Jaswinder Kaur Sangha v The Estate of Diljit Kaur Sangha & Ors

This second appeal arises after judgment in a probate action that was tried before Deputy Master Bowles, an experienced former Chancery Master who was sitting in retirement (“the Deputy Master”), and raises two main questions. The late Hartar Singh Sangha died on 3 September 2016 in Chandigarh, Punjab, India aged 72 leaving substantial assets both in England and in India. He made a will in 2007 dealing with both his English and Indian assets. He made another will in 2016 which disposed only of his Indian assets. This was declared to be his last will and contained a revocation clause revoking “all such previous documents”. The first question is whether this was effective to revoke the 2007 will in its entirety, as the Deputy Master held after trial of the action, or only effective to revoke the 2007 will to the extent that it dealt with his Indian assets, as Mr Simon Gleeson, sitting as a Deputy High Court Judge, (“the Deputy Judge”) held on appeal.

The second question only arises if the 2007 will was not wholly revoked, and concerns the requirements for due execution in s. 9 of the Wills Act 1837 as amended. The question is whether a will is duly executed if the order of events is (i) testator signs in the presence of witness 1; (ii) witness 1 attests the will; (iii) witness 2 then comes into the room and testator acknowledges his signature in the presence of both witnesses; and (iv) witness 2 attests the will. Is this sufficient (as both the Deputy Master and the Deputy Judge held)? Or must in such a case witness 1 acknowledge his signature after the testator has acknowledged his signature in the presence of both witnesses? There is also a question, if the latter is the case, whether the Deputy Master was entitled to find as a fact in the present case that witness 1 did acknowledge his signature after the testator had done.

There is also a subsidiary issue as to the form of order the Deputy Judge made.

There was a complex background to the dispute, and the Deputy Master heard a considerable amount of evidence over the course of a 9-day trial not only in relation to the probate claim but also in relation to a proprietary estoppel claim that was tried at the same time. He set out his findings of fact in a lengthy, careful and detailed judgment handed down on 16 June 2021 at [2021] EWHC 1599 (Ch). This deals with a large number of factual and legal issues. What follows is only a brief summary, but sufficient to explain the questions that arise. I will refer, as both judges below and counsel have done, to Hartar Singh Sangha as “Hartar” and to the other members of his family by their first names. Numbers in square brackets in this section of my judgment refer to paragraphs of the Deputy Master’s judgment.

Hartar was born in Punjab in 1944. When he died in 2016, he left two families, consisting respectively of his first wife Diljit, their son Sundeep and their daughter Mandeep or Mandi, and of his second wife Jaswinder and their son Harbiksun. I have referred to Diljit and Jaswinder as his “wives”, as he himself did in different wills, but as will appear there is a dispute, not yet resolved in these proceedings, as to which of them he was lawfully married to. He was also survived by his sister Jagpal.

There were four wills made by Hartar, all in English, which were in evidence. I give the details below but in summary they were:

Hartar is said to have married Diljit in Punjab in a customary Sikh ceremony in October 1962, but Jaswinder has disputed whether Diljit was ever lawfully married to him [4]. In 1963 he moved to England and in 1965 Diljit joined him [5]. Sundeep was born in 1969 and Mandi in 1972, both being born in the UK [5]. In these proceedings Diljit, who survived Hartar by some 18 months, was initially the 1^st Defendant. She died in March 2018 however, aged 75, and her estate is now represented by Sundeep [3]. Sundeep himself was the 2^nd Defendant and Mandi the 3^rd Defendant. Diljit’s estate, Sundeep and Mandi are the Appellants in this Court.

In 1979 Hartar made the 1979 will. Its validity was accepted by all parties [64], although it is not disputed that it has been revoked. It is typed, professionally drawn by a firm of English solicitors in Essex, and in conventional English form. It is witnessed by a secretary and a receptionist with English addresses and evidently made in England. It is dated 20 July 1979. It appointed the partners in the firm as executors and trustees and left “all my estate” subject to payment of expenses and debts to his trustees on trust as to half for “my wife Diljit” for life, with remainder to his children Sundeep and Mandi, and as to the other half for his children, thereby closely following the provisions under English law on intestacy. There is no suggestion in the will of separate provision being made for any assets in India, there being no reference to India in the will at all, nor indeed is it known if Hartar did then own any assets in India.

Hartar in due course became a British citizen. By 1990 he had acquired a substantial English estate in the shape not only of a successful bathroom fittings business but also of various properties [10]-[11]. He had in 1983 bought a plot in Hainault Road, Chigwell and had been building a house there over the years as a family home [5]. It was finished in 1990 and the family moved there [5]. Diljit in fact continued to live in Hainault Road until her death [7], although she never had any legal or beneficial interest in the property.

Hartar’s sister Jagpal had some time before 1990 joined Hartar in England, and she moved with the family to Hainault Road [6]. But it appears that Diljit and Jagpal did not get on, and in 1991 Diljit asked Jagpal to leave [8]. That seems to have been the catalyst for the breakdown of the relationship between Hartar and Diljit, and when Jagpal left Hartar did too [9]. They both returned to India and went to live in Chandigarh, where in due course Hartar bought a house [9]. Thereafter Hartar seems to have divided his time between India and England [16].

In August 1992 Hartar married (or purportedly married) Jaswinder in Jalandhur in India [2]. The validity of this marriage is disputed, Diljit’s estate claiming that he lawfully married Diljit in 1962 and was never divorced [4]. That was an issue raised on the pleadings in these proceedings, but it was agreed that it would not be addressed in the trial before the Deputy Master and it remains an unresolved issue [2]. In these proceedings Jaswinder was the Claimant, and the only active Respondent in this Court.

Hartar did not tell Diljit of his marriage to Jaswinder, and only told Sundeep in 1994 when the two of them travelled together to Chandigarh [18]-[19]. By then Hartar had developing business interests in India and he told Sundeep that he, Hartar, would look after those but that Sundeep should look after the family’s affairs in the UK and would be his successor and would get everything in the UK [19].

In 1998 Hartar and Jaswinder set up home together in England, although again he did not tell Diljit or Sundeep, and Sundeep only learned of it from his father in 2002 [23].

In 2001 Hartar’s and Jaswinder’s son Harbiksun was born [2]. Harbiksun was the 4^th Defendant, but took no part in the proceedings either below or on appeal. The 5^th and final Defendant was Jagpal, who was represented at trial but has again taken no part in the appeal, which does not affect her.

Diljit became unwell in 2002. It fell to her son Sundeep to be primarily responsible for looking after her, and increasingly so after 2011 [42]-[45]. The Deputy Master found as a fact however that Hartar was concerned not to be seen in the Punjabi/Sikh community to have abandoned Diljit, and not only did Diljit continue to live in Hainault Road, but he made available to her and Sundeep an income derived from another of his properties [21].

Hartar also promised Sundeep that he would transfer his English properties into his name [23]. These, and later similar promises, formed the basis for the proprietary estoppel claim which was advanced by Sundeep. The Deputy Master found that such promises were repeatedly made (in 2002, 2009, 2011/2, 2015 and finally July 2016) but that the proprietary estoppel claim failed because of a lack of detrimental reliance [26]-[27], [362]-[391]. In fact however, as Sundeep discovered only in 2016, Hartar over the years transferred a large number of his English properties (including Hainault Road where Diljit was still living) into the joint names of himself and Jaswinder [27], [51].

In 2002 Jagpal married. The Deputy Master found that that appears to have precipitated a very serious estrangement between Hartar and Jagpal (who had previously worked together with him in developing his Indian business and property affairs, and who also had a 25% interest in the Chandigarh property) [29]. Hartar apparently took the view that now that she was married Jagpal should “return” her interest in the Chandigarh property to him [29].

