Examining Wills Secret Trusts and fraud
Wills Act is an anti-fraud device in the way to make sure will of the testator is genuine by complying with different formality requirements. Under S9 of Wills Act 1837  , last wishes should be made in writing, signed by the testator whose signature is witnessed by two people. However, there is exception to S9  , namely the secret trust.
Generally, secret trust arises in the situation where a settlor communicates to the intended trustee that the settlor would pass property to him for his holding on trust for the intended beneficiary after death of settlor even if there is no formal mention of the trust in the will. The court would uphold the trust even if it does not satisfy the formality requirement in S9 unless the intended trustee disagrees to be a trustee in the very first place. 
There are two types of secret trust, half secret trust and fully secret trust. In fully secret trust, both the existence of the trust and the terms of the trust are absent in the will while in half secret trust, trusteeship appears in the will but there is no mention of any terms and beneficiary  . There are indeed several justifications for this abnormal operation of secret trust over the centuries, significantly the fraud theory and dehors the will theory.
The fraud theory is based on the maxim: equity will not permit a statute to be used as an instrument of fraud. The operation of fraud theory aims to prevent any denial of trust based on unsatisfactory statutory formalities.  In this way prevents the intended trustee from keeping the property of the settlor for his own good while depriving that of the beneficiaries. This theory gained main support from the case Rouchefoucauld v Boustead  .
Although the theory seems to be a fair justification for the operation of secret trust, it is not without flaw.
Firstly, it would be difficult to attest the evidence brought about by the claimant. Secondly, the theory fails in half secret trust on the fact that trusteeship is identified in the will and there could be no benefit to the trustee even though the trust fails by the operation of resulting trust as suggested by Blackwell v Blackwell  .
Dehors the Will Theory
Dehors the will theory stands by arguing secret trust operates outside the will for it is in fact an inter vivos trust declared by communication of trust to trustee and constituted on settlor’s death. The theory does not treat secret trust as testamentary disposition and thus not to be included in S1 of the Wills Act 1837  applicable to any testamentary disposition. The elements for a valid secret trust are: intention of settlor, communication of the intention between settlor and the intended trustee and the trustee’s acceptance of it and constitution of trust by vesting of property into hands of trustee upon death of settlor  .
There are judicial supports for the theory as in Blackwell v Blackwell  by Viscount Sumner “I do not see how the statute-law relating to the form of a valid will is concerned at all", application of the theory in Cullen v Attornay-General for Ireland  and the speech by Megarry in Re Snowden  . Moreover cases following Blackwell seem to accept the theory as a concrete justification for operation of secret trust, more particularly in Re Young  where a beneficial interest in a secret trust was upheld even though the beneficiary was a witness to the will contrary to the Wills Act 1837  where a witness could not be a beneficiary.
Dehors the will theory is probably commonly recognized as the modern theory of secret trust.  However, the theory is indeed problematic.
Secret Trust under Law of Trust or Law of Probate?
If secret trust is outside the will, then should it follow the law of trust instead of law of probate? Careful study of the characteristics of secret trust below reveals that it should not be so and it should instead be a kind of testamentary disposition.
First, secret trust and the will are closely inter-related. If the will fails, it would not be possible to vest the property in the secret trustee and the secret trust will also fail due to want of subject matter. The interdependency between the two could be seen in the following situations:
(i) Revocability of secret trusts
Wills are revocable at any time before death of testator while trusts are irrevocable once it is created unless the testator reserves the right to do so.  If it is said that secret trusts are dehors the will, they should be irrevocable as that of normal trusts. However, common laws established that they are not.
Secret trusts could be revoked in two different ways. Settlor could revoke the will so that there could be no vesting of property upon potential secret trustee.  Moreover, settlor could revoke the secret trust without changing anything in the will as in the case of Guest v Webb  in which Starke J. expressed the view that as a matter of conscience under equity, a secret trust should not be set up contrary to wishes of the testatrix.
Justification for revocability of secret trust mainly falls on the reason that secret trust is not constituted until death of settlor where the vesting of property into hands of secret trustee happens via the will.
(ii) Witness as secret trustee and secret beneficiary
It is accepted that as secret trustee does not receive any beneficial interest from the will, the secret trustee can be a witness to the will as in Re Ray’s Will Trust  and Re Young  . However, Hayton  argues that a fully secret trust under this situation should fail as a beneficial interest is given to the fully secret trustee under the will.
