them more consistent with our present beliefs. Studies have also shown that memory is particularly vulnerable to interference and alteration when a person is presented with new information or suggestions about an event in circumstances where his or her memory of it is already weak due to the passage of time.

19.

The process of civil litigation itself subjects the memories of witnesses to powerful biases. The nature of litigation is such that witnesses often have a stake in a particular version of events. This is obvious where the witness is a party or has a tie of loyalty (such as an employment relationship) to a party to the proceedings. Other, more subtle influences include allegiances created by the process of preparing a witness statement and of coming to court to give evidence for one side in the dispute. A desire to assist, or at least not to prejudice, the party who has called the witness or that party's lawyers, as well as a natural desire to give a good impression in a public forum, can be significant motivating forces.

20.

Considerable interference with memory is also introduced in civil litigation by the procedure of preparing for trial. A witness is asked to make a statement, often (as in the present case) when a long time has already elapsed since the relevant events. The statement is usually drafted for the witness by a lawyer who is inevitably conscious of the significance for the issues in the case of what the witness does nor does not say. The statement is made after the witness's memory has been ‘refreshed’ by reading documents. The documents considered often include statements of case and other argumentative material as well as documents which the witness did not see at the time or which came into existence after the events which he or she is being asked to recall. The statement may go through several iterations before it is finalised. Then, usually months later, the witness will be asked to re-read his or her statement and review documents again before giving evidence in court. The effect of this process is to establish in the mind of the witness the matters recorded in his or her own statement and other written material, whether they be true or false, and to cause the witness's memory of events to be based increasingly on this material and later interpretations of it rather than on the original experience of the events.”

The claimants were cross-examined at some length during the course of the first day of the trial. As a result, I am bound to record that the substance of what they said in their witness statements was not materially shaken. I bear in mind the strictures of Leggatt J in relation to the limitations of memory. We all remember what we want to remember. In Olins v Walters [2007] EWHC 3060 (Ch), Norris J said of one witness:

“I have a deep sense that her evidence is not based upon a real recollection of two brief incidents (putting her signature on a document one or two decades ago) but upon a reconstruction of what she would have done having regard to her inability to recollect the Deceased's attendance at Battersea and her willingness to do whatever Mr Walters requested. I am also alert to the danger that recollection can be coloured by the context in which it is invited.”

But there are too many features of the story in the present case which ring true for me to reach the conclusion that this story has been misremembered in any significant particular. These include the justifiable pride of Mr and Mrs Clark in achieving home ownership from humble beginnings, the purpose of acquiring it to have something to leave, the often-repeated desire to pass this on to the next generation, the wish in their retirement for Mr and Mrs Clark to engage professional help to ensure that this happened, the attendance of the solicitor and his clerk at the family home, the need to engage the children themselves (the claimants) in the process, and above all (for first-time home-owners and testators) the desire for doing it once and doing it right. And, although he was not ill when he made his will, Mr Clark in fact died only about ten months afterwards. For this family these events were very unusual, and taken together constituted a ‘one-off’. Their uniqueness makes it more likely that they would stick in the claimants’ memories. In my judgment, what they say is not just plausible but, taken as a whole, in my judgment very likely.

In addition, the second claimant in cross-examination brought out an incident which occurred in 2004. According to her evidence (which I accept), the testatrix told her that she was thinking of adding her grand-daughter Michelle (the first claimant’s daughter) to the will. It appears that Michelle had just become engaged to be married. The second claimant however reminded the testatrix that she had promised their father not to change her will, and the testatrix having reflected on this said that she would not after all do so.

Yet, during the trial it became apparent that in June 2004 the testatrix had made two wills, five days apart. In the first she had added Michelle both as an executrix and also to share the house or its value. In the second she had removed her as a beneficiary of the house. It is clear that, when the testatrix told the second claimant that she was thinking of adding Michelle, she was not quite telling the truth. She had already done so. Having then been reminded of her promise to her late husband, a few days later she changed the position back again. This incident, supported as it is by the evidence of the second claimant and the existence of the two wills, supports the view that, at that time at least, the testatrix accepted that in 2000 she had given her word not to change her will.