In 2003 Hartar made the 2003 will. By the end of the trial the validity of this will was not in issue [65]. It is typed but the Deputy Master found that the words used were very substantially Hartar’s own words [292]. It is dated 5 March 2003 and was made in Chandigarh. Hartar gives his Chandigarh address and then recites that he is the owner of “the following properties” which are listed from (a) to (i). Those at (a) to (e) are each a specific parcel of land in India, that at (h) as set out below is also confined to Indian property, and that at (i) is a general sweep-up clause of “all moveable and immoveable properties in India at the time of my death”. Those at (f) and (g) do not specifically refer to India, being respectively “Shares of Limited Company and profits from it” and “All bank accounts and deposits with any banks and pending claims and cases of properties”, but in the circumstances the Deputy Master held that these items were intended to refer to Hartar’s Indian company and Indian bank accounts and that the will made no disposition of his English estate [288]-[289].

All these properties were left to “my wife Jaswinder” and Harbiksun, or failing them to Sundeep. Paragraph (h) contained a similar specific provision in relation to what were termed “deprived rights” as follows:

“Deprived rights such as house Plot 508, Sector 18, Chandigarh and share of 768, Sector 8, Chandigarh and house/plot at Kapurthala, which I have purchased with earned funds from UK but in joint names of my sister which she should transfer back to myself after she got married. Her share should be given to my wife and my son Harbiksum [sic] Singh. In case of my death Jaswinder Kaur, my wife and Harbiksun Singh Sangha my son died within 28 days then all should go to my older son Sundeep Singh Sangha who is settled in UK.”

After the substantive dispositions the will contained two paragraphs as follows:

“And whereas I cancel all my Wills and testaments made earlier.

This Will is my last Will and testament.”

It was argued for Diljit, Sundeep and Mandi before the Deputy Master that this was ineffective to revoke the 1979 will in relation to English assets, because giving the revocation clause its literal effect would be to create an intestacy as to Hartar’s English estate that he cannot have intended [291]. But the Deputy Master rejected that on the basis that there was nothing in the will to support this view and the bald point that the literal construction would give rise to an intestacy in respect of the English estate and a want of provision for the first family was not enough to establish that this was Hartar’s intention [293].

A note has been added to the 2003 will in manuscript as follows:

“make sure that no property goes to Jagpal Kaur all my property to transfer to my wife Jaswinder and son Harbiksun”.

Jaswinder’s evidence, not significantly in dispute, was that this note was written by Hartar at some point following execution of the will and reflected his ongoing dispute with Jagpal [73].

The Deputy Master found that Hartar told Jaswinder that so far as his English estate was concerned he had provided for her by transferring properties into their joint names [212]; and in 2004 Hartar bought more property in England, again in the joint names of himself and Jaswinder [51].

In 2007 Hartar executed the 2007 will. The validity of this will was very much in issue at trial. Diljit’s estate, Sundeep and Mandi contended that it was a fabrication, having been concocted by or on behalf of Jaswinder, and that the entire story of its creation and execution was untrue [74]. In the event the Deputy Master rejected this case for reasons set out by him in detail at [217]-[279]. That has not been appealed and the will is therefore now established to have been genuine.

But there was also, and remains, a question as to its due execution. I will have to look at the details of this question and the evidence as to its execution in more detail below but in summary the Deputy Master found that the sequence of events was as follows. First, Hartar signed the will in the presence of the first witness, a Mr Balraj Singh; second, Mr Balraj Singh signed the will as attesting witness in Mr Hartar’s presence; third, the second witness, a Mr Khaira, was summoned into the room where he joined Hartar and Mr Balraj Singh; and fourth, Hartar asked Mr Khaira to sign the will (thereby acknowledging his own signature in the presence of both witnesses), which Mr Khaira then did [75]. That in the Deputy Master’s view satisfied the requirements of s. 9 of the Wills Act 1837 as amended, and it was not necessary to show in addition that Mr Balraj Singh had acknowledged his own signature after Hartar had acknowledged his [121]. But if the latter were necessary, he would have been able to conclude that the probability is that Mr Balraj Singh did acknowledge his signature to Hartar after Hartar had himself acknowledged his signature to Mr Khaira and Mr Balraj Singh [122]. In those circumstances the will was duly executed [128].

The 2007 will was made on a printed will form but completed in manuscript, and executed in Chandigarh in India. It would appear that Hartar had had the form for some time as it had been signed by Hartar in blank with a date of 25 August 2000 in blue biro on the second page, but apart from this the will was written in black biro, and the 2000 date was struck through and replaced with the date 16 April 2007 (again in black biro). It appointed “my wife Jaswinder” and Harbiksun as executors, and left “all my estate” to Jaswinder, or failing her to Harbiksun, with a further provision that:

“above estate means all my Bank accounts, removeable and immoveable property in India and U.K (England) should go to my wife Jaswinder Kaur and also all my pending court claims and other disputes Jaswinder Kaur should benefit”.

Like the note added to the 2003 will, it also contained remarks directed at Jagpal, including a provision that:

“I hereby confirm that my sister Jagpal Kaur should not make any claims on my property and company and should stay away from my family.”

Despite the making of the 2007 will Hartar’s promises to Diljit and Sundeep continued. But so did his transfers of properties into the joint names of himself and Jaswinder, including in 2010 Hainault Road (where Diljit was still living) [51]. By the time Hartar died in 2016 over £10m worth of English property in total was in their joint names and shown in the IHT account as passing to Jaswinder by survivorship [48]. This constituted the large majority of Hartar’s English estate, his net estate for probate purposes otherwise being valued at some £1.1m [48]. (That may be contrasted with his Indian estate on which Jaswinder put a value of £30m at the date of his death [47].) The Deputy Master found that Hartar thought he was thereby providing for Jaswinder and that when making his 2016 will assumed that these properties would pass to Jaswinder by survivorship [211]-[215]. In fact Hartar was made bankrupt in 2013, the bankruptcy being annulled shortly thereafter. The bankruptcy had the effect, as the Deputy Master held, of severing the beneficial joint tenancy of those properties which had previously been transferred (which was most of them) [332]; and he further held that the effect of the annulment was to revest in Hartar the interests in common arising from the severance, not to reconstitute the joint beneficial ownerships [341]. He found however that there was no reason to believe that Hartar was aware when he made his 2016 will of the effect of his bankruptcy on the provision he thought he had made for Jaswinder [214].

In 2016 Hartar made his final will, the 2016 will. Again its validity was hotly contested at trial. In this case it was Jaswinder who contended that the will, which was propounded by Jagpal, was not authentic but fabricated by Jagpal, or on her behalf, the entire story of its preparation and execution being invented [84]. Diljit’s estate, Sundeep and Mandi, having initially contested the genuineness of the will in their pleading, ultimately took a neutral stance at trial [85]. But the Deputy Master found that this will too was genuine [144]-[191], and again that has not been appealed, with the result that it is now established that this will is valid and effective, and is the operative will for Hartar’s estate. There remains however a question, as already referred to, whether it had the effect of revoking the 2007 will in its entirety or only insofar as the 2007 will disposed of Hartar’s Indian estate.

The will is typed and dated 21 March 2016 (and so about 6 months before Hartar’s death). The evidence as to its preparation and execution came primarily from an Indian advocate, Mr Bedi, and was to the effect that the will had been executed at his office, in the old courthouse in Kapurthala [82]. Mr Bedi had first written out the will in Punjabi at Hartar’s dictation; he had then read it out to Hartar; and had then had it typed out in English. It had then been duly executed [150].

The will does not appoint executors. It refers to Hartar’s ownership of various specific properties in India, and bank accounts and other moveable and immoveable properties in India, and provides that this should all be divided after his death in equal shares between “my following four legal heirs”, namely Sundeep, “my wife Daljit [sic]”, Harbiksun and Jagpal, adding by way of explanation “Although we did split due to some reasons, but I do not want to ignore my sister’s support to me on many aspects.” It makes no provision for, or reference to, Jaswinder at all. Jagpal also produced an undated letter, said to have been sent to her in India by Hartar, in support of his supposed reconciliation with her. This too was attacked by Jaswinder as a self-serving forgery, but the Deputy Master found it to have been genuine, and persuasive evidence that Hartar was indeed seeking to reconcile with his sister [171]-[179].

After the dispositive provisions the will provides:

“This is my last and final WILL and all such previous documents stand cancelled.”