It is also suggested in Re Young  that as secret beneficiary may not be aware of the existence of a secret trust at the time when witnessing the will, the secret beneficiary could be a witness to the will. However, if the secret beneficiary knows about the existence of the secret trust, there would be a possible conflict of interest  and that would be hard to justify the decision in Re Young  than that of Re Fleetwood  , which it was held that the legacy to beneficiary fails because the beneficiary was a witness to the will.
(iii) Death of Beneficiary
Under a will, when the beneficiary predeceases the testator, the gift lapses. However, in secret trust, if a secret beneficiary dies before the settlor, the gift could still be upheld. The situation is well-reflected by Re Gardner No.2  in which Romer J said that communication of the obligation to hold the property on trust was like a declaration which binds the secret trustee to hold the property upon trust as specified and thus the trust was constituted at that time before death of the testatrix upholding the gift to the secret beneficiary. Although Romer J has a strong argument to support that gift in secret trust was not to be treated as given under the will for secret trust could be created in the case of intestacy, Re Gardner No.2  was not without flaw.
The secret trust in Re Gardner No.2  could not be created at the time of communication and acquiescence of the trust as the subject matter at that time was future property which could not be valid under a trust and thus the time of constitution should fall on the death of testatrix.
Hodge  suggests further requirement to the constitution of a secret trust in that not only the trust is constituted on death of settlor and vesting of trust property to trustee but also needs to be further manifestation of intention of trustee to constitute the trust and acts as trustee.
Moreover, if Re Gardner No.2  was decided rightly, it would mean that the secret trust is constituted upon communication and acquiescence as in Blackwell v Blackwell  . However, this would contradict the revocability of secret trust as discussed earlier which could occur before death of settlor.
Secondly, the normal rules of trust do not accept declaration of immediate trust of future property while secret trust accepts the trust to be binding on property even if they are acquired afterwards. If it is to argue that secret trust could get around the problem by constitution upon death of settlor, it would mean that Re Gardner No.2  should be decided wrongly for there was no trust when the beneficiary died.
Thirdly, if secret trusts are express trusts, the trust over land would require to be evidenced in writing in accordance to S53(1)(b) of the Law of Property Act 1925  , while in Ottaway v Norman  , it was suggested that in the case of fully secret trust, no writing was required although writing was still required under half secret trusts of land as in obiter statement in Re Baillie  .
Fourthly, the argument for secret trusts to be under law of trusts instead of law of probate fails to justify the reason for the need of declaration of trust to occur before the date of the will if under the assumption that the date of constitution of trust is the date of the death of settlor and the will only acts as a trigger to the constitution purpose.
Lastly, it is also suggested that there are two basic characteristics for testamentary disposition: ambulatory and revocability, which mean the valid will walks along until the testator dies and it can be revoked or changed respectively before death of testator.  Clearly, the two characteristics of testamentary disposition are well satisfied by characteristics of secret trust from the discussion above.
Wrong interpretation of “Will"
One of the academics Patricia Critchley  argues the theory interprets meaning of will in a wrong sense and the theory confused “outside the will" with “outside the Wills Act" for secret trust is a kind of testamentary disposition not to “arises outside the will". Critchley  distinguished meaning of will between board and narrow sense. She pointed out that the theory is founded only on the narrow interpretation of will to be a formal document executed by a testator contrary to what the statute means. Critchley’s interpretation of “will" in the broad sense seems to be more reasonable on the fact that “will" should be intention of the way the testator wishes to deal with his own property rather than of a piece of writing on which dehors the will seems to focus on the latter.  Moreover, the narrow interpretation of “will" could not justify the admission of inherent evidence by court and could not explain the importance of acceptance of trust by trustee  . Critchley  also pointed out that S9 is indeed just an evidential provision setting out requirements for evidence to be admitted in order to prove the intention of testator. 
From the above discussion, it should be clear that secret trust is instead a kind of testamentary disposition to be included in S1 of the Wills Act 1837.
It does not operate completely outside the will but the two are instead inter-dependent on each other. The key element would be on constitution of a secret trust on death of settlor and vesting of property into hands of trustee either by terms of will or following rules under intestacy. Although the statement in Re Young  suggests that the gift was taken under the secret trust but not the will, it does not say that secret trust and will are independent to each other. Moreover, dehors the will theory interprets meaning of the will wrongly. Therefore dehors the will indeed fails to justify the informality of secret trust.
Although Re Gardner No.2  seems to establish a strong point for dehors the will theