Marley v Rawlings & Anor

On 17 May 1999 a married couple, Alfred Thomas Rawlings and Maureen Catherine Rawlings, had an appointment with their solicitor at their home for the purpose of executing the wills that he had drafted for them. The wills were very short and, except for the obvious differences necessary to reflect the identity of the maker, were in exactly the same terms. Each spouse left his or her entire estate to the other spouse or, if the spouse failed to survive him or her by one calendar month, to Terry Michael Marley who was not related to them but whom they treated as their son.

Mr Rawlings’ will read:

“This is the last will of me ALFRED THOMAS RAWLINGS of [address].

1.

I REVOKE all former wills and testamentary dispositions.

2.

IF MY wife MAUREEN CATHERINE RAWLINGS of [address] survives me by a period of one calendar month then I appoint her to be the sole Executrix of this my will and subject to my funeral and testamentary expenses fiscal impositions and all my just debts I leave to her my entire estate.

3.

IF MY said wife MAUREEN CATHERINE RAWLINGS fails to survive me by a period of one calendar month I appoint TERRY MICHAEL MARLEY of [address] to be the sole Executor of this my will and subject to my funeral and testamentary expenses fiscal impositions and all my just debts I leave to him my entire estate.

IN WITNESS whereof I the said ALFRED THOMAS RAWLINGS have hereunto set my hand the day of 1999.

SIGNED by the testator in our presence and then by us in his:- [space for witnesses]”

In Mrs Rawlings’ will, references to Maureen Catherine Rawlings were substituted for the references to Alfred Thomas Rawlings, references to “my wife” were replaced with references to “my husband”, and references to “her” were replaced with references to “him”. Instead of referring to the “testator”, the reference was to the “testatrix”.

By mistake, Mr Rawlings signed the will meant for Mrs Rawlings and Mrs Rawlings signed the one meant for Mr Rawlings. The solicitor and his secretary attested their signatures. No one noticed the error then, or on the death of Mrs Rawlings which occurred in 2003.

When Mr Rawlings died in August 2006, the error came to light. Mr Rawlings owned the property in which he had lived with Mr Marley on a joint tenancy so that passed to Mr Marley by survivorship. However, there was some £70,000 net in the estate and a dispute arose between Mr Marley on the one hand and Mr and Mrs Rawlings’ two sons, Terry and Michael Rawlings, on the other as to whether the will that Mr Rawlings had signed (to which I will refer hereafter as “the will”) was a valid will in which case Mr Marley would inherit or whether he had, in fact, died intestate in which case the two sons would inherit.

Mr Marley began proceedings in the Chancery Division for probate of the will which came before Proudman J. Her judgment records that the claim advanced before her was only for rectification of the will and probate in solemn form of the will as rectified though counsel for Mr Marley, Ms Teresa Peacocke, told us that it had not been conceded that the will could not be admitted to probate as it was. In any event, on 3 February 2011 the judge dismissed Mr Marley’s claim. It is against that determination that he now appeals, the two Rawlings sons being the respondents to the appeal. If Mr Marley fails in the appeal, he will look to Mr and Mrs Rawlings’ solicitor for damages reflecting the value of the estate that he has lost.

There can be no doubt as to what Mr and Mrs Rawlings wanted to achieve when they made their wills and that was that Mr Marley should have the entirety of their estate and their sons should have nothing. Unfortunately, that certain knowledge is not what determines the outcome of this appeal. The answer is contained in the law relating to the making and rectification of wills.

Wills Act 1837 in its present form

The first port of call is the Wills Act 1837 which I will set out as amended over time.

Section 3 provides that:

“It shall be lawful for every person to devise, bequeath, or dispose of, by his will executed in manner herein-after required, all real estate and personal estate which he shall be entitled to, either at law or in equity, at the time of his death, and which, if not so devised, bequeathed, or disposed of, would devolve upon his executor or administrator….”

Section 1 deals with the meaning of certain words in the Act and includes the following elucidation of the word “will”:

“The words and expressions herein-after mentioned, which in their ordinary signification have a more confined or a different meaning, shall in this Act, except where the nature of the provision or the context of the Act shall exclude such construction, be interpreted as follows; (that is to say) the word “will” shall extend to a testament, and to a codicil, and to an appointment by will or by writing in the nature of a will in exercise of a power, and also to an appointment by will of a guardian of a child, and to any other testamentary disposition;…”

Section 9 deals with the signing and attestation of wills. So far as is material, it is 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)

[requirements of the witnesses]”

Some history of the Wills Act and other provisions relating to wills

In order to understand the jurisprudence concerning the Wills Act, it is necessary to know a little about the history of the statute itself, there having been significant amendments to it since it was passed in 1837, and about other developments in the law.

A fairly early amendment was to relax the requirement that the will be signed by the testator at the foot or end of it. As originally drafted, section 9 read:

“no Will shall be valid unless it shall be in Writing and executed in the manner herein-after 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.”

The Wills Act Amendment Act 1852 provided that a will could be valid if the signature was

“so placed at or after, or following, or under, or beside, or opposite to the end of the will, that it shall be apparent on the face of the will that the testator intended to give effect by such his signature to the writing signed as his will;….”

There was further relaxation of the formal requirements for making wills when section 9 in its current form was substituted for the original section 9 by the Administration of Justice Act 1982. The changes included the abandonment of all direction as to where on the document the testator’s signature had to be, it only being required now that “it appears that the testator intended by his signature to give effect to the will”.

The 1982 Act also introduced, for the first time, provisions which permitted rectification of a will by inserting new words into it as opposed to omitting from probate words of which the testator did not know and approve. So far as material to this appeal, those provisions are set out below under the next heading.

Section 20(1) establishes the power to order rectification. It reads:

“20 Rectification

(1)

If a court is satisfied that a will is so expressed that it fails to carry out the testator’s intentions, in consequence—

(a)

of a clerical error; or

(b)

of a failure to understand his instructions,

it may order that the will shall be rectified so as to carry out his intentions.

(2)

An application for an order under this section shall not, except with permission of the court, be made after the end of the period of six months from the date on which representation with respect to the estate of the deceased is first taken out.”

Section 21 deals with the interpretation of wills. It reads:

“21 Interpretation of wills—general rules as to evidence

(1)

This section applies to a will—

(a)

in so far as any part of it is meaningless;

(b)

in so far as the language used in any part of it is ambiguous on the face of it;

(c)

in so far as evidence, other than evidence of the testator’s intention, shows that the language used in any part of it is ambiguous in the light of surrounding circumstances.

(2)

In so far as this section applies to a will extrinsic evidence, including evidence of the testator’s intention, may be admitted to assist in its interpretation.”

Proudman J considered that section 9(b) of the Wills Act 1837 resolved the case. She said,

“21.

In my judgment s 9 (b) of the 1837 Act provides a complete answer to the claim, namely that the testator did not intend by his signature to give effect to the will which he signed. If asked whether he did he would not have said, ‘yes, subject to correction of errors by substituting my wife’s name for mine wherever it occurs’. He would simply have responded, ‘no, of course not, that is my wife’s will’.”

Therefore, she said (§27),

“It seems to me that the claimant must fail on the ground that the requirement of s. 9(b) of the Wills Act 1837 as amended is not met.”

She went on to consider whether, if she was wrong about that, she had jurisdiction under section 20 of the 1982 Act to rectify the will and determined that she did not. It was common ground that there had not been a failure to understand the testator’s instructions within section 20(1)(b) and she concluded that there had not been a clerical error within section 20(1)(a) because the error was not to do with the wording of the will which was correctly expressed but simply given to the wrong person for signature.

The prevailing theme of the appellant’s submissions was that the role of the courts is to give effect to the intention of the testator and that Proudman J could and should have done this here if she had applied section 9 of the Wills Act correctly. Although the only English authorities which are directly in point, In the Goods of Hunt (1875) LR 3 P&D 250 (Re Hunt) and In the estate of Fanny Deborah Meyer [1908] P 353 (Re Meyer) (and see also In the Goods of F.S (1850) 14 Jur.402) were unhelpful to the appellant’s case, they were either wrongly decided and/or distinguishable and/or outdated and not representative of the current state of the law. There are instructive authorities from South Africa, New Zealand, Australia, Canada, the United States and Jersey in which the courts have given effect to wills in similar circumstances to the present ones. I will say more later about the international authorities on which the appellant relies.

The appellant submitted that both section 9(a) and section 9(b) were satisfied. They require simply that the testamentary document is in writing and signed in such a way that it is apparent that the testator intended by his signature to give effect to it as a will.

Here there was a document of testamentary character. When Mr Rawlings applied his signature to it when it was placed in front of him, he did “intend…by his signature to give effect to the will” as his will. He did not intend that it should be expressed as in fact it was, but that is irrelevant for the purpose of section 9(b). What section 9(b) is directed at is the requirement that the testator should intend the signing of the will to be an act of execution. The new section 9(b) introduced by the Administration of Justice Act 1982 was designed to relax the law governing the position of the testator’s signature on the document and that context must be borne in mind in construing the subsection.

Extrinsic evidence is admissible in determining what was to be regarded as the deceased’s will for the purposes of probate; the will prepared for Mr Rawlings but signed by his wife is an important part of that evidence. It confirms that the provisions contained in the will he signed were (with the necessary adjustments) what he intended and approved.

It seemed to me that Ms Peacocke envisaged four possible alternative courses by which the position that has arisen could have been corrected:

In the respondents’ submission, the will fails on the grounds both of formal validity (it does not comply with section 9(b)) and substantial validity (the terms of it did not have the testator’s knowledge and approval).

Section 9 was not satisfied, they argued, because whilst Mr Rawlings intended to sign a will, he did not intend to sign this will. He did not, in fact, sign his will at all and nor did Mrs Rawlings sign hers. This defect of execution means that there is no will which can be admitted to probate and/or construed or rectified. Provisions alleviating the consequences of a failure to comply strictly with formalities, for example by allowing a will to be operative where there had been substantial compliance with them, could have been introduced in the 1982 Act but were not. The court should not take the route that Parliament did not take. Neither should the court circumvent the formalities in an effort to give effect to the testator’s intentions.

The respondents submitted that the original requirement for a signature at the end of the will showed that the purpose of the signature was to authenticate the actual provisions in the will and to prevent anything being slipped into it after signature. The current provision must be interpreted bearing that in mind. The intention required by section 9(b) is an intention to give effect to this document as a will; the testator must intend to make this will not another one.

This case is not the same as a case (which may satisfy section 9(b)) in which the testator signs his own will but that will contains a mistake. The difference between the two situations is not simply one of degree. In the latter situation, the testator signs his own will. Here he has signed someone else’s. Putting it another way, this case does not simply concern omissions or other errors in a will but a wholesale substitution of different provisions for the provisions that should have featured in it.

Re Meyer and Re Hunt have survived unchallenged for over a century and were cited with approval by the Court of Appeal in Corbett v Newey [1998] Ch 57 @ 68 and Proudman J was right to follow them.

If section 9(b) is not satisfied, there is no valid will so there can be no rectification, no probate.

If, however, the court were to find that section 9(b) is satisfied, the will would still not be effective because the testator did not have knowledge of or approve the terms in it.

This is not the sort of situation in which words have been included without the knowledge and approval of the testator by inadvertence (for example a clause which accidentally revokes a will in another jurisdiction) and can be excluded from the grant of probate. The power to do that was established before Re Hunt and Re Meyer and mentioned in those cases but it did not save the wills there because, as here, the deceased did not know of or approve any of the contents of the document s/he signed because it was a totally different document from the one intended. There can be no distinction in principle between admitting the document to probate where it is the will of the deceased’s spouse and admitting to probate another document completely (such as the will of a complete stranger or another type of document altogether such as a contract) that the deceased accidentally “executed” in the belief that it was his will.

Furthermore, it is not permissible to focus on the one provision of the will which remains relevant at the time the question of probate arises and to admit that provision alone to probate because it happens to coincide with what the testator intended; the validity of the will cannot depend on the factual situation at the time when the matter falls to be decided.

As to the argument that the will could be construed in such a way as to correct the errors, a reference to “my husband X” cannot be construed as a reference to “my wife Y” – that would not be construction but re-writing. The cases cited by the appellant in support of some such construction are entirely different.

Neither can the will be rectified under section 20. The error was not clerical. A clerical error is an error in the process of recording the testator’s intended words, see Wordingham v Royal Exchange Trust Co Ltd [1992] Ch 412 as approved in Re Segelman [1996] Ch 171. Section 20 is aimed at drafting errors and this was not a drafting error but a defect of execution of the will of the same type as failing to ensure that the witnesses were present at the same time as required by section 9(c). It would be wrong to circumvent the requirements for due execution by rewriting what the would-be testator has signed.