As with the 2003 will (see paragraph 20 above), the question was raised whether this was effective to revoke the previous will (here the 2007 will) in its entirety or only in respect of its disposition of Hartar’s Indian estate, leaving the previous disposition of his English estate untouched. Here it was Jaswinder arguing for a partial revocation, with Diljit’s estate, Sundeep and Mandi contending that the 2007 will had been revoked in its entirety.

The Deputy Master found that the plain meaning of the words of revocation could not be clearer: all other wills are cancelled [194], [203]. There was no extrinsic evidence of Hartar’s actual, subjective intention in relation to the revocation or cancellation clause [197], [200]. In those circumstances the Deputy Master held that there was nothing in the will, or the material surrounding circumstances, to sufficiently contradict the literal meaning of the words of cancellation used in the will, or to import, or denote, any unequivocal intent in the testator to modify, or limit, that meaning [205]. He was not persuaded that the fact that the literal effect of the revocation clause in the 2016 will gave rise to an intestacy in regard to Hartar’s English estate was sufficient to establish the clear and unequivocal intention that the revocation clause must have been intended to be confined to the revocation of those wills, or parts of wills, affecting only Hartar’s Indian estate [209]. The fact that the literal interpretation of the revocation clause would lead to an intestacy in respect of Hartar’s English estate had to be set in the context of the steps that he had taken, or believed he had taken, to transfer the bulk of his English assets into the joint names of himself and Jaswinder inter vivos, such that his expectation in 2016, when he made his 2016 will, would have been that those assets would not fall into his estate at all and would simply vest in Jaswinder by survivorship [211]. In that context a decision to revoke all previous wills, including those dealing with his English estate, was not at all implausible [212].

Hartar, as already mentioned, died on 3 September 2016 in Chandigarh. One of the other issues raised in these proceedings was where he was domiciled at the date of his death, Jaswinder contending for an Indian domicile and Diljit’s estate, Sundeep and Mandi for an English domicile [72]. This issue, like the issue of the validity of Hartar’s marriages to Diljit and Jaswinder, was however excluded from the trial. The result of the Deputy Master’s judgment was that the 2016 will was valid and revoked the 2007 will in its entirety (and also the 1979 and 2003 wills) and therefore resulted in an intestacy in relation to his English estate [205]. That therefore meant that both these unresolved issues would require to be resolved, the question of domicile because it would determine which set of intestacy provisions, Indian or English, applied, and the question of the validity of Hartar’s marriages as it would determine which of Diljit and Jaswinder was his lawful widow [204].

The Deputy Master gave effect to his judgment by an Order dated 21 September 2021. So far as concerns the probate claim this recited that the Court had found that the 2016 will was Hartar’s last valid will, that it disposed of his Indian estate only, and that it revoked all previous wills such that Hartar died intestate save in respect of his Indian estate. It then provided that the Court pronounced for the force and validity of the 2016 will, and made declarations in line with the recited findings. It then gave directions for the determination of the outstanding issues, namely as to Hartar’s domicile and whether he lawfully married Diljit in 1962 or Jaswinder in 1992, and which of them was his surviving spouse for the purposes of intestacy.

It did not grant probate of the 2016 will. The Deputy Master had noted in his judgment that there was a question whether it would be appropriate for the English Court formally to admit to proof the 2016 will which was made in India and limited to Hartar’s Indian estate. But it had not been the subject of any argument so he left it over for further discussion after hand-down of the judgment [193]. We have not seen any further judgment but the Deputy Judge recorded in his supplemental judgment (see below) that the Deputy Master was specifically asked to grant probate of the 2016 will but declined to do so “on the relatively uncontroversial ground that it is not the practice of the English court to grant probate of a Will which deals exclusively with assets in another jurisdiction.”

Jaswinder appealed the declaration that the 2016 will revoked the 2007 will in its entirety, contending that it only revoked the 2007 will in relation to Hartar’s Indian estate. Diljit’s estate, Sundeep and Mandi responded by contending that if the 2007 will was not wholly revoked it was invalid for want of execution.

The appeal came before the Deputy Judge. He handed down judgment on 12 August 2022 at [2022] EWHC 2157 (Ch). Numbers in square brackets in this section of my judgment refer to paragraphs of his judgment.

He first considered the presumption against intestacy which had been relied on by Jaswinder in support of her appeal, but he concluded that it was of no assistance to her in the present case [20]. He then addressed the question of construction of the revocation provision in the 2016 will, and specifically the word “such” in “all such previous documents stand cancelled” [26]. He said that the word could refer to the word “will” so that the reading could be expanded as “this is my last and final will and all [previous wills] stand cancelled”. Or it could refer to the document in which it was contained so that it could be expanded as “this is my last and final will [in respect of my Indian property], and all [wills in relation to that property] stand cancelled” [26].

He said that he did not find it easy to say which was the correct interpretation. But he found assistance in two material facts. The first was that it seemed to him that if Hartar had intended to revoke the will dealing with his English property he would have mentioned the fact or at least alluded to it, and some significance must be attached to his decision not to do so [28]. The other was that there was a plausible alternative explanation for the wording, namely that it was intended to refer to the 2003 will which also related exclusively to Hartar’s Indian property. It was more likely that that was the purpose of the clause than that it was an indirect attempt to revoke by implication the provisions of a different will on a different continent dealing (in good part) with different property [29]. On the basis of these matters he concluded that the Deputy Master was wrong to construe the words as wholly revoking the 2007 will [30].

He then referred to certain authorities which I will have to look at in more detail below, concluding as follows (at [34]):

“It seems to me that where a will is expressed to apply to specific, identified property in a particular jurisdiction, is made in that jurisdiction with the assistance of lawyers established and qualified in that jurisdiction, and has no other connecting factor with any other jurisdiction, the starting point should be an assumption that the will as a whole is only intended to apply to that property in that jurisdiction unless there is some good reason to believe otherwise. In this case there was none.”

He then dealt with, and specifically rejected, a number of arguments put forward by Mr William East, who appeared before him on behalf of Diljit’s estate, Sundeep and Mandi. One was that Hartar could have, but did not, specifically restrict the revocation clause to Indian wills; the Deputy Judge said that that seemed to him to reverse the natural approach [36]. He also rejected a submission that Hartar cannot have intended to allude to the 2003 will as he would have known that it had been revoked by the 2007 will, saying that it was unlikely that Hartar could have had that degree of certainty [36]. He rejected a submission that his construction departed from the natural and ordinary meaning of the words used as he did not agree that the Deputy Master’s construction was the natural and ordinary one [37]. Finally he rejected a submission that Hartar’s intention to revoke the 2007 will could be inferred from the steps which he had taken in respect of his English assets, referring to the Deputy Master’s conclusion that it was “not at all implausible” that Hartar might have taken the view that his English estate was already dealt with. The problem with that was that not all of Hartar’s English estate had been put into joint names and there might have been in excess of £1m of assets that did not pass by survivorship. He agreed that the construction put forward was not implausible but he did not find it compelling [38].

He therefore held that the 2007 will was not wholly revoked. He went on to consider whether it had been validly executed. He agreed with the Deputy Master’s conclusion on what was required to comply with s. 9 of the Wills Act 1837 [46]; and he also held that the Deputy Master was entitled on the facts to infer that Mr Balraj Singh had acknowledged his signature after Hartar had acknowledged his own signature, and that there was no room on appeal to challenge a finding of fact by the tribunal which had actually heard the witnesses [51].

He therefore allowed the appeal and dismissed the cross-appeal [52].

After judgment was handed down there was argument on a number of consequential issues. One of these was whether a grant of probate should be made in respect of both the 2016 will and 2007 wills, or only in respect of the 2007 will.

The Deputy Judge handed down a supplemental judgment on 13 September 2022 dealing among other things with this issue. It was argued by Mr East that the appropriate order was to grant probate of both the 2007 and 2016 wills, but the Deputy Judge concluded that only the 2007 will should be admitted to probate [38]-[39].

