Til death do us part: predatory marriage and probate


The concept of ‘predatory marriage’ may not mean a great deal to English lawyers, and certainly not probate practitioners. But it is a phenomenon which can have serious and permanent testamentary effects. With an aging population, and legislation in apparent need of reform, the harm caused by predatory marriage seems set only to increase.
The term, which may be more familiar to those with experience of Canadian law, refers to an individual (henceforth ‘A’), whose mental capacity is in doubt. If A marries, the effect of the Matrimonial Causes Act 1973, section 12(c), is that the marriage will not be deemed void, but rather voidable – as in contract law.
Section 16 of the 1973 Act then provides that any decree by the Court that A’s marriage is null and void is prospective, not retrospective:
‘A decree of nullity granted after 31st July 1971 in respect of a voidable marriage shall operate to annul the marriage only as respects any time after the decree has been made absolute, and the marriage shall, notwithstanding the decree, be treated as if it had existed up to that time.’

This can have unsettling implications for the law of probate. As most practitioners will be aware, a valid marriage ordinarily revokes any previous wills made by that individual (Wills Act 1837, section 18(1)). So, even if A’s marriage is later declared a nullity, its effect will still have been to revoke any previous wills A may have made. A will need to reinstate those wills, or else die intestate.
But what if A has since lost capacity to make a will? Or, worst of all, what if A dies before the marriage can be nullified?
This is the essential problem on which this article is based. Its ingredients are these:

  1. A produces a will, the benefit of which is received primarily by B (‘the Will’);
  2. A then marries C (‘the Marriage’), the effect of which is to revoke the Will and, for these purposes, disinherit B (Wills Act 1837, section 18);
  3. A dies, and because of the Marriage, dies intestate. B has lost the benefit of the Will but cannot unravel the Marriage (Matrimonial Causes Act 1973, sections 12(c) and 16).

This scenario can all too easily be procured deliberately: a predatory marriage. Say A’s capacity is uncertain, and C knows this. C can connive to marry A, creating a marriage which is voidable and not void thanks to the 1973 Act. The Marriage revokes the Will, and, if A dies soon after, C takes the benefit of the intestacy rules as spouse. B is disinherited and without recourse – even though the Will was made when A did have capacity.
In light of the dangers of predatory marriages, this article investigates (i) the origins of the 1973 Act and its predecessor, the Nullity of Marriage Act 1971; (ii) how courts have subsequently grappled with the problem of voidable marriages and capacity; (iii) the policy rationale for the 1973 Act; (iv) potential avenues for reform; and (v) pointers for practitioners.

The law prior to 1971

The problem of void and voidable marriages is not a new one.1 For instance, before the Marriage Act 1835, even a marriage between brother and sister may merely have been voidable, until annulled by order of an ecclesiastical court. Originally, ‘the Ecclesiastical Courts were in the habit of annulling these marriages, even after the death of the parties, after the death of both, or of one only’ (Ray v Sherwood (1836) 1 Curt. 193). However, by the nineteenth century, the civil courts preferred to restrict the availability of annulments to the lifetime of the parties, raising a sinister twist on the scenario proposed above in which A and C are siblings.
Parliament then legislated for mental deficiency as a ‘new ground’ for the nullity of a marriage in the Matrimonial Causes Act 1937, section 7:

1 See Paul J Goda’s The Historical Evolution of the Concepts of Void and Voidable Marriages 7 J. Fam. L. 297, tracing the distinction back to 1085. Professor Goda argues the term ‘voidable’ is first used in the context of marriages in a divorce case of 1598.

‘(1) In addition to any other grounds on which a marriage is by law void or voidable, a marriage shall be voidable on the ground—


(b) that either party to the marriage was at the time of the marriage of unsound mind or a mental Defective within the meaning of the Mental Deficiency Acts, 1913 to 1927, or subject to recurrent fits of insanity or epilepsy;

The Mental Deficiency Acts provided for a range of conditions (so-called idiots, imbeciles, feeble-minded persons, and moral imbeciles). Although the language of the Acts has aged poorly, in many ways the capacity question was handled with more nuance then than under the modern law. Judges were prepared to treat pronounced cases of incapacity as questions of voidness, thereby taking them outside the voidability limitations of the 1937 Act.
An example is Re Park [1954] P. 112, whose facts mirror the scenario set out above. A made the Will in 1948, principally for the benefit of B. The Marriage to C took place in 1949. C claimed as widow for the benefit of A’s intestacy, the Marriage having revoked the Will. B sought to rely on the Will, and to do that, he claimed a declaration that the Marriage was a
nullity due to A’s mental capacity.