The respondents submitted that there are foreign authorities which are against the appellant as well as those upon which he relied. All of the authorities upon which the appellant relied derive from the New Zealand case of Guardian Trust & Executors Co. of New Zealand Ltd v Inwood [1946] NZLR 614. The New Zealand court there deliberately departed from Re Hunt and Re Meyer for reasons which were specious and the effect of the decision was subsequently confined by the Supreme Court of New Zealand in Re Foster [1956] NZLR 44 although later followed in McConagle v Starkey [1978] 3 NZLR 635. Decisions in other jurisdictions have been made per incuriam and/or in a different legal context from the law applicable here.

I agree with the respondents that the logical place to start – indeed, it seems to me the only place to start - is with the question of the formal validity of the will.

Furthermore, I do not see how the question of formal validity can be approached except upon the basis of the will as it stands, albeit in the context of the circumstances surrounding its making. I find the suggestion that the court can rectify the would-be will under section 20 before considering whether it meets the formal requirements of section 9 of the Wills Act unappealing in principle. I also find it difficult to envisage how it would operate in practice.

Section 20(1) provides that if the court “is satisfied that a will is so expressed that it fails to carry out the testator’s intentions….it may order that the will shall be rectified so as to carry out his intentions”. The question that arises is what is meant by the word “will” in that context. It does not appear to be defined anywhere for the purposes of the section. Whilst not inserted into the Wills Act 1837 in the way that certain other provisions are by the 1982 Act, section 20 is plainly intended to operate alongside the Wills Act and I would have thought that a similar approach should be taken to the terminology in both. True it is that there are in the Wills Act certain provisions in which the word “will” means no more than “would-be will” – section 9 is one such and section 7 (wills made by under-18 year olds) is another – but the restricted meaning is in each case clear from the context. Elsewhere, the word “will” is used to signify a valid will and that seems to me the sensible interpretation of it in section 20 also.

Nothing in section 20 indicates to the contrary. In fact section 20(2) offers some, albeit possibly weak, support for this interpretation. It establishes the time limit for the making of an application for rectification. The time period is fixed by reference to the date on which representation with respect to the estate is taken out and representation could only be taken out in relation to a valid will. I say that the support offered by this is possibly weak because there is nothing in the words of the subsection to show that the draftsman contemplated that in every case representation would be taken out or that it was a pre-requisite to an application for rectification. However, it would be odd if the section was intended to permit applications to be made in relation to wills which were not valid as they stood and in relation to which representation could not therefore be taken out but, in contrast to the situation pertaining to valid wills, imposed no time limit at all as to when in such a case an application for rectification had to be made.

In construing section 20, it is also material to bear in mind the function of that section in comparison to the function of section 9. Section 9 is directed not at the content of the will but at questions of form. Is it in writing? Is it signed by the testator or some other person in his presence and by his direction? Does it appear that the testator intended by his signature to give effect to the will? Are the witness requirements met? In contrast, section 20 of the Administration of Justice Act 1982 is directed at the substance of the will. Does the will fail to carry out the testator’s intentions for a reason that falls within the section? As a matter of logic, issues of substance naturally come after issues of form have been resolved. Making amendments of substance by way of rectification of a would-be will which, in its unrectified state, would not satisfy the formalities strikes one instinctively as putting the cart before the horse and I find it very difficult to conceive of a set of circumstances in which rectification ordered under section 20 could enable an otherwise invalid would-be will to satisfy the requirements of section 9 without rocking the very foundations of that section at the same time.

For all these reasons, I would therefore treat the word “will” in section 20 as referring to a valid will. It follows that I must commence my analysis with section 9 of the Wills Act 1837 and the question of whether the document signed by Mr Rawlings satisfies the requirements of that section. Recognising that the law at the time that Re Hunt and Re Meyer were decided was not precisely as it is now, I intend to start my examination by referring only to the present terms of section 9. Nevertheless, Re Hunt and Re Meyer merit careful consideration as do the authorities from overseas to which the appellant invited our attention and I will turn to them later to test my provisional conclusions.

The document Mr Rawlings signed undoubtedly had the necessary testamentary quality for section 9. It was a will not a utility bill or a contract for example. There was no problem in relation to the witnesses either so no question arises in connection with sections 9(c) and (d). The focus can therefore be on sections 9(a) and (b).

Section 9(a) contains several hurdles. The will must be in writing and it was. The other feature of section 9(a) is the requirement that the will be signed. This is not a case of signature by someone else on behalf of the would-be testator so what is required is that it be signed “by the testator”. Argument was not directed to this particularly during the hearing before us but there is, to my mind, a real question as to whether this will was signed by “the testator”. The will had been drawn up for Mrs Rawlings and said that it was “the last will of me Maureen Catherine Rawlings”. The obvious person to describe as the testator in relation to this will was therefore Mrs Rawlings and she did not sign it. Mr Rawlings was intending to be a testator but not through the medium of this will.

If Mr Rawlings’ signature did constitute a signature by the testator, the question arises as to whether it appears that he intended by his signature to give effect to the will as required by section 9(b). I doubt very much that the draftsman of section 9 expressly considered how the subsection would operate in the “crossed will” situation. However, the words of the subsection must yield a solution for that conundrum.

A practical approach must be taken to the requirements as this court said in Wood v Smith [1993] Ch. 90 (C.A.) 3 WLR 583. Scott LJ said @ 111:

“There can be no doubt but that the Parliamentary intention in substituting the new section 9 for the original section was to simplify the requirements for the execution and witnessing of a will. The requirements of paragraph (a) and paragraph (b) are in my judgment complementary. Paragraph (a) requires a signature. Paragraph (b) requires that the signature be intended to give effect to the will. These requirements demand a practical approach.…. ”

Having declined to draw assistance from the authorities on section 40 of the Law of Property Act 1925 where the object of the signature was to identify the party to the contract, Scott LJ continued:

“The object of a signature by a testator, by contrast, is to authenticate the written document in question as the will of the testator. A normal signature, placed at the foot of a testamentary document would in most cases carry the implication that the testator intended the signature to give testamentary effect to the document….”

Wood v Smith concerned a rather different problem from the present case. The testator had written the testamentary document himself, heading it “My Will by Percy Winterbone” but he had not otherwise signed it, saying when this was pointed out to him that he had signed it at the top and it could be signed anywhere. The court accepted this as sufficient for due execution of the will notwithstanding that the testator wrote his name rather than signing it in the conventional sense and notwithstanding that he did this before the dispositive provisions of the will had been written. It held that by writing his name and the dispositive provisions in a single operation, the deceased had provided evidence to show that he intended his signature to give effect to those provisions.

Scott LJ’s observation that the object of a signature by a testator is “to authenticate the written document in question as the will of the testator” is interesting. There seem to me to be two elements in that. By his signature, the testator is not only executing the document as a will with immediate effect but also (at least in a broad sense, subject to adjustment arising from issues of want of knowledge and approval and matters within the scope of section 20) confirming that the document represents his testamentary intentions. This dual function is consistent with the historical roots of the present provision. Part of the motivation for the original requirements as to the position of the signature on a will was, it seems to me, the desire to provide a simple and reliable way of establishing, without oral evidence, that the will contained the provision that the testator wished to make. So initially the signature executing the document had to be at the foot or end of it and therefore almost inevitably had to be applied after the substance of the will had been set out, and even when that was relaxed in the 1852 amendment, it was still provided that no signature would be operative to give effect to any disposition or direction underneath or following it or which was inserted after the signature was made. In support of this view of the purpose of the signature, it is worth noting that in devising rules to apply after the Wills Act, in Guardhouse v Blackburn (1865-69) LR 1 P & D 109 @ 116, Sir J P Wilde included the following:

“Secondly, that except in certain cases, where suspicion attaches to the document, the fact of the testator’s execution is sufficient proof that he knew and approved the contents.”

I return to Mr Rawlings, keeping in mind these considerations. Undoubtedly he intended to give effect to a will when he signed the document placed in front of him but not to the will that was in front of him and which he signed. It is also beyond doubt that he intended to give effect to testamentary provisions which were very like those in the will that he signed but they were not precisely the same testamentary provisions because the names were different. I am inclined to the view, for reasons which I shall attempt to set out, that in these circumstances, section 9(b) was not satisfied.

The wording of the subsection does not seem to me to accommodate what happened here. It refers to “the will” which must, in my view, mean the will which is in writing before the would-be testator, the one Mr Rawlings actually signed. That is the will to which the testator must have intended to give effect and he did not so intend.

What is required is that “it appears” that the testator intended by his signature to give effect to the will. The suggestion was made in the hearing before us that what is material in this regard is the subjective intention of the testator. However when I consider the wording of the subsection and think again of its forbears, as well as reviewing the authorities concerning both the old and new provisions, it seems to me that the concern of the subsection is, at least in part, with what is apparent on the face of the document. The 1852 requirement was that the signature should be “so placed….that it shall be apparent on the face of the will that the testator intended to give effect by such his signature to the writing signed as his will” and one suspects that the words “it appears” in the present subsection owe something to the words I have picked out in italics in this earlier formulation. I would not wish to overstate the significance of this. Wood v Smith demonstrates that for the purposes of the present subsection, enquiry goes wider; Scott LJ placed reliance on the evidence that the deceased indicated in clear terms to his witnesses that he regarded his name at the start of the will he was engaged in writing as his signature (@589E). But, for what it is worth, what appears from the face of the will in our case is that Mr Rawlings cannot have intended to give effect to what was written as his will because it was plainly his wife’s.

Leaving the confines of the will itself does not assist in establishing the appellant’s case and achieving what Mr Rawlings wanted either. Subjectively, he intended to make a will but not this one, not the will before him. As Proudman J said, if asked whether he intended to give effect to the will which he signed, he would have said “no, of course not, that is my wife’s will”.

I cannot put this case into the same class as a will which is the will of the testator but contains errors which can be corrected by construction, omission or rectification. If the difference between them is a question of degree, then there is ample clear water between the two situations to justify why this one falls on the wrong side of the line and the other does not. This is simply not at all the same sort of situation as, for example, that in which the testator was described by her married name at the commencement of the will and in the testimonium clause but her maiden name was used in connection with her mark by which she executed the will (Re Clarke (1858) 1 SW. & Tr. 22). Nor is it the same in my view, pace some of the judges in the overseas authorities, as the testator signing a carbon copy of his own will. But, for my part, I am actually inclined to see the difference as a difference in kind. It is difficult to articulate why any more specifically than to say that one situation involves the testator’s will but with errors in it and the other does not involve his will at all. In no way could Mr Rawlings be seen as authenticating the contents of the will in front of him by his signature because they were not the testamentary provisions he intended to make; he was, furthermore, in no position to execute that will because it was simply not his will.

How does my provisional conclusion compare with the reasoning in Re Hunt and Re Meyer and withstand the challenge of the overseas authorities?

The deceased in Re Hunt prepared wills for herself and for her sister. The legacies in each were almost identical except that one left a legacy to Stockwell Orphanage and the other a legacy to Haverstock Hill Working Orphan School. Their intention was that on the death of either of them, the survivor should enjoy their joint property for life and this was achieved in each will by the words “The residue….I leave to my sister [name] for her life”. One will was headed “The last will and testament of me, Ann Hunt” and the other was for Sarah Hunt. In the last will and testament of Ann Hunt, the name of the sister in the residuary legacy was Sarah and the position was the reverse in Sarah Hunt’s will. By mistake, each sister executed the will which had been prepared for the other. Most of those interested on intestacy consented to the document executed by Sarah Hunt being recognised as her will and to probate being granted but some were abroad and could not be contacted so an application was made, ex parte, for probate. It was argued that if probate were to be granted, a court of construction would receive evidence that the sister called Sarah in the paper was in fact Ann and the mere fact that a wrong name was inserted at the commencement of the will would not vitiate it.

Sir J. Hannen clearly wished to grant the application but refused to do so. His judgment is short and I will set it out in full. The thread that runs through it is that the document that Sarah Hunt signed as her will was not her will and could not be read as such, even bearing in mind the court’s power to give effect to a will containing an accidental misdescription or to exclude from a will a feature of which the testator did not know or approve. It reads:

“I should be glad to give effect to the intentions of the testatrix, by granting probate of this instrument, if I could, but I must not allow myself to be led away from what appears to me to be very plain ground by such a desire. No doubt there has been an unfortunate blunder. The lady signed as her will something which in fact was not her will. If I were to attempt to read it as her will, it would lead to a variety of absurdities. She leaves to her sister Sarah, that is to herself, a life interest in a portion of her property, and all the furniture, plate, &c, which she holds in part with herself. I am asked to treat this as a misdescription. If by accident a wrong name had been introduced, and it was clear what person was intended, the Court would give effect to the instrument, providing the mistake could be corrected. But it would be contrary to the truth in this case if I acted upon such an assumption. If I were to put such a construction upon this will, I should be assuming in order to do substantial justice, what everyone who hears me would know is contrary to the fact. And no Court ought to base its judgment on something wholly artificial, and contrary to what everyone must see is the real state of the circumstances. It is enough to say that there has been an unfortunate blunder. A paper has been signed as the lady’s will, which, as it happens, if treated as her will, would to a great extent, although not entirely, carry out her wishes. But in one respect it does not, for by it a legacy is bequeathed to one charity which she intended to leave to another. As regards this legacy, it is suggested that it might be treated as if the deceased did not know and approve of that part of the will, but she did not in fact know and approve of any part of the contents of the paper as her will, for it is quite clear that if she had known of the contents she would not have signed it. I regret the blunder, but I cannot repair it.” [my italics]

In Re Meyer two sisters executed mutual wills. Later, they executed codicils which were in similar terms but, by mistake each executed the one intended for the other sister. When Fanny Meyer died, the court was asked to grant probate of the will and codicil, omitting from the codicil the names of the testatrix and her sister wherever they were mentioned. The other sister, who was the residuary legatee, consented to the motion.

Sir Gorrell Barnes, President, refused probate in relation to the codicil. This was a decision he arrived at without reference to Re Hunt which was not cited to him. His short judgment includes the following:

“…it is quite clear that this lady, though her signature is on the document, never meant to sign this particular codicil at all. She meant to sign a totally different document. It may be that this document contains provisions corresponding with what she wished to sign, because the two documents were cross-codicils by two sisters. But, as a matter of fact, the deceased in signing her name to this codicil never intended to do that at all, but intended to put her signature to another document; and unless some authority can be produced to me to shew that in such a case the document she did not intend to sign is to be treated as the one she did intend to sign, I do not mean to support it. In my opinion this codicil cannot stand.” [my italics]

I find myself in sympathy with these two judgments. The thinking in the italicised passages in particular resembles my own although each judge managed to express himself with an enviable brevity which I have conspicuously not achieved. Sir J. Hannen was influenced by the fact that Ms Hunt did not know and approve any part of the contents of the paper she signed as her will and would not have signed if she had and Sir Gorrell Barnes would not be persuaded to treat the document Ms Meyer did not intend to sign as the one that she did. In Re Hunt consideration was given to the possibility of correcting the sister’s name or omitting provisions so, whilst the powers of the court did not yet extend to rectification of the type introduced by section 20, the sort of approaches for which the appellant in this case has contended were considered and I am not persuaded that the development of the law since the two decisions has removed the force of the decisions. I draw support from them for my own provisional conclusion.

What therefore of the overseas authorities?

The natural place to begin is with Guardian Trust and Executors Company of New Zealand Ltd v Inwood [1946] NZLR 614. The situation was very similar to that in this case, with sisters (Jane and Maude Remington) executing each other’s wills by mistake. Apart from the names, the wills were in identical terms. The New Zealand Court of Appeal was considering an action for probate in solemn form of the will which Jane had signed. The relevant part of the dispositive clause left assets to “my trustee” on trust for various purposes and then “to pay the income thereof to my sister Jane Remington….during her life”.

The submission made by the defendants who were entitled under the intestacy was that the document should not be admitted to probate because “it was not executed animo testandi – that is, that the testatrix did not intend to sign this document and that this document was never intended by her or by any one to be her will” (p 623 of the report). Fair J, delivering the judgment of the court, said:

“This would appear, upon careful examination, to be a very technical basis for its rejection, and, upon an exact appreciation of the true facts, to lack substance. True, the physical document was not the paper that the testatrix intended to sign, but it was a paper that contained everything that she wished included in the paper she intended to sign except the Christian names of her sister. She adopted it believing that it expressed her intentions in every respect. It does in most, and can be read as carrying out her intentions. It appoints the executor she intended to appoint in the exact terms she intended to appoint it. That in itself if it stood alone, would be enough, apart from this formal objection, to entitle it to probate: Mortimer on Probate Law and Practice, 2^nd Ed 205,246. It also disposes of the residue after the life interest in the exact terms which the other will contains. The life interest is in correct terms except for the Christian name. There is no doubt that she intended the document to which she put her signature to operate as her will.

If she had intended to sign the document in the original typewriting, and she had, by mistake, been given a carbon copy, she would have been executing a paper physically different from that which she intended to sign, but if it had contained a duplicate carbon copy it appears unarguable that document in carbon would be invalid on that ground. The present will seems to us to differ from such copy only in degree and not in substance. No doubt the circumstances of the recital with the wrong Christian name would call for explanation as the preliminary headings of the will in Whyte v Pollock. But the fact that the paper put before the testatrix was different from that which she thought she was signing should not, we think, prevent that part of the document which she wished and believed, and which was, in fact, included, being her testamentary act. The testatrix did really know and approve of the effective provisions contained in it: Parker v Felgate and Perera v Perera.”

The court went on to consider the English authority against that point of view, including Re Hunt and Re Meyer. Re Hunt and the earlier similar decision of In the Goods of FS (1850) 14 Jur 402, it put to one side as perhaps to be explained “on the ground that they were decided before more recent decisions above referred to on the power of the Court to correct errors in the language of wills”. Having referred to Re Meyer, the court then observed:

“Decisions by Judges of such eminence are entitled to the greatest respect. But these were founded upon ex parte applications, which were not fully argued and none of which appear to have been matters of substantial importance to the applicants, as appears [sic] from passages in the judgments that effect would probably be given to the wishes of the various testatrixes by those succeeding on an intestacy. In such circumstances, one can understand that the first impression which might well be formed in a case of this kind would be that the signing of a document drawn as somebody else’s will could not be taken as the expression of the intention of a testator, and that the whole document must be rejected. But it would seem that that view is not so much in accordance with the real position, and the principles of the law as the conclusion that the document does express, as it was intended to, the real intention of the testatrix except of the omission of the two words ‘Maude Lucy,’ and the substitution for them of the word ‘Jane’. Moreover, the rule is now well established that words or clauses that have been introduced without the knowledge and approval of the testator may be rejected, and the remaining portion of the will alone admitted to probate: Morrell v Morrell, Fulton v Andrew and Mortimer on Probate Law and Practice, 2^nd Ed 86,87. This had not been clearly established when In re FS and In re Hunt were decided. In such circumstances, this Court feels that, as the language and terms of the will in Re Meyer do not appear to have been the same as those of this will, that decision does not preclude the Court from acting upon the opinion it has come to as to its validity and effectiveness. It therefore holds that the paper executed by the deceased was duly executed, is effective as her will, and should be admitted to probate. Probate will accordingly be granted with the omission of the word ‘Jane’ from the will.”

The reasoning of the New Zealand court is, of course, worthy of careful study in its own right. But the bases upon which that court departed from the English authorities do not, I am afraid, hold water. Whilst the applications in Re Hunt and Re Meyerwere ex parte, I doubt that that had the effect that the New Zealand court thought. Granted, the fact that there was no opposition should have made it easier, not harder, to succeed in the application. However I doubt very much that that would have been seen by the applicant’s lawyers as discharging them from preparing the application fully. It would inevitably have been recognised that even if there was no active opposition from another interested party, the court itself would be astute to ensure that the order sought was a proper one and that argument would therefore have to be carefully prepared so as to be able to meet any difficulties which the judge felt with regard to it. Furthermore, it is unsafe to assume that the outcome of the motion was of no real moment to the applicants because of the particular circumstances of the cases. Counsel was instructed to appear on the motion in both cases and, indeed, in Re Meyer, both King’s Counsel and a junior appeared. Whilst in Re Meyer it appears that the testatrix’s wishes might still have been destined to be carried out notwithstanding the failure of the motion, I can find nothing in the judgment in Re Hunt to suggest that that would be the case there and the judge’s regret as to his inability to help does not provide confidence that it was going to be so. As to the suggestion that a lack in the powers of the court was behind the decision in Re Hunt, this is tenuous. Sir J. Hannen was in fact invited to treat the name in the will as a misdescription and to correct it and did not dismiss this proposal on the basis that he had no such power but on the basis that it would be contrary to the truth of the case.

The Guardian Trust case has had fluctuating fortunes in New Zealand.

The Full Court of the New Zealand Supreme Court distinguished it in In re Foster (deceased) [1956] NZLR 44. A husband and wife had made wills in similar terms. Each signed the other’s will by mistake. On the husband’s death, probate was sought of the document he had signed omitting words which referred to the wife.

The court saw the Foster case as much more intractable than the Guardian Trust case. In the latter, it had been possible to achieve the exact expression of the testatrix’s intentions by simply omitting the word “Jane”. In Foster the dispositive clause said “I give devise bequeath and appoint all the estate both real and personal of whatsoever kind and wheresoever situate of which I shall be possessed to which I shall be entitled or over which I shall have any disposing power at the time of my death unto my husband the said William Foster absolutely”. Striking out the words “unto my husband the said William Foster” would not cure the problem, merely make the disposing clause in the will nugatory. The court saw the ratio of the Guardian Trust case as being the finding of fact that the document propounded was the testamentary instrument of the deceased. They did not consider that the same could be said in the case before them. The questions they posed were “Did the testator intend the document to which he put his signature to operate as his will? Did he really know and approve of the effective provisions contained in it?”. They noted that the testator had read the will he signed and that there was a strong presumption that when a will had been read over to or by a capable testator before execution, he knew and approved all its contents. But this had only occurred because each spouse read over both wills before they were executed and the court therefore took the view that:

“….the testator did not read over this document as his will, but read it over as his wife’s will. He executed the document by mistake. The one dispositive intention he had, namely, to give his estate to his wife was not included; in its place there was a disposition to ‘my husband William Foster’ which is a palpable absurdity. No striking out of the words will make this provision tractable and in accordance with the proved intention of the testator. The evidence tendered fails to establish to this Court those factors which were accepted as being proved to the satisfaction of the Court of Appeal in Guardian Trust”.

The court refused to grant probate and leave it to a court of construction to construe the document because it could not be satisfied that the testator knew and approved a document which disclosed a provision which did not carry out his intention.

When Holland J in the High Court in Christchurch in McConagle v Starkey [1997] 3 NZLR 635 came to consider these two earlier decisions in another mixed up wills case, it is possibly not surprising that he found difficulty in reconciling them and in following the grounds for distinguishing between the facts of the two. Procedural considerations as to appeals in New Zealand were such that he took the view that his duty was to follow the Court of Appeal decision, that is to say the decision in the Guardian Trust case. He thought that the Full Court in Foster had defined the ratio of that case too narrowly. He restated the ratio as:

“Where a document has been duly executed in accordance with the Wills Act 1837 (UK) and the testator intends to execute a document as his last will, and the document actually executed can with the deletion of a word or words give true effect to the testator’s testamentary intentions then probate may be granted of such document with appropriate deletions.”

Holland J was satisfied that, in his case, deletion of words left the will in a shape which enabled the testator’s intentions to be achieved and he granted probate on that basis.

Meanwhile, the Guardian Trust case had been influential in mixed up will cases elsewhere. In the British Columbia Supreme Court in In re Brander Estate [1952] 4 DLR 688, Wilson J relied on it, not adding any reasoning of his own, to admit a will to probate with alterations which in fact went further than mere deletions from the will and included substituting the name of the wife in each case. The Alberta District Court followed the Brander decision in Re Knott Estate (1959) 27 WWR 688 and, as will signing errors crept slowly across Canada in an easterly direction, the Saskatchewan Surrogate Court followed both decisions in Re Bohachewski Estate (1967) 60 WWR 635.

In In the Estate of McDermid [1994] Can LII 4950, the Saskatchewan Court of Queen’s Bench took a fresh look at things without apparently having reference to these earlier decisions. The judge concluded that the requirements of section 7 of the relevant Wills Act (which replicate our present section 9) had been fully complied with except that each of the deceased and his wife had inadvertently executed the will intended for the signature of the other. In contrast to the position in this country, there was in Saskatchewan a provision (section 35 of their Wills Act) which enabled the court to order that a will was fully effective notwithstanding that it had not been executed in compliance with all the formal requirements of the Act if satisfied that it embodied the testamentary intentions of the deceased. This applied and probate was granted.

We have been shown an extract from a Canadian textbook (Feeney’s Canadian Law of Wills; 2011 update) which says that the traditional approach would not permit a will to be admitted to probate where the testator mistakenly executes the wrong paper even if it contains the testamentary dispositions that were intended and all that is required to make the will make sense is to substitute a few words. There is reference in it to “a trend in the western provinces that would contradict this proposition” and Re Brander and Re Bohachewski are cited.

Australia had encountered the problem before the Guardian Trust case. In In re Petchell deceased (1945) 46 WALR 62, Wolff J refused probate. Section 9 of the 1837 Wills Act applied. The judge did not consider that the execution by the husband of the wife’s will complied with the requirement that the testator should execute the document as his will. Dealing with transposed wills in the Supreme Court in South Australia in In the Estate of Blakeley deceased (1983) SASR 473, White J was, however, prepared to admit to probate the husband’s will which had been inadvertently signed by the wife as if the wife’s signature were expunged from it and the husband’s signature appeared in lieu. He commented:

“The courts of England and Australia have refused to recognise unexecuted wills, while the courts of Canada have performed mental gymnastics to do what they can to recognise them. The general rule hitherto has been that ‘when a testator executes a will prepared for another, the document is not his will, even if some of the provisions therein were intended by the testator’: Williams and Mortimer, Executors, Adminstration and Probate (1970)”

He was able get round the problem using the new power in section 12(2) of their Wills Act which, in the manner of the Canadian provision in McDermid, permitted a will to be effective notwithstanding that it did not comply with all the formalities if the court was satisfied that there could be no reasonable doubt that the deceased intended the document to constitute his will.

A similar result was achieved by a similar route in the later South Australian decision of In the Estate of Hennekam (deceased) [2009] SASC 188.

The testator was not so fortunate in the Supreme Court of Queensland in Re Goward [1996] QSC 247. This was not a case of mirror wills erroneously signed by the wrong party. The wills in question differed in substance. Queensland had a provision permitting a testamentary instrument to be admitted to probate if it was executed in “substantial compliance” with the formalities but it was held that recourse could not be had to those provisions unless there was some signature associated with the document in question evidencing the deceased’s adoption of it. The document intended to be the deceased’s last will and testament was not signed by her and it could not be admitted to probate.

In Re Vautier [2000] JLR 351, the Jersey Royal Court (Probate Division) solved the problem (spouses signing each other’s reciprocal will by mistake) by rectifying the will that the deceased wife had signed to overcome the errors caused by the transposition of the wills. The decision contains the following interesting resume of the court’s understanding of the jurisprudence from England (prior to the Administration of Justice Act 1982), Canada and New Zealand:

“22.

To summarise, the common law of England recognised a power in the court to delete words from a will which were included by mistake but did not allow for power in the court to rectify by altering or adding to the wording of the will. The law of New Zealand has recognised a similar power to delete. It has not yet recognised a power to rectify by other alterations although, in the case of McConagle v Starkey, the court indicated that it was supportive of such an approach. In Canada the courts have exercised a power to rectify a will by altering the wording but the initial decision could be said to be based on a misunderstanding of Guardian Trust.”

The court set out why it considered that it had power “to rectify a will whether by deleting or altering (e.g. by substituting or adding) words” despite the absence of any statutory power so to do, the reasons being in brief:

It is of note that the Jersey court proceeded on the basis that the will signed by the wife, being a document which bore her signature which was duly attested as required by law, complied with the formalities for a will (see §14). The decision neither sets out, nor analyses, precisely what the Jersey formalities were and therefore sheds no light on the area which I perceive to be that of greatest difficulty in this case, that is section 9 of the Wills Act.

The appellant also relies on a decision of the Court of Appeals in New York, In the matter of Harvey Snide, deceased (1981) 52 NY 2d 193 (transposed wills of husband and wife). The question was whether the will signed by the husband could be admitted to probate. Against that, it was argued that the husband lacked the required testamentary intent because he never intended to execute the document he actually signed. The matter was finally resolved in favour of probate by the Court of Appeals of New York (4 judges in favour and 3 dissenting). The majority were more attracted to the decisions of Re Brander and Guardian Trust than to the American cases in which probate had been denied in such circumstances. They were much influenced by the fact that the two instruments were “reciprocal elements of a unified testamentary plan”, “executed with statutory formality….at a contemporaneous execution ceremony”. They said:

“….we decline the formalistic view that [testamentary] intent attaches irrevocably to the document prepared, rather than the testamentary scheme it reflects. Certainly, had a carbon copy been substituted for the ribbon copy the testator intended to sign, it could not be seriously contended that the testator’s intent should be frustrated…..Here the situation is similar. Although Harvey mistakenly signed the will prepared for his wife, it is significant that the dispositive provisions in both wills, except for the names, were identical.”

The dissenting opinion began with the view that “[t]his is indeed an instance of the old adage that hard cases make bad law” and that departing from consistent precedent in the United States and English courts would prove troublesome in future. It was pointed out that any statute of wills operates frequently to frustrate the identifiable dispositive intentions of the deceased but the prescribed formalities must normally nonetheless be observed. Judge Jones, giving the opinion, said:

“We are not concerned on admission to probate with the substantive content of the will; our attention must be focused on the paper writing itself. As to that, there can be no doubt whatsoever that Harvey Snide did not intend as his will the only document that he signed on August 13, 1970. ”

Levelling a charge of “lack of rigorous judicial reasoning” at the majority view, he said:

“….why should the result be any different where, although the two wills are markedly different in content, it is equally clear that there has been an erroneous contemporaneous cross-signing by the two would-be testators, or where the scrivener has prepared several drafts for a single client and it is established beyond all doubt that the wrong draft has been mistakenly signed. Nor need imagination stop there. ”

The final case in this foreign tour is that of Henriques v Giles [2009] ZASCA 64 in the South African High Court. Again, the crossed wills were those of a husband and wife. Although the South African Wills Act contained a provision permitting a document to be accepted as a will in certain circumstances notwithstanding a failure to comply with all the formalities, it was conceded that that did not apply. The question was therefore whether the will signed by the husband could be rectified. Rectification was opposed on the basis that as the will did not comply with the requirements for formal validity set out in the relevant Wills Act, it could not be rectified.

The extent of the South African courts’ power to rectify can be ascertained from the following two paragraphs in the judgment:

“15.