By his Order dated 9 September 2022 he therefore declared that the 2007 will was validly made and that the 2016 will did not revoke the 2007 will in relation to Hartar’s English estate, and granted probate of the 2007 will to Jaswinder and Harbiksun limited to Hartar’s English estate.

Diljit’s estate, Sundeep and Mandi now appeal to this Court. Four grounds of appeal are advanced. They are, in summary:

There was little dispute as to the relevant legal principles. First, in deciding which will or wills are valid and effective and should be admitted to probate, the Court is sitting as a court of probate. As such it is always open to the Court to consider the intention of the testator and for that purpose to consider evidence outside the confines of the will in question. This is well established law: see the statement of Sir John Nicholl in Methuen v Methuen (1817) 2 Phil 416 at 426, cited by Lord Wilberforce in re Resch’s Will Trusts [1969] 1 AC 514 at 547D, that:

“In the court of probate the whole question is one of intention: the animus testandi and the animus revocandi are completely open to investigation.”

The authorities are usefully collected and considered in the judgment of Mr Roger Wyand QC, sitting as a Deputy High Court Judge, in Lamothe v Lamothe [2006] EWHC 1387 (Ch) at [17]-[31]. He concluded at [32] that on the facts of the case before him:

“For those reasons I find that it is open to me, and indeed my duty in a case such as this, to consider all the evidence as to the surrounding circumstances of the drafting and execution of the 1995 will to determine whether Mrs Lamothe intended thereby to revoke the 1993 will.”

That seems to me to be an accurate application of the law.

Second, it follows that even where a will contains a general revocation clause (a clause revoking all previous wills) there can in any particular case be evidence which establishes that this was not the testator’s intention. An example can be found in one of the cases referred to by Mr Wyand in Lamothe v Lamothe, namely re the Goods of Oswald (1874) LR 3 P & D 162. There the testatrix’s last will contained a revocation clause but the evidence clearly established that it was not her intention to revoke her previous will, and both wills were admitted to probate with the revocation clause omitted from the second will.

Third, however, it is also well established that if a will, duly executed, does contain a general revocation clause, that is in itself powerful evidence that that was the testator’s intention and that a “heavy burden” lies on those seeking to establish otherwise: see Lowthorpe-Lutwidge v Lowthorpe-Lutwidge [1935] P 151 at 156 per Langton J. Having referred to Gregson v Taylor [1917] P 256 at 261 per Hill J (“when it has been proved that a will has been read over to or by a capable testator and he then executes it, these circumstances afford a very grave and strong presumption that he knew and approved all the contents”), Langton J said that although the Court has power to admit an earlier will to probate despite a later will with a revocation clause if satisfied that it was not the intention of the testator to so revoke, continued:

“But the burden, in that case, is heavy. It is a heavy burden upon a plaintiff who comes into this Court to say: “I agree that the testator was in every way fit to make a will, I agree that the will which he has made is perfectly clear and unambiguous in its terms, I agree that it contains a revocatory clause in simple words: nevertheless I say that he did not really intend to revoke the earlier bequests in earlier wills.” Quite obviously the burden must be heavy upon anybody who comes to assert a proposition of that kind.”

Mr Alexander Learmonth KC, who appeared with Mr East for Diljit’s estate, Sundeep and Mandi, submitted that in this case the 2016 will did contain a clear general revocation clause. That was certainly how the Deputy Master saw the matter (see paragraph 31 above). The Deputy Judge saw it differently as he regarded the revocation clause as open to two interpretations depending on whether “all such previous documents” meant “all previous wills” or “all wills in relation to Indian property” (see paragraph 37 above). Ms Penelope Reed KC, who appeared with Mr Mark Blackett-Ord for Jaswinder, sought in her written submissions to uphold this as the correct interpretation in context, but accepted that the natural and ordinary meaning of the words taken on their own was as a general revocation clause. That indeed had been accepted by Mr Blackett-Ord below. In oral argument she concentrated on the proposition that even a general revocation clause can be given a narrower effect in appropriate circumstances: see Dempsey v Lawson (1877) 2 PD 98 at 107 per Sir James Hannen P:

“Even if the second instrument contains a general revocatory clause, that is not conclusive, and the Court will, notwithstanding, consider whether it was the intention of the testator to revoke a bequest contained in a previous will.”

I agree that the clause here is a general revocation clause. The natural meaning of the provision that “all such previous documents stand cancelled” is that “such” refers back to “my last and final WILL”, and hence that the clause cancels all previous wills. I also agree however that that is not conclusive. In accordance with the authorities I have referred to, there is always a question in a court of probate whether the testator did intend the clause to have its natural effect. But there must be some justification for the Court concluding that the testator did not intend it to do so.

Ms Reed relied on a number of matters. The first was that this was a will made in India and dealing with Indian property. She referred by way of analogy to re Wayland [1951] 1 All ER 1041. There a testator made a will in 1947 in Belgium expressed to deal only with Belgian property, and a will in 1949 in England. The 1949 will contained a general revocation clause but went on to say “this will is intended to deal only with my estate in England”. Pearce J had little difficulty in holding in the circumstances that the revocation clause was to be read as only revoking former wills dealing with English property and did not revoke the 1947 will dealing with Belgian property. That was on the basis that if it was read as revoking the Belgian will the 1949 will would deal with estate other than the testator’s estate in England which would be inconsistent with the provision in the will (at 1043E-F). It may be noted that there was also clear evidence that the testator was unaware that the revocation clause might operate to revoke the Belgian will and did not intend it to do so, including evidence that he told his English solicitor not to worry about his Belgian estate as his Belgian lawyer would deal with it as he had made a Belgian will (at 1042G).

A somewhat similar case is Benjamin v Bennett [2007] All ER (D) 243. There a testator made a will in England. He then made a will in Barbados. This contained a general revocation clause, but was headed “Barbados” and only dealt with his assets in Barbados. Mr Richard Sheldon QC, sitting as a Deputy High Court Judge, held that the revocation clause did not revoke his English will. The case is only reported in a very brief digested form which does little more than refer to the claimant (the executrix and residuary beneficiary under the English will) having satisfied the onus of proof and having established that the deceased had not intended to revoke the English will by the Barbados will. The researches of counsel however have resulted in a note from counsel who had appeared for the claimant in the case which gives rather more information. From this it appears that there was evidence from a partner in the Barbados firm which drafted the Barbados will to the effect that the deceased said nothing about intending to revoke an English will, his instructions being that the Barbados will was to deal with Barbados property; that both the English and Barbados wills were found together in the deceased’s safe; and that it had been conceded by the first defendant (the executrix of the Barbados will) that she had understood that the deceased intended no conflict between the two wills.

In both these cases therefore the facts went a good deal further than the mere fact that the later will containing the revocation clause only disposed of assets in the country where it was made. In each case the later will contained an indication on its face that it was intended to be confined in its operation to that country either by saying expressly that it was “intended to deal only with my estate in England” or by being headed “Barbados”; and in each case there was powerful supporting evidence as to the testator’s actual intention.

The Deputy Judge took as his “starting point” an assumption that a will expressed to apply to specific, identified property in a particular jurisdiction, made in that jurisdiction with the assistance of lawyers established and qualified in that jurisdiction, and having no connecting factor with any other jurisdiction, is only intended to apply to that property in that jurisdiction unless there is some good reason to believe otherwise: see his judgment at [34] (cited at paragraph 39 above). That seems to me to go further than is justified by any of the authorities we were shown. As set out above both re Wayland and Benjamin v Bennett were cases where the evidence was much stronger than the mere fact of a will made in England disposing only of assets in England or a will made in Barbados disposing only of assets in Barbados.

The closest one comes to support for the Deputy Judge’s starting point in any of the cases we were referred to is in some comments by Mr Wyand in Lamothe v Lamothe. In that case the testatrix made a will in 1993 in England which contained a specific devise of her major asset (a house in London) as well as disposing of her property generally. In 1995 she made a will in Dominica disposing of certain less valuable property in Dominica. It did not contain any specific reference to her London house but it did dispose of residue (“all my real and personal property whatsoever and wheresoever situated not hereby … specifically disposed of”). It also contained a general revocation clause and the question was whether that revoked the 1993 will. On the facts Mr Wyand held that there was clear and unequivocal evidence from the testatrix’s son that the testatrix had a positive intention to revoke the earlier will (at [52]). But at [51] he said:

“When one looks at the two wills and the way in which the 1993 will deals with the house at 89 Dunlace Road and the 1995 will deals with the less valuable property in Dominica the first impression is that the fourth defendant must be correct and they were intended to deal with the property in the respective jurisdictions.”