Bizarrely, A had purported to make a second will on the same day as the Marriage. He had been married at 11 AM and the second will had been made at 2.45 PM. In a previous action, this second will had been found to be invalid.
The question for the Court of Appeal was whether ‘at the time of the [marriage] ceremony the deceased was mentally capable of understanding the nature of the contract of marriage so that the marriage could be regarded as valid’ (per Birkett LJ at 129). A finding of fact had been made by the jury at first instance that A did so understand. The Court of Appeal did not interfere with this finding, despite the inconsistency with the invalid second will made four hours later.
Therefore, prior to 1971, a marriage could be found to be void for lack of capacity. A void marriage would leave a prior will intact, and beneficiaries could apply after the testator’s death to have that marriage declared null.

The 1971 transformation

This position was changed by the Nullity of Marriage Act 1971, the effect of which was, for these purposes, identical to the 1973 Act. The two Acts set out the situations in which a marriage could be void, including parties who were prohibitively close relations, unlawful intermarriage, bigamy, and polygamy (1973 Act, section 11).
The Acts then set out vitiating factors that would make a marriage voidable (1973 Act, section 12). The voidable factors include duress, unsoundness of mind, and lack of mental capacity.
Unlike the 1937 Act, both void and voidable factors are given exhaustively. Not only that, but section 16 of the 1973 Act, quoted above, provided that a declaration of the nullity of a voidable marriage had no retrospective effect.
The enormity of this change was seen five years later in Re Roberts [1978] 1 WLR 653. A made the Will for the benefit of B in 1973. A then married C in 1974, allegedly in a state of senile dementia. B sought to rely on the Will to oppose C’s application for a grant of letters of administration.
As B conceded, any retrospective declaration of nullity was defeated by section 16 of the 1973 Act. The ingenious argument put forward was therefore that section 18 of the Wills Act 1837 only revoked prior wills where the marriage was valid. This was rejected. Parliament had expressly legislated for marriages to be voidable for lack of capacity. A voidable marriage subsisted until it was annulled and ‘where a marriage is voidable the persons who are concerned with the grounds which make it voidable are the parties to the marriage and nobody else’ per Walton J at 656. B had no remedy.

Subsequent case-law

In Re Davey [1981] 1 W.L.R. 164, a not dissimilar scenario came before the Court of Protection. A, aged 92, married C, an employee at his nursing home. The danger of a repeat of Re Roberts was invoked successfully as a ground for urgently making a statutory will on A’s behalf. This proved to be prescient; A died a week after the statutory will was made. That Re Roberts should artificially impose such urgency (indeed mandating that the statutory will be executed without notice to C) is clearly unsatisfactory.

The City of Westminster v IC [2008] EWCA Civ 198, at least at first instance ([2007] EWHC 3096 (Fam)), contains the clearest judicial assault on Re Roberts that the authors have been able to find.
In this case, the Family Court was required to consider an individual referred to as IC, who had been married in Bangladesh. There were various conflict of laws issues that need not concern us here. What is relevant is that IC was a demonstrably vulnerable individual lacking capacity. In such circumstances, Roderic Wood J clearly felt uneasy about a finding in English law that the Marriage was merely voidable up and until a declaration was made pursuant to the 1973 Act. The Judge’s creative solution was as follows (paragraph 104):

    1. The 1973 Act made voidable a marriage where inter alia one party’s consent was vitiated by ‘unsoundness of mind’ (section 12(1)(c));
    2. Unsoundness of mind covered a spectrum of conditions, from individuals whose conditions were subtle to individuals with obvious unsoundness, such as IC;
    3. It was only the former that was contemplated by the draughtsman of section 12(1)(c);
    4. Where an individual ‘demonstrably had no mental capacity’ then it would be ‘repugnant to public policy’ to recognise a marriage as valid (and so presumably it fell outside the procedure of the 1973 Act). The Marriage was therefore declared invalid.

The authors have some sympathy with the Judge’s dilemma, which is comparable to the treatment of incapacity in Re Park and the old Matrimonial Causes Act 1937. The difference of course is that the 1973 Act had provided exhaustive categories of voidness and voidability, and expressly denied judges the power to make retrospective declarations in voidable cases. The Court of Appeal therefore concluded that Roderic Wood J had no power to make the declaration he did. But even here, the judges’ reasoning differed subtly.
Thorpe LJ described Re Roberts as ‘clear and binding’ (paragraph 25). Parliament had expressly provided for this situation under the 1973 Act.
Wall LJ appears to have ascribed to Re Roberts a more limited view. Though binding, it was ‘simply not relevant’:
‘What it decides is that a voidable marriage which has not been the subject of a decree of nullity operates to revoke the wills made by the parties to it. In my judgment, that does not

affect the question of whether the marriage in the instant case is entitled to recognition in English law.’ (paragraph 96).
Hallett LJ somewhat unhelpfully agreed with both. In the end, the mischief of Re Roberts was avoided by the Court refusing nonetheless to recognise the Bangladeshi marriage under conflicts of law rules, such recognition being against public policy.
Interestingly, Thorpe LJ at paragraph 3 made the following pronouncement:

‘The role of marriage in the life of one so handicapped is inconceivable in our society. Furthermore as a matter of law marriage is precluded. IC lacks the fundamental capacity to marry.’