The second claimant was incidentally criticised because her witness statement does not refer to this incident in 2004. The relevant evidence was elicited only in cross-examination. The implication of the criticism is that this story is a recent invention by the second claimant. I reject any such implication. The second claimant explained, as the first claimant had done before her, that each of them typed up her own first draft of her witness statement on a computer and emailed it to their solicitors, who are based in Manchester. The solicitors presumably made some suggestions for alteration but the style, the substance and the formatting remained those of the claimants themselves. It does not surprise me that the relevance of the 2004 incident did not strike the second claimant at the time of preparing her own statement and submitting it to the solicitors. Nor am I surprised that the solicitors did not elicit it from the second claimant so that it could be included. This incidentally deals with Jonathan Gaunt QC’s third point in Charles v Fraser, cited above.

I have already given my views of the first and second claimants as witnesses. I was perhaps more satisfied that what the first claimant had said in her witness statement was what had actually happened then I might have been if it was the second claimant alone, but taken together, and with the benefit of the further 2004 incident, I consider that the evidence that they gave of the discussions with their parents at and after the time of the execution of the wills on 25 July 2000 was substantially correct. In particular, I do not accept that they have subsequently made up the critical ideas (i) that their father was concerned to lay down once and for all the devolution of the family property, so that the wills should be “set in stone”, and (ii) that their mother agreed to this.

From these critical ideas proceeded the question which Mr Clark put to the solicitor and also the discussion between Mr and Mrs Clark and their daughters after the solicitor had gone. I accept that Mr Clark was satisfied with the assurance he had received from the solicitor (whether he understood it accurately is irrelevant), and did not wish to take the matter further, but considered that, because he and his wife had promised each other, there would be no changes to their wills thereafter. Mr and Mrs Clark were novices in will making, but they had very clear ideas about what they wanted to happen. The fact that more sophisticated will makers would or might have thought differently in the same circumstances is therefore nothing to the point. I also accept that Mrs Clark expressly made the exclamation (from the kitchen) that she would not change her will, as reported by both the first and the second claimant. The fact that they use the same words in their statements does not surprise me, and it is all the less likely that each is mistaken. It certainly does not make them liars.

Absent other evidence to the contrary, in my judgment this evidence would establish two agreements between Mr and Mrs Clark. The first is an agreement at some time before the execution of the will, and the second is one just afterwards. Each was to the effect that the wills they were to make, or had just made, were irrevocable. Their daughters were to benefit from the gift of the house.

It is clear that mutual wills need not be proved on direct evidence of an agreement between the testators, but may be proved on extrinsic evidence. In Re Cleaver deceased [1981] 1 WLR 939, for example, where the testators made mirror wills, there was no direct evidence of an agreement. But there were several pieces of extrinsic evidence.

Nourse J held (at 949C-F):

“But to the later events I do attach considerable importance. First there was the ‘That has settled’ or ‘That has cooked his hash’ incident. That suggests to me that the testator did think that he had tied everything up and that that did have the testatrix's tacit agreement. Then there are the events after the testator's death, in particular the two conversations which the testatrix had with Mr. Arthur Cleaver, the first on the evening of the testator's death and the second at the end of June 1977. I have already dealt with those conversations at some length. Having added them to all the other evidence in the case I find that I am in the end fully satisfied as to the existence of an enforceable agreement. I should, however, add that I do also attach some importance to the fact that within three months after the testator's death the testatrix did make a fresh will which faithfully followed her 1974 will in every material respect. There was no need for E her to make a fresh will. It could perhaps be partly explained as a tidying up operation and I can understand that she might not have wished to have a will which made any mention of her deceased husband. However, in the absence of any evidence as to the circumstances in which that will was made it must I think be of some significance that the testatrix apparently regarded herself as being under more than a moral duty to dispose of her estate in accordance with her 1974 will.”

Similarly, in Fry v Densham-Smith [2010] EWCA Civ 1410, the Court of Appeal upheld the finding of the judge at first instance that there was an oral agreement between two testators (Denny and Laura, each with a son from a previous marriage, Martin and Jonathan) for mutual wills, based on extrinsic evidence alone.