He then referred to a letter from the testatrix’s lawyer in Dominica which he found inconclusive: it said that the lawyer had explained the formal parts to her, but it also said that when he asked her to list the properties she intended to dispose of, she only listed the properties in Dominica. He continued:

“If the clause in the 1995 will dealing with the residue of Mrs Lamothe’s estate had not referred to her property ‘wheresoever situated’ then the two wills would have been easily reconcilable on this basis. The inclusion of those words in the definition of the residue in the 1995 will suggests an intention to deal with her entire estate in that will and not just the Dominican assets…

Had matters rested there I would have been in considerable doubt as to Mrs Lamothe’s intentions when she executed the 1995 will.”

That undoubtedly suggests that in the absence of the reference in the Dominican will to residue “wheresoever situated” then he might well have concluded that both wills should stand, and even with the inclusion of those words he would have found the question doubtful but for the other evidence. But I do not think Mr Wyand in these comments (necessarily obiter) was seeking to lay down any principle of general application. I read them as simply an illustration on the facts of the particular case of the principle that the court of probate can take into account any evidence which sheds light on the question whether a testator did have the necessary intention to revoke. It was certainly a striking fact that her Dominican will said nothing about her major asset in England which she had recently dealt with in an English will, and that her instructions to her Dominican lawyer were to deal with Dominican property. I can readily understand why in those circumstances Mr Wyand thought the position would have been doubtful without the evidence of her actual intention.

I do not therefore think that the Deputy Judge’s starting point is justified by the authorities if it was intended (as it naturally reads) as a statement of general principle. Ms Reed said that all he was doing was explaining where his analysis started from in this particular case but even on this basis, I think it goes too far. A testator must of course make a will somewhere. Where a testator is spending the majority of his time in a particular country (as Hartar appears to have been doing in India – we were told that although he often came to England he limited his visits for tax reasons and did not by 2016 keep a permanent home here, staying in whichever of his properties happened to be available) it is not surprising if that is where he makes a will. Nor is it surprising if a testator who makes a will in India uses the services of an Indian lawyer. And given that the vast majority of his estate was Indian (the value of his Indian estate being about £30m) it is also unsurprising that he dealt specifically in that will with his Indian properties.

None of this to my mind tells one anything about Hartar’s intentions in relation to his much smaller English estate. On the basis that he assumed that the jointly held properties would pass by survivorship to Jaswinder, the remainder of his English estate would have been only about £1m, a large sum for most people but a comparatively small one for him (and as explained below even this may overstate what Hartar would have thought his estate to be worth). It is not like a testator in England taking the trouble to instruct a foreign lawyer (in Belgium or Barbados or Dominica) to prepare a will dealing specifically with foreign property. In such a case the suggestion that the overseas will is a subsidiary one, intended only to affect property in that jurisdiction, is often a much more plausible one. Hartar’s 2016 will was undoubtedly his primary will, made in Punjab where he then mainly lived, and dealing with his main assets. I do not think this justifies any assumption at all, whether as a starting point or otherwise, that he intended it to be confined to his Indian estate. It is not as if it was headed “Will for India” or the like. When Mr Learmonth was asked what the position would have been if it had been, he said that even this would not have been enough. I am rather doubtful about that, but it is not necessary to pursue the question as those are not the facts.

The starting point in the circumstances I think is that Hartar is prima facie to be taken as having intended the 2016 will to be not only his primary will but his only will (which is what the revocation clause on its face provides), and if the Court is to reach some other conclusion it needs some basis in the evidence to do so. This is the approach adopted, correctly in my view, by the Deputy Master, who said in his judgment at [202]:

“The revocation clause, itself, constitutes strong evidence of the testator’s intention to revoke and clear and unequivocal evidence is required in order to establish that the literal language of the revocation clause did not represent the true intention of the testator.”

What such evidence was there? As the Deputy Master recorded, there was no evidence of Hartar’s actual intention in relation to revocation (see paragraph 31 above). This was despite the fact that Mr Bedi, who was responsible for drawing up the will at Hartar’s dictation, gave evidence and could have been asked what, if anything, was said about revocation at the time of execution of the 2016 will, and what Mr Bedi understood Hartar to have intended by the clause. In fact Mr Bedi was not asked any questions about this by any party. We received some submissions as to how realistic it would have been for Mr Blackett-Ord, who appeared for Jaswinder at trial and whose case was that the entire story of the execution of the 2016 will was a fabrication, to have asked questions of Mr Bedi as to what he understood Hartar to have intended by the revocation clause. I can see that that might have been difficult for Mr Blackett-Ord to do, even as a fall-back position, but I do not propose to consider this point any further. The fact is that there was no evidence at all before the Deputy Master as to Hartar’s actual intention one way or the other.

That meant that apart from the fact that the 2016 will was made in India and only disposed of Hartar’s Indian estate Ms Reed could only rely on various matters drawn from the surrounding circumstances. These were that Hartar did have English assets as well as Indian assets; that the 2016 will did not appoint any executors or make any provision for payment of debts; and that there was no suggestion that Mr Bedi had given Hartar any legal advice about the will, his role being simply that of an amanuensis. These seem to me to do no more than raise the question whether he intended to revoke the will so far as it dealt with English assets, and to fall a long way short of providing any answer to that question, let alone the clear evidence needed to displace the prima facie effect of the clause.

In addition Ms Reed sought to rely on two other matters. One, which I deal with below, is the presumption against intestacy. The other was that there was evidence which suggested that Hartar thought of himself as having wills for England and wills for India. This was objected to as a new point that was not taken either at trial or on the first appeal, and Ms Reed applied (if necessary) to amend her Respondent’s notice expressly to take the point. We permitted Ms Reed to show us the evidence that she sought to rely on de bene esse (on a provisional basis), but I will say straightaway that I did not find the evidence of any real assistance.

What it consists of is the following (these are not verbatim transcripts but taken from a very full note of the evidence):

Having considered all the passages from the evidence to which Ms Reed referred us, I remain unpersuaded that they provide any support for the submission that Hartar regarded himself as having wills for England and wills for India, or that this is a reason to regard the 2016 will as intended to be limited in its operation to India. It is not therefore necessary to consider whether this is a new point that can be objected to on that ground.

That leaves the final matter relied on by Ms Reed, which is the presumption against intestacy.

That there is such a presumption is beyond doubt. It is referred to in the standard textbooks such as Theobald on Wills (19^th edn, 2021) §26-061:

“It has often been said by the courts that it is unlikely that a testator intended to die intestate, and that therefore the courts should lean against a construction which creates intestacy, although perhaps not too heavily.”

And see Williams on Wills (11^th edn, 2021) §51.1 which is to similar effect.

It is possible to cite from judgments which exhibit varying degrees of judicial enthusiasm for the presumption. In re Harrison (1885) 30 Ch D 390 at 393-4 Lord Esher MR said:

“There is one rule of construction, which to my mind is a golden rule, viz., that when a testator has executed a will in solemn form you must assume that he did not intend to make it a solemn farce,—that he did not intend to die intestate when he has gone through the form of making a will. You ought, if possible, to read the will so as to lead to a testacy, not an intestacy. This is a golden rule.”

In re Edwards [1906] 1 Ch 570, however, Romer LJ refused to apply the presumption so as to distort the plain meaning of a will, saying (at 574):

“It is said that the Court leans against an intestacy. I do not know whether that expression at the present day means anything more than this, that in cases of ambiguity you may, at any rate in certain wills, gather an intention that the testator did not intend to die intestate, but it cannot be that, merely with a view to avoiding intestacy, you are to do otherwise than construe plain words according to their plain meaning. A testator may well intend to die intestate. When he makes a will he intends to die testate only so far as he has expressed himself in his will.”