South Africa has no legislation on the rectification of wills and the ambit of our courts’ powers to rectify mistakes in a will has been the subject of considerable judicial disagreement. While there has never been any dispute in regard to the court’s power to correct clerical errors or an erroneous description of a beneficiary or a benefit in a will, or to delete form a will words or provisions included in it by mistake, there were conflicting decisions concerning the court’s power to rectify a will by inserting words or provisions which have been omitted in error or by substituting the correct words or provisions for incorrect ones which have been mistakenly included in a will. It is now generally accepted that the South African courts do have this latter power.

16.

Rectification is an equitable remedy, the purpose being to give effect to the true intention of the relevant parties or of the testator or testatrix concerned. A court will rectify a will where, due to a mistake, be it on the part of the testator or testatrix or on the part of the drafter, the will does not correctly reflect their testamentary intention. The applicant for rectification must establish that (a) the alleged discrepancy between expression and intention was due to a mistake; and (b) what the testator or testatrix really meant to provide. The onus, which must be satisfied on the balance of probabilities, is on the party seeking rectification.”

In considering the court’s decision, it is important to recognise that whilst the formalities required by the South African Wills Act bear some resemblance to those in section 9 of the English Wills Act, they are far from identical to them. In particular, there is no requirement such as that in section 9(b) that “it appears that the testator intended by his signature to give effect to the will”.

The court permitted rectification. Refusing it would, it considered, “be to sacrifice testamentary intention to excessive and needless formalism”. Permitting it would reflect the testator’s intention without doing violence to the principles governing rectification of documents or defeating any of the functions of testamentary formalities. The court considered that its conclusion would accord with international jurisprudence and remarked that the English provision in section 20 of the Administration of Justice Act “has been criticised as being unduly restrictive”. The decision contains little detailed analysis directed specifically to the formalities provisions in the South African Wills Act and, obviously, no discussion about the provisions of our section 9.

Whilst I have found a review of these international authorities illuminating, nothing in them causes me to change my provisional view.

Firstly, they have to be approached with caution given the differing provisions of the law in the various jurisdictions.

Secondly, what comes over clearly from them is that judges have generally not found the problem of transposed wills easy and that there is no unanimity about how it should be approached. The camp which favours a facilitative approach includes McConagle v Starkey, the three Canadian cases in the 1950s and 1960s, Vautier in Jersey, Snidein New York (although there is a persuasive dissent in that case) and Henriques v Giles in South Africa; much of this camphas been influenced by the Guardian Trust case. This line of jurisprudence focuses on the facts that the document signed contained essentially what the testator intended and that he signed it thinking it was his will. The testator’s intentions have been achieved in other cases by reliance of legislation of the “substantial compliance” type (McDermid, Blakeley, and Hennekam). As we have no such legislation, these cases are not really of assistance. A more conservative/traditional approach appears in Foster and Petchell and may also underlie the decision in Goward where even substantial compliance provisions did not help the testator. In this group of authorities, what is influential is that however near the contents of the document the testator signed may be to his own intended will, the document he signed was not the one he meant to sign at all and cannot therefore be his will. I find myself more in sympathy with this way of thinking, and with that expressed by way of dissent in Snide, than with the more liberal approach and it fits better, in my view, with our domestic provisions on formal validity.

Accordingly, I remain of the view that the formalities in section 9 were not satisfied in this case. As Proudman J found, the will is therefore not valid. For the reasons I give above, I do not consider there is any room for rectification in these circumstances. It is not therefore necessary to determine the issue that has arisen as to whether what happened here was the consequence of a clerical error within the meaning of section 20 of the Administration of Justice Act 1982 and I do not propose to say anything further about that.

I would dismiss the appeal.

Iagree.

I agree with the judgment of Black LJ.

In paragraph 26 of her judgment, Black LJ summarises the four alternative submissions made on behalf of Mr Marley. Two key questions are raised by the first and fourth submissions which can be expressed as (1) whether the issue of the validity of the document as a will has to be determined before it can be rectified or (2) whether the issue of rectification of the document can be determined first and then the issue of its validity determined on the basis of the document as rectified.

I do not see how the power under s.20 of the Administration of Justice Act 1982 to rectify can be exercised unless the document before the court is the will of the testator. The section refers to “a will being so expressed that it fails to carry out the testator’s intentions...”

Thus the first question to be determined is whether the document put forward by Mr Marley is the will of the testator. That can only be determined by considering whether it complies with the formalities of s.9.

The difficulty for Mr Marley, it seems to me, is s.9 (b), as that provision requires it to be established that Mr Rawlings “intended by his signature to give effect to the will”.

As a power to rectify a will by the omission of words the testator did not know and approve existed at common law, it has never been necessary that the testator intended by his signature to give effect to everything contained in the document as his will. Nonetheless, for s.9(b) to be satisfied, the document must be a document that, without rectification, the testator intended as his will.

In Corbett v Newey [1998] Ch 57, Morritt LJ identified at page 68 the policy of the Wills Act 1837:

The question that must be asked therefore is whether the document Mr Rawlings signed was intended by him to operate as his will or in the words of Scott LJ in Wood v Smith did he sign “to authenticate the written document in question as” his will.

In my view, the answer to the question is that Mr Rawlings did not authenticate the document as his will and he did not intend it to operate as his will. He may well have intended to give effect to a will, but that is insufficient. He must have intended to give effect to the document he signed as his will, not simply as a will. Had he been asked, he would, as Proudman J observed at paragraph 21, said that it was not his will, but his wife’s. He would have been correct for the will that he signed was that of his wife, not his will.

My view accords with the decisions in Hunt and Meyer. The reasoning in each was the same. The testatrix in each case did not intend to sign the document in question as her will and would not have signed it had either known what it was. Nothing material to the provisions of the Wills Act 1837 has changed since the decisions of those distinguished judges.

I agree with Black LJ’s analysis of the extensive number of overseas authorities to which we were referred; they are useful in showing how difficult the judges have found the issue and the ways in which the issue has been approached. However my decision rests upon the clear provisions of s.9(b).

In the circumstances therefore as the will is not one that meets the requirements of s. 9(b) of the Wills Act 1837, the question of rectification does not arise. I therefore do not wish to express a view on whether the powers under s.20 of the Administration of Justice Act 1982 are as narrowly circumscribed as has been indicated in the first instance authorities to which Proudman J referred or as Proudman J considered them to be at paragraph 29 of her judgment. The scope of s.20 must await a decision where the issue arises. I would simply observe that there are powerful arguments for the section to be given a wide and generous scope.

This is a conclusion I have reached with great regret, but Parliament made very limited changes to the law in 1982 and it would not be right for a court to go beyond what Parliament then decided.