As a statement of principle, this is difficult to reject. However, as a statement of domestic law, its accuracy is doubtful. Assuming that an individual could be found to officiate over the ceremony, IC may have been entirely able to marry in England and Wales. Not only that, but the marriage would have remained valid until the point of its annulment and, if IC died before it could be annulled, it would have endured indefinitely. This article has focused on the injustice to B, as a disinherited beneficiary, but the harm to A personally should not be overlooked.
There has been little substantive case-law since (aside a perhaps surprising citation on a capacity point in an insolvency case (Fehily v Atkinson [2016] EWHC 3069 (Ch)). It is hoped the area is ripe for reform.

Policy rationale

The 1973 Act tries to balance a historic tension: between certainty and morality. On the one hand, marriages define wide-ranging and important legal relationships. There would be unsettling consequences if they could be easily and retrospectively set aside. On the other, our law and society prize consent and free will, insisting that those marriages be formed in a particular manner and according to various conditions, mandating their annulment if not.
This is exactly the same policy tension as can be seen as far back as Harris v Hicks (1692) 2 Salk. 548. The question in that case was incest, not incapacity, but in all other respects is identical. The Ecclesiastical Court of Coventry moved to annul an incestuous marriage, for obvious moral reasons. The Court of King’s Bench prohibited the annulment (though allowed

punishment for incest to proceed) in order to protect the issue of the marriage. Voidability was, as it is now, the uneasy compromise struck between morality and certainty.2
Another part of the problem appears to derive from the imperfect analogy with contract law. Voidability is a contractual import, but the law of marriage has failed to carry the analogy to its fullest extent. When a voidable contract is set aside (‘rescinded’), it is broadly speaking rescinded ab initio. It is ‘treated in law as never having come into existence’ (Johnson v Agnew [1980] A.C. 367 at 393 per Lord Wilberforce), unlike section 16 of the 1973 Act, which more closely reflects the contractual remedy of termination.
Another inconsistency is with the criminal law. Forced marriage is a criminal offence under section 121 of the Anti-social Behaviour, Crime and Policing Act 2014. C might in theory be imprisoned for up to seven years for forcing A into a predatory marriage, though that marriage would be otherwise lawful and capable of having significant, even irrevocable, testamentary consequences.


It was said by Karminski J that ‘the essence of a valid marriage is the consent of the contracting parties’ (H v H [1954] P. 258 at 265 – 266). The law as it stands does not always enact that principle, and the result can be abuse.
A sad example of this is the case of Mrs Blass, which was put before Parliament in November 2018 by the MP for Leeds North East, Fabian Hamilton. Mrs Blass was A: a widow suffering from dementia. She formed a close relationship with C, a younger individual who, according to Mrs Blass’ family (B), came to exercise a controlling influence over Mrs Blass. The two were married in secret. This secret marriage unmade a previous will and Mrs Blass’ family found themselves disinherited. Apparently Mrs Blass may not even have been aware that she had re-married.
As a result, Fabian Hamilton MP introduced the Marriage and Civil Partnership (Consent) Bill. The Bill, which has since withered on the parliamentary vine, proposed the following:

2 See also the judgment of Singleton LJ in Re Park: ‘consent is absolutely requisite to matrimony’ and yet: ‘The annulment of other contracts would only affect property; but [the annulment of a marriage] would do that, and more – it would tell upon the happiness, character, and peace of the parties. The appalling character of these consequences is well calculated to impress the courts with the solemn duty of requiring a clear case for the application of the general principle to this delicate and important contract’ ([1954] P. 112 at 125 – 126).

  1. Marriage would no longer revoke prior wills (reversing section 18 of the Wills Act 1837 and mirroring the position in some Canadian provinces, including Alberta’s Wills and Succession Act 2010, section 23(2));
  2. Superior training for marriage registrars to spot vulnerable individuals;
  3. Marriage questionnaires to alert registrars to the need for possible capacity assessments;
  4. Public notices of intention to marry, to enable families to intervene.

Provision (3) raises interesting questions about the capacity test for marriage, and the extent to which it might differ from that of testamentary capacity. In Re Park at first instance, Karminski J had found that the mental capacity required to marry was less than that required to make a will. This distinction was rejected by the Court of Appeal; ‘there is no sliding scale of soundness of mind’ as Hodson LJ put it at 135.
A different conclusion was reached by the Ontario Supreme Court in Banton v Banton [1998] OJ No 3528, where A, a widower with Alzheimer’s disease, married C, a waitress fifty years his junior. A’s marriage to C was found to be valid, though a contemporaneous will was not.
This is troubling. Logic would dictate that, so long as section 18 of the 1837 Act remains, a marriage is a testamentary disposition, and so the capacity to marry should at least align with that of making a will, if not exceed it. As Mostyn J observed recently in Mundell v Name 1 [2019] EWCOP 50 at paragraph 7:
‘It would be surprising if the degree of mental capacity that is needed to execute a will is in fact less than the degree of mental capacity that is needed validly to contract a marriage’.

This aligns with the position taken by the Court of Protection in EJ v SD [2017] EWCOP 32 at paragraph 10, and it is surely to be preferred:
the fact that a second marriage revokes the will is information that a person should be able to understand, retain, use and weigh to have capacity to marry.’
An alternate solution would be to update the Court’s armoury to deal with the problem. Section 55 of the Family Law Act 1986 sets out the declarations a court may make about a marriage. As the Court of Appeal observed in The City of Westminster v IC at paragraph 26:

‘What is significantly absent is a sub-paragraph permitting a declaration that the marriage was at its inception an invalid marriage.’

Perhaps it is time that judges were given that power. The pernicious effect of predatory marriage stands only to worsen as the population ages, and a single sentence from the draughtsman’s pen may go some way to alleviating that harm.

Practical pointers

Practitioners will be familiar with questions of capacity in relation to wills. With elderly or vulnerable clients, it is suggested that similar attention should be paid to capacity to marry.
In the case of a marriage that has already taken place, A’s capacity should be considered. Swift action may be needed to challenge that marriage, or make a new will to counteract section 18 of the 1837 Act. This could involve having a deputy appointed to act on A’s behalf, as in Re Davey.
Practitioners should be aware that a beneficiary may not know if a testator has been married (B may be happily unaware of A’s marriage to C, until they find themselves disinherited). In rare cases, A themselves may not realise that they have married C – as appears to have been the unhappy case of Mrs Blass.
There is also the danger of a marriage having taken place abroad, in a jurisdiction with more relaxed capacity rules, which was the precise issue in The City of Westminster v ICIn less extreme cases, foreign marriages are routinely recognised in English courts. It may be difficult to establish if such a marriage took place, and if so, whether it was valid according to the law of that jurisdiction. Practitioners should be aware also of the possibility of bigamy,
which, in the authors’ experience, is surprisingly common.

In the case of a prospective marriage, an individual’s capacity may have to be confirmed in advance. This is what happened in Mundell v Name 1. Mostyn J, at paragraph 6, also made the following suggestion:
‘However, it is open to him, between now and [the marriage], to execute a codicil to his will which provides that the will shall survive his marriage and be effective thereafter. I would suggest that it is in his interests, although it is not part of the decision I have to make, that he

should execute a codicil to his will to that effect in the next couple of days, if I permit the marriage to proceed.’

Mostyn J’s suggestion takes effect through section 18(3) of the 1837 Act, which makes an exception to ordinary section 18(1) revocation where, when a will is made, ‘the testator was expecting to be married to a particular person and […] intended that the will should not be revoked by the marriage’.
Mostyn J’s suggestion is an imperfect solution. By the marriage, the testator has granted their spouse a right to receive reasonable financial provision from their estate under the Inheritance (Provision for Family and Dependants) Act 1975, regardless of the terms of their will. This would risk the unfortunate consequence of the will beneficiaries being locked into high risk and costly litigation following the Deceased’s death in order to defend their loved one’s testamentary wishes and their beneficial interest.
If it appears that a party already lacks capacity, a deputy will naturally consider the merits of a statutory will. In the circumstances, it might not be appropriate for an individual to make a will at all, but that would appear to be a rare case (Re Jones [2014] EWCOP 59). If a statutory will is required, again it may be necessary to act quickly – perhaps even by way of urgent application, as in Re Davey – if the would-be testator’s health is declining.
Unfortunately, as is the premise of this article, once A has died, options are limited. B may wish to consider a White v Jones [1995] UKHL 5 style claim against A’s legal advisers, but such an action is likely to be speculative. B may have to throw themselves on the mercy of the intestacy rules.

This article has been prepared for the Northern Contentious Probate Group seminar on 8 January 2020 by James McKean, barrister at New Square Chambers, and Hollie Richardson and Andrew Bishop of Shoosmiths LLP.