And in re Abbott [1944] 2 All ER 457 at 459 Lord Greene MR said:

“Speaking for myself, I have always thought the suggested presumption against intestacy a very dangerous line of thought. It involves speculation as to the intentions of a class of persons, namely testators, which, as anybody with experience knows, is a highly capricious class. Some persons deliberately die intestate; some deliberately die intestate save as to certain items. It is very dangerous to place too much reliance on the supposed wish of testators in general not to die intestate. In my opinion, it is quite inadmissible to place any reliance upon it where, by so doing, violence is done to some clear disposition in a will.”

By contrast in the most recent decision of this Court on the question, Partington v Rossiter [2021] EWCA Civ 1564, Lewison LJ endorsed the presumption as follows (at [32]):

“The principle was applied to a potential partial intestacy in Barrett v Hammond [2021] WTLR 51. Although Mr Saifee accepted the existence of this principle of interpretation, he suggested that it could have no application where there would be a partial intestacy in any event; and the court should not attempt to minimise any possible intestacy. I disagree. The policy underlying the principle is, in my judgment, threefold. First, a court strives to give effect to the testator’s intention and purpose as expressed in a will; and the purpose of a will is (at least generally) to dispose of all the testator’s estate. Second, the rules of intestacy are to some extent arbitrary (to the extent that they may not represent the wishes of an individual testator, but are default rules for the population at large). Third, the testator’s own dispositions promote legal certainty. ”

Neither re Edwards nor re Abbott was cited in Partington v Rossiter, but I do not think these cases are as difficult to reconcile as might at first appear. Re Harrison was a case where there were two rival constructions, one of which would have meant that the will would always have been completely ineffective. Where the testatrix had clearly executed something in the form of a will, it is perhaps not surprising that Lord Greene thought that one should assume she intended it to have some dispositive effect rather than making no disposition of her property at all. Re Edwards on the other hand was a case where the will would have operated in most circumstances but the relevant clause had the effect that in the events which happened it made no effective disposition; re Abbott was a case where the will did contain an effective disposition but the question was whether it included all the testatrix’s estate. In each case the Court considered the relevant words of the will to be clear and was not willing to allow the presumption to displace them. Partington v Rossiter was unlike that: it was a case where the relevant disposition was ambiguous and the presumption was helpful to resolve the ambiguity. I was myself party to the decision and it was indeed a case where it was unlikely on the facts that the testator had intended to die partially intestate, as Lewison LJ said at [41]. What I think the cases taken overall illustrate is that although the presumption exists, and can in certain cases be useful for resolving ambiguities, its force varies from case to case with the circumstances.

The question therefore is whether the presumption is sufficient in the present case to overcome the natural meaning of the revocation clause. The Deputy Master thought not (see paragraph 31 above): it was not sufficient to establish a clear and unequivocal intention that the revocation clause was to be confined to revoking dispositions of Hartar’s Indian estate. On this point the Deputy Judge agreed, holding that the presumption was of no assistance and suggesting that Hartar might even have positively intended to leave the position as to his English estate unresolved.

That last suggestion is not one that strikes me as very plausible: Hartar had previously made provision for his English estate, and if he knew anything about the laws of intestacy he must have appreciated that to leave it undisposed of would be likely to give rise to disputes given the potential issues around his marriages, which he could scarcely have been ignorant of. What however seems to me far more plausible is the Deputy Master’s suggestion that he assumed that the bulk of his English estate had already been taken care of by putting it into joint names so that it would pass to Jaswinder by survivorship. That would I think have applied not only to the £10m of property shown in the IHT account as passing to her by survivorship, but also to another property (Balmoral Road) which had also been bought in their joint names. This was not included in the IHT account as passing to Jaswinder by survivorship, but that was only because a restriction had been entered on the register by the Official Receiver as Hartar’s trustee in bankruptcy, and the Deputy Master does not suggest that Hartar had appreciated the effect this had on the interests in Balmoral Road any more than the consequences of his bankruptcy, and its subsequent annulment, on the interests in the other properties.

All this means that Hartar might well have assumed that most of his English estate, including Balmoral Road, had already been dealt with, and that there would be a comparatively small rump of assets, worth somewhat less than £1m, still to deal with. We are not told what it consisted of although it included another property in his sole name (Marlborough Parade) which he might have been intending to transfer as well. I do not find it at all surprising that he might have intended to get round to dealing with this and such other remaining assets as there were in due course, but regarded this as less pressing and never did so before he died. That does not mean that he positively intended to die intestate, but it does mean the force of the presumption against intestacy is quite a weak one in the circumstances. I do not think it is enough to overcome the natural meaning of the general revocation clause. That was the view taken by the Deputy Master, and as can be seen that was informed by his view of the factual context, in which he was immersed in a way we cannot be. That would make it difficult in any event to reverse his decision, but for the reasons I have given I in fact take the same view as he did.

It follows that none of the matters relied on by Ms Reed is in my judgement sufficient to displace the natural meaning of the revocation clause in the 2016 will. I would therefore allow the appeal against the Deputy Judge’s decision and restore the conclusion of the Deputy Master that the 2016 will revoked the 2007 will in its entirety.

That makes it unnecessary to resolve the question whether the 2007 will was duly executed, but it raises a point of statutory construction of some general significance on which there is no other authority of this Court and it was fully argued, and it may be helpful if I express my views.

The 2007 will was made in India. In those circumstances s. 1 of the Wills Act 1963 provides:

“1 General rule as to formal validity

A will shall be treated as properly executed if its execution conformed to the internal law in force in the territory where it was executed, or in the territory where, at the time of its execution or of the testator’s death, he was domiciled or had his habitual residence, or in a state of which, at either of those times, he was a national.”

That means that the 2007 will will have been properly executed if executed in accordance with the law of India (the internal law in force in the territory where it was executed), or in accordance with English law (Hartar being a UK national at the time both of its execution and of his death). There was however no evidence of Indian law so it is to be treated as if it were the same as English law.

The relevant English law is found in s. 9 of the Wills Act 1837 (“s. 9”), as amended by the Administration of Justice Act 1982 (“AJA 1982”). As originally enacted, s. 9 provided as follows:

“IX Every Will shall be in Writing, and signed by the Testator in the Presence of Two Witnesses at one Time

And be it further enacted, That no Will shall be valid unless it shall be in Writing and executed in manner hereinafter mentioned; (that is to say,) it shall be signed at the Foot or End thereof by the Testator, or by some other Person in his Presence and by his Direction; and such Signature shall be made or acknowledged by the Testator in the Presence of Two or more Witnesses present at the same Time; and such Witnesses shall attest and shall subscribe the Will in the Presence of the Testator, but no Form of Attestation shall be necessary.”

That was supplemented by s. 1 of the Wills Act Amendment Act 1852 to clarify what would count as a signature “at the foot or end” of a will.

As amended by s. 17 AJA 1982, s. 9 provided as follows:

“9 Signing and attestation of wills

No will shall be valid unless—

(a)

it is in writing, and signed by the testator, or by some other person in his presence and by his direction; and

(b)

it appears that the testator intended by his signature to give effect to the will; and

(c)

the signature is made or acknowledged by the testator in the presence of two or more witnesses present at the same time; and

(d)

each witness either—

(i)

attests and signs the will; or

(ii)

acknowledges his signature,

in the presence of the testator (but not necessarily in the presence of any other witness),

but no form of attestation shall be necessary.”

That was the form in which s. 9 stood at the time of the execution of the 2016 will. (It has since been amended to allow, on a temporary basis, for remote witnessing, but this does not affect the present case.)

I have summarised above (paragraph 24) the facts as found by the Deputy Master. In more detail, the primary facts he found were as follows [75]-[76], [252]-[253]:

On the basis of these facts, the Deputy Master also found that Hartar had acknowledged his signature after Mr Khaira came into the room and joined Hartar and Mr Balraj Singh. There was no direct evidence of this, and no relevant questioning directed to any witness, but he was satisfied that it was the clear probability that in requesting Mr Khaira to sign the will, Hartar would have acknowledged by word or deed that he had already placed his signature on the will [113]. That conclusion is not challenged on this appeal.