Mummery LJ (with whom Smith and Wilson LJJ agreed) said:

“32.

Direct evidence is not available on the two critical points in Martin’s claim: the existence of a mutual wills agreement between Denny and Laura and the execution of a will by Laura pursuant to it. No will, or copy will, or instructions for a will relied on as the mutual will made by Laura during Denny’s lifetime has been produced. There is no indication in Denny’s professionally drafted will or in the solicitor’s attendance note that it is a mutual will.

33.

The deficiency of direct evidence and the piecemeal nature of the evidence relied on by the judge for his conclusion do not, in my view, defeat Martin’s claim or undermine the judgment. It is, of course, unfortunate that Denny’s personal and financial papers were unavailable as direct evidence. However, evidence can be evaluated by a court and facts can be perceived and proved by the process of drawing reasonable and probable inferences from other facts, such as primary facts specifically found, undisputed events and uncontroversial circumstances surrounding them. All of those matters may be safe points of departure for the judicial process of drawing inferences that can lead to an evidentially satisfactory conclusion.”

Mummery LJ went on to say:

“36.

I think that the judge both made, and satisfactorily explained how he made, reasonable and proper inferences from the facts, events and circumstances summarised earlier in this judgment about what was probably agreed to be done and probably done by Denny and Laura about the disposition of their estates to each other and to their respective sons.

37.

In his detailed critique of the evidence (and the lack of it) Mr Norris paints a picture of a sparsely documented case based on unreliable oral evidence about events of almost 25 years ago involving two principal parties now both dead. Retrospective critical analysis is, of course, a necessary part of the process of evaluating the evidence and deciding what has been proved on the balance of probabilities. The critical process should not, however, deflect the court from the incontrovertible fact that, at the heart of the case, there was a testamentary predicament which was more likely to have been addressed than ignored by this elderly couple, each with a son from a previous marriage with prospects of inheritance of their parent’s assets. The probabilities are that they discussed the situation and reached an agreement along the lines alleged by Martin for the re-ordering of their affairs following the marriage. To me the judge’s findings on what was probably agreed and done have the ring of truth. They are based on reliable and satisfactory evidence given by Martin, which the judge was entitled to accept, even though he did not accept other aspects of his evidence; on the timing and contents of Denny’s last will; on Mundays’ note of what they were told by Denny about Laura making a will on their marriage; and on Laura’s unsent letter and her telephone conversations with Martin.”

In the present case, the defendants obviously could not give any direct evidence about what happened when the wills were executed. They were children at the time. They can only give slight, indirect evidence about the intentions of the testatrix and her husband by referring to what the testatrix said or did not say to them or in their hearing in later years. Essentially this is negative in substance, ie that the testatrix never mentioned any such agreement with her late husband as the claimants now contend for, and showed no reluctance to change her will in her later years (indeed, with some regularity). As Victoria Brooks said (and is obvious from the terms of the many wills the testatrix made) the testatrix was not consistent, changing her mind significantly from one will to the next. A similar point is made in her witness statement by the cousin (daughter of the first cousin) of the testatrix, Maryanne Dean.

The documents before the court show that in 2011 the testatrix made no fewer than four wills. The first and second of these, both made in January, gave pecuniary legacies to the grandchildren (amounting to about £100,000), but left the residue of the estate to the two claimants. The last of them, in December, gave only £5 each to the claimants, and left the rest to the grandchildren. I am satisfied from the evidence that neither of the claimants knew of the December 2011 will before the first defendant told the second claimant about it in 2012. The second claimant confronted the testatrix about the will, but the testatrix denied making it. There was then an incident which I find followed this confrontation, when the claimants searched the testatrix’s house while the testatrix was out at her cousin’s one Saturday (an event at which Maryanne Dean was present, and of which she gave evidence). Although this does not cast the claimants in a good light, it is nonetheless consistent with their case that in 2000 there was an agreement between Mr and Mrs Clark for mutual wills, as they allege. It does not demonstrate that the claimants must have known for years that the testatrix had been making wills but had done nothing about it.