On these facts the Deputy Master, upheld by the Deputy Judge, held that each of the requirements of s. 9 was satisfied. No question arises over those in s. 9(a) or s. 9(b), and on the basis that Hartar acknowledged his signature when requesting Mr Khaira to sign, no question arises over that in s. 9(c) either as that acknowledgment was in the presence of both Mr Balraj Singh and Mr Khaira. The only question is whether the requirement in s. 9(d) was satisfied, and this depends on whether this requires each witness to either (i) attest and sign the will or (ii) acknowledge his signature after the testator has complied with s. 9(c). Hartar only complied with s. 9(c) when he acknowledged the will after Mr Khaira came into the room. If therefore each witness has to comply with s. 9(d) after that has taken place, then although Mr Khaira’s signature would be compliant, Mr Balraj Singh’s signature would not be, as that took place before the acknowledgment. It would then be necessary to show that Mr Balraj Singh acknowledged his signature after Hartar had acknowledged his own.

Under s. 9 as originally enacted, it was established that the section required the following: (i) that the testator signed the will (or it was signed at his direction); (ii) that the testator’s signature was made or acknowledged in the presence of both witnesses at the same time; (iii) that each witness signed the will in the presence of the testator (but not necessarily each other); and (iv) that the signature of the attesting witnesses as required by (iii) took place after the testator had signed or acknowledged his signature in their presence as required by (ii). That was decided by Sir Herbert Jenner (later Sir Herbert Jenner Fust), very soon after the Act was passed, in re Allen (1839) 2 Curt 331 where he said:

“The natural construction of the words of the act, which are in the future tense, seems to be that when the signature is made or acknowledged, the witnesses shall then attest it”.

He expressed a similar view in Moore v King (1843) 3 Curt 243 at 253:

“I am inclined to think that the Act is not complied with, unless both witnesses shall attest and subscribe after the testator’s signature shall have been made and acknowledged to them when both are actually present at the same time.”

See also his decision to like effect in Cooper v Brockett (1843) 3 Curt 648 at 659. Those cases were consistently followed: see eg re Davies [1951] 1 All ER 920, and re Colling [1972] 1 WLR 1440.

In the latter case the facts were as follows. The testator was a patient in hospital. He started to sign his will in the presence of two witnesses, a nurse and another patient. But before he had completed his signature the nurse was called away to attend another patient. During her absence the testator completed his signature, and the other patient signed as witness. The nurse then returned and both the testator and the other patient acknowledged their signatures to her and she signed as witness. Ungoed-Thomas J felt compelled, with great regret, to hold that the will had not been duly executed. That was because the testator had not completed his signature in the presence of both witnesses, and it was essential that he should have signed the will or acknowledged his signature in the presence of both before either of them had attested to and subscribed: see at 1442D.

In 1977 the Law Reform Committee was asked to consider the law relating to the making and revocation of wills. It reported in 1980 in its Twenty-Second Report (The Making and Revocation of Wills), Cmnd 7902, and it was this report which led to the amendments to s. 9 made by s. 17 AJA 1982. No objection was made to our looking at it, and I think it would be admissible on ordinary principles at least to show the mischief at which the legislation was aimed.

At paragraph 2.9 of the Report the Committee referred to the requirement for two witnesses and recorded that none of those submitting evidence to them had suggested it be relaxed. They therefore recommended that there be no change in the requirement. At paragraph 2.10 they referred to the requirement that both attesting witnesses be present at the time when the testator makes or acknowledges his signature. Some of those submitting evidence thought this requirement could be relaxed but the Committee said:

“However, we do not consider that this requirement causes any great injustice and on the whole we think it is right that the three necessary participants in the “ritual” of execution of a will should be present together during the essential part of it, namely the signature or acknowledgement of his signature by the testator.”

At paragraph 2.11 they said that they would however like to prevent the recurrence of a case such as re Colling. Having set out the facts of the case they continued:

“The will was held to be invalid and some have thought that the invalidity resulted from the failure on the part of the testator to sign in the simultaneous presence of the two witnesses. However, in our view, it was not the requirement of simultaneity which invalidated the will in re Colling because in that case the testator did acknowledge his signature in the presence of both witnesses. What invalidated the will was the requirement that both the attesting witnesses must subscribe after the operative signature or acknowledgement of the testator. In Colling, when the testator acknowledged his signature one of the witnesses had already subscribed and did not actually subscribe again although he acknowledged his earlier signature. We do not think that testators’ intentions should be defeated by such a technicality. We therefore recommend that the effect of re Colling should be reversed by providing that an acknowledgement by a witness of his signature should have the same effect as his signature, just as under the present law a testator’s acknowledgement of his signature is as operative as his actual signature. The method of making would then consist of two successive steps whereby the testator would sign or acknowledge his signature in the simultaneous presence of the two witnesses and the witnesses would then sign or acknowledge their respective signatures.”

That being the background to s. 17 AJA 1982, it is now possible to address the rival arguments. The Deputy Master took the simple view that each of Mr Balraj Singh and Mr Khaira had satisfied the requirement of s. 9(d) because each had signed in the presence of Hartar and that was enough [116]. He considered that there was nothing in the amended version of s. 9 to replicate the meaning or effect of s. 9 as it stood before amendment [120].

He found support for that in Williams, Mortimer & Sunnucks on Executors, Administrators and Probate (21^st edn, 2018) at §9-24, §9-25. At §9-24 the editors explain the old law and say that many wills were refused probate for failure to comply with its requirements, adding:

“The most common circumstance was where the testator signed in the presence of only one witness, W1, who then duly attested and subscribed, and the testator later acknowledged his signature in the presence of W1 and a second witness, W2, who then also attested and subscribed. The will failed because it was witnessed by W1 before the testator acknowledged his signature to W2.”

They then give other examples, and continue at §9-25:

“In the case of deaths on or after 1 January 1983, s.9(d) as substituted provides that a will made in any of the circumstances listed above will now be valid, provided each of the witnesses acknowledges his signature to the testator after the testator has made or acknowledged his signature.”

I do not think these passages support the Deputy Master’s view as he thought they did. The execution of the 2007 will precisely matches the example cited from §9-24, and the statement in §9-25 that a will made in those circumstances “will now be valid” is qualified by the requirement that each of the witnesses acknowledges his signature to the testator “after the testator has made or acknowledged his signature”. That seems to me to be most naturally understood as referring to the testator having made or acknowledged his signature in accordance with s. 9(c), that is in the presence of both witnesses. If so, far from supporting the Deputy Master’s view, the editors would be espousing the opposite view that the requirements of s. 9(d) had to take place after the requirements of s. 9(c) were satisfied.

The Deputy Judge took the same view as the Deputy Master, and said that the editors of Williams, Mortimer & Sunnucks expressed the view at §9-25 that the rewording “removed the concept of time sequencing” [46]. But I do not think that this can actually be found in §9-25.

Ms Reed sought to uphold the decisions of both judges below. She said that the requirement that s. 9(d) be satisfied after s. 9(c) could not be found in the section and that it was hard to see what, in a case like the present, the purpose was of requiring the first witness to acknowledge his signature after the testator had acknowledged his to both witnesses. It could not be for the benefit of the other witness as it did not need to be in the presence of the other witness. Nor did it make sense to say it was for the benefit of the testator as the testator had already seen him sign.

There is some force in these points. But I prefer the construction put forward by Mr Learmonth. This is that the essential requirement for due execution by the testator is that he do something while all three individuals are together. This is what the Law Reform Committee called the “ritual” of execution and “the essential part of it” (see paragraph 2.10 of their report, cited at paragraph 85 above). The purpose of requiring this to be done with all three together at the same time – what the Committee called the requirement of simultaneity – is no doubt to guard against fraud and collusion. If one then asks what is the purpose of the requirement for attestation, it seems to me that it is for the two witnesses to confirm that they have seen this essential element being duly carried out. One cannot attest something that has not yet happened. So if the signature by witness 1 takes place before this essential part of execution, it is not a signature that attests that it has happened.

Seen in this light, it remains a requirement of s. 9 as amended by s. 17 AJA 1982 that the steps required take place in sequence. The testator must sign (or the will be signed at his direction). The testator must either do this in the presence of both witnesses together, or must acknowledge his signature in the presence of both together. This is the essential act. The witnesses must then confirm that they have witnessed this essential act. They do this either by signing the will, or, if they have already signed, by acknowledging their signature after the essential act.

This interpretation has the benefit of being consistent with the mischief identified in the Law Reform Committee’s report. It can be seen from paragraph 2.11 (cited in paragraph 86 above) that their concern was with the case such as re Colling where there was acknowledgment by the testator of his signature in the presence of both witnesses, but there was no due execution because witness 1 had signed already and did not sign again after such acknowledgment but only acknowledged his own signature. The solution they proposed was that it would suffice if, as in re Colling, witness 1 acknowledged his signature. But there is nothing to suggest that they thought it would be enough if witness 1 had simply signed before the essential event, and the last sentence of paragraph 2.11 is directly contrary to such a view.

This interpretation also has the support of all the textbooks we were referred to. As well as Williams, Mortimer & Sunnucks which I have referred to above, it is the view taken by Borkowski’s Law of Succession (4^th edn, 2020) at §5.2.6.1 (“the only logical possibility: until the testator does what he is required to do there is nothing to witness”); Megarry & Wade, The Law of Real Property (9^th edn, 2019) at §13-026; Parry & Kerridge, the Law of Succession (13^th edn, 2016) at §4-18 (“There is still a required order of events”); Theobald on Wills (19^th edn, 2021); and Williams on Wills (10^th edn) §12.5 (“the AJA 1982 does not change this basic rule relating to the order of signing and attestation”).

It also has the support of Lewison LJ, albeit obiter, in Barrett v Bem [2012] EWCA Civ 52 at [18] where he said that the conditions in s. 9 “set out what is in effect the temporal sequence for their fulfilment”. And most recently HHJ Paul Matthews, sitting as a Judge of the High Court, considered the question at some length in James v Scudamore [2023] EWHC 996 (Ch) at [94]-[105]. That too was obiter but in an erudite and cogent analysis HHJ Matthews disagreed with the Deputy Judge in the present case and held that the event referred to in s. 9(d) must follow that in s. 9(c) (at [101]). I agree with his reasoning and with his conclusion.

In my judgement therefore the Deputy Master and Deputy Judge were wrong to regard Mr Balraj Singh’s signature, which took place before Hartar did anything in the presence of both witnesses, as sufficient to satisfy the requirements of s. 9(d).

That is sufficient to answer the question of principle on what s. 9 requires. The remaining question under this ground of appeal is whether the Deputy Master was entitled to find, as he did, that on the balance of probabilities Mr Balraj Singh did acknowledge his signature after Hartar acknowledged his own signature to Mr Khaira. This raises no point of general application and turns on the particular facts of the present case, and given the view I take on ground 1 (which I understand Arnold and Carr LJJ are in agreement with), has no effect on the outcome of the appeal.

What the Deputy Master found was as follows. At [123] he found that as with Hartar’s own acknowledgment to the two witnesses there was no direct evidence that Mr Balraj Singh made the requisite acknowledgment. He thought that unsurprising given that the emphasis at trial was not on the formalities but on authenticity, and that Mr Khaira’s evidence that he was not initially present only came out in his oral evidence and was “un-foreshadowed and, I think, unexpected.” Nevertheless the result was, as he said, that there was a “lacuna in the evidence” such that the Court had to decide the point by having “regard to all the circumstances and the core probabilities”.

At [125] he said that this was not a meeting of strangers: Mr Balraj Singh had been working for and with Hartar for some 30 years, Mr Khaira was living in the Chandigarh property at the time of execution of the 2007 will, and Mr Tiwana had been Hartar’s factotum for many years. He continued at [126]:

“In that context, there is no reason to believe that when Mr Khaira was summoned by Hartar and asked to witness the will, the parties and, in particular, Mr Balraj Singh stood in silence. It is much more likely, looking at the circumstances, that there was, among these close acquaintances, some conversation, or discussion, and that that discussion would have included an assertion, or indication, by Mr Balraj Singh that he had already signed the will, as witness. The question, then, is as to sequence and, as to that, the greater probability, as it seems to me, is that such an assertion, or indication, would have occurred after Hartar had, himself, requested Mr Khaira to sign, as witness and, in so doing, acknowledged his own signature on the will. It seems to me very unlikely, given Hartar’s dominant personality, as adverted to at various later stages of this judgment, that Mr Balraj Singh would have adverted to his own signing as witness before Hartar had, himself requested Mr Khaira to witness his signature and, thereby, acknowledge that signature.”

At [127] he concluded that had it been necessary he would, while recognising the decision as “marginal”, have been prepared to determine that, on the balance of probability, both Mr Khaira and Mr Balraj Singh had attested the will after Hartar had acknowledged his signature in their joint presence.

Mr Learmonth said that there was no evidence at all as to what happened and that this was pure speculation. He pointed out that the burden of proof lay on those propounding the will, that is Jaswinder, and that no such evidence was elicited, although questions could have been asked of the witnesses; and that it would not be enough if Mr Balraj Singh had said nothing at all when Mr Khaira came into the room, or if he had spoken first to explain that he had signed, and then Hartar had agreed. The Deputy Master had inferred what had happened in order to fill in a gap in the evidence but had no real basis for doing so at all.

Ms Reed said that it was no-one’s fault that there was no evidence directed at this point. Before trial everyone had expected the evidence to be to the effect that the will was signed by Hartar and the two witnesses while they were all together (which is what the attestation clause said), and Mr Khaira’s evidence that he was not there when Hartar and Mr Balraj Singh signed came as a complete surprise. In those circumstances the Deputy Master had to do the best he could; he had heard the witnesses over 9 days and was in a position to form a judgment as to the probabilities. Very little was needed in order to find an acknowledgment, and he was entitled to reach the conclusion he did.

These arguments seem to me finely balanced with points well made on both sides. Had it been necessary to resolve them I would of course have done so. But, as I have already said, in the light of my conclusion on ground 1 nothing now turns on this particular question. Given that it raises no point of principle, I do not think it necessary to reach any conclusion on the point, and do not propose to do so.

Ground 3 complains of various suggested irregularities in the way the Deputy Judge dealt with the revocation issue. Given my conclusions on ground 1, nothing now turns on this ground and I do not think any useful purpose would be served by considering it. I propose to say no more about it.

Ground 4 concerns the question whether the Deputy Judge should have granted probate of only the 2007 will (as he did) or of the 2007 and 2016 wills together. The argument depended on the 2007 will not having been wholly revoked. If, as I would hold, the appeal is allowed on ground 1, the 2007 will will have been wholly revoked and the question does not arise. Again in those circumstances I see no useful purpose in considering the question.

I would allow the appeal on ground 1, set aside the Deputy Judge’s order, and restore the Deputy Master’s order declaring that the 2007 will was wholly revoked.

I agree.

I agree with the judgment of Nugee LJ. I would only add a short point concerning the interpretation of section 9 of the Wills Act 1837 as amended by the Administration of Justice Act 1982. As Nugee LJ notes in paragraph 96, Lewison LJ said in Barrett v Bem that “the conditions” in section 9 “set out what is in effect the temporal sequence of their fulfilment”. I agree with this observation, and I think it is worth spelling out a little more fully what I understand Lewison LJ to have meant by it. At first blush, section 9 does not say anything about the temporal sequence in which the requirements it imposes must be satisfied. When one considers the requirements imposed by paragraphs (c) and (d), however, it is implicit that paragraph (c) must be fulfilled prior to paragraph (d). This is because the witnesses can only “attest”, which is one of the ways in which paragraph (d) can be satisfied, if they have witnessed the signature or acknowledgement required by paragraph (c). This was the reasoning of HHJ Matthews in James v Scudamore, with which I also agree.