It has been said that it is inherently improbable that a testator should be prepared to give up the possibility of changing his or her will in the future, whatever the change of circumstances: Charles v Fraser [2010] EWHC 2154 (Ch), [64]; see also Re Oldham [1925] Ch 75, 88. And it is well known that textbooks say that making mutual wills is not sensible, and that private client lawyers do not often advise them. After all, they take away some of the testator’s ability to adapt his or her will to changing circumstances. It is therefore suggested by the defendants that it is inherently improbable that any testator would want to make a mutual will.

But there are a number of problems with this approach. One is that the circumstances of individual testators making wills may be very different one from another. A testator who knows he is dying may have little interest in preserving his freedom of testation for the future, but every interest in tying down that of someone else, who may be prepared to acquiesce, whether by feelings of sentiment, or by the desire to acquire property from the dying testator. Or the testator may simply not have the benefit of sound advice as to the possible impact of the restraints on future will-making which a mutual will may have. It is wrong to treat every testator as if he or she were a private client lawyer or textbook writer, seeing the whole range of possible cases. Each testator, however, sees only his or her own case.

Another is that the surviving testator might never have obtained control of the property of the first testator to die without making the agreement (or giving the promise). So, to the extent that the survivor’s estate is thereby augmented, the survivor may be losing no freedom of testation at all. And gaining that augmentation may be a considerable advantage if, as one would normally imagine to be the case, it is impliedly agreed that the survivor may “dip into” the capital as and when needed for the survivor’s needs. So there may be more to be said for entering into a mutual wills agreement than the textbooks give credit for. To elderly people, extra security is particularly valuable. We should not judge them by the risk-taking attitudes of younger people.

A further, important, point is that the reality in England in 2017 is that testators are in practice far less free than is popularly supposed to make their wills as they see fit. Contrary to popular mythology, complete freedom of testation in England only existed between 1891, or (if we ignore limits on gifts to charity) 1833 (for men) or 1883 (for women), and 1938. And even between those times the high tide of Victorian family morality (which lasted well into the twentieth century) ensured that testators in practice did “the right thing”. Today there are not just family and societal pressures on what you may do with your estate, but also legal ones, in the form of the Inheritance (Provision for Family and Dependants) Act 1975, which all sensible testators take into account in making their wills. The number of cases where emancipated adult children make a successful claim under that Act demonstrates a sea-change in judicial attitudes even since 1937 (when apparently all the chancery judges save one were against the proposed new legislation in principle: see Tyler’s Family Provision, 3^rd ed by Oughton, p 18). So testators who make mutual wills today are giving up rather less than complete freedom of testation.

A fourth point is that promises not to change or revoke wills do not have to be absolute. They may have limits. For example, the promise may be not to change or revoke without the agreement of the other party, or without notice to the other party, or may be limited to certain assets, or a fraction of the estate. The parties are free to agree what they like. Even after the death of the first to die, it will be possible for the survivor to agree with the persons who would benefit under the mutual will agreement nevertheless to change her will. For all these reasons I decline to hold that intending to make mutual wills in England today is inherently improbable. It all depends on the particular circumstances.

Of course the words or actions of the testatrix in subsequent years may shed some light on what she promised or did not promise in 2000. But the mere fact that the testatrix subsequently acted inconsistently with what she is alleged to have promised would not of itself prevent my finding, on the totality of the evidence before me, that she indeed promised her husband not to change or revoke her will, or that he relied upon her promise.

In my judgment, on the basis of all the evidence and other material before me, Mr and Mrs Clark expressly promised each other that having made their wills in the form they had they would not revoke them, and thereby engaged the principle of mutual wills. That being so, the testatrix was not free unilaterally to alter her will and make a new one inconsistent with that of July 2000, in the sense that, if she did so, her personal representatives would hold her estate on trust to perform the equitable obligations laid upon her under the doctrine of mutual wills.

There is some controversy as to when and how the constructive trust that may be produced by the operation of the doctrine of mutual wills takes effect. For example, what happens on the death of the first testator to die? Is there some kind of “floating trust” over (i) all the assets passing to the second testator and/or (ii) the assets already in the second testator’s estate (see Birmingham v Renfrew, 675; Re Goodchild [1996] 1 WLR 694, 702)? Does this kind of constructive trust satisfy the so-called “three certainties” of other trusts? It is not necessary to resolve all these questions in this case. But I will shortly state my own view that the so-called “three certainties” rule is not a rule about trust law at all. Instead it is a rule about property law, and, trusts being part of property law, they follow that rule too. If A is to make a gift to B of some asset valid at common law, it is obvious that A must intend a gift and not a loan (or some other legal construct), that it must be clear exactly what it is that is being given, and that it must be clear to whom it is being given. A failure in any respect causes the gift to fail. So too with trusts.

Here, in my judgment, that rule (so far as it applies) is complied with. There is no doubt that both Mr and Mrs Clark intended that their wills should not be changed. They did not specify that there was to be a trust to implement their intention, any more than the parties in Paul v Constance [1977] 1 WLR 527, CA, did so. And indeed this would be a constructive trust, albeit one arising from an agreement, so the rule cannot apply in the same way as an express trust. But their intention is clear nonetheless. It is equally clear that they intended to benefit their daughters, in equal shares, and no one else. Lastly, it is clear that the idea of not changing or revoking your will carries with it the notion that everything that you leave at your death shall pass to the ultimate beneficiaries. So the subject matter of this trust is everything which is left at the death of the survivor. And that is the point at which usually the constructive trust is imposed, to the extent that the provision then made by the will of the survivor, or the intestacy rules so far as they are applicable, or a combination of both, is or are inconsistent with the original agreement.

There is no trust imposed on the death of the first to die, unless that is so agreed. How far the survivor may deal with the assets which come from the first testator and with the assets which remain in the second testator’s estate is a matter of construction of the agreement (or other binding promise) between them. In the ordinary case, one may imagine that their agreement would allow for the survivor to spend capital (or some agreed amount or fraction of the capital) on herself if that should prove necessary, but, on the other hand, not to give it away to third parties during her life. It may be that, if the survivor attempts to give away (or otherwise deal with) such property in breach of the agreement, that too is an event which would justify the imposition of a constructive trust on the purported gift (see eg Lewis v Cotton [2001] 2 NZLR 21, [43]). But it is not necessary to deal with that now. That did not happen here. The consequence is that there is no uncertainty of subject matter for the trust imposed by virtue of the mutual wills doctrine.

Once I reach that conclusion, it does not matter if the claimants, or either of them, after the death of their father behaved towards their mother, or indeed any other members of their family, in a manner which may be criticised. I emphasise that I make no finding that either of the claimants has so acted. But, short of agreeing a release, or making a representation to found an estoppel, in law nothing that they could do thereafter could undo the equitable obligation binding their mother in relation to both the property which she received from her husband and also her own property which she had at the time of making the will. In substance, this refers to the house which they had bought, which, by their wills of July 2000, Mr and Mrs Clark at that time intended to pass on to their daughters (the claimants). I have no doubt that when Mrs Clark, the testatrix, made the many further wills after 2000 which she did, she had either forgotten that she had made a promise not to change her will, or, if she remembered that (and on at least one occasion she was reminded of it and changed her will back accordingly) she was nevertheless not entitled to disregard it. In equity at least, the clock of her testamentary freedom had stopped with the death of her husband, and thereafter she no longer retained the unilateral right to dispose of her assets that she had once enjoyed.

It follows that there must be judgment for the claimants. This does not mean that they have proved any kind of moral superiority over the defendants. I am not concerned with any such questions. What it simply means is that they have established that Mr and Mrs Clark in the year 2000 made their wills subject to an agreement that they would not thereafter revoke them or change them without notice to the other and, Mr Clark having died without having changed his will and without having received notice from Mrs Clark that she had changed hers, Mrs Clark’s estate must be held by her personal representatives on trust to give effect to the form of the will which she made in July 2000, and not as set out in the will of December 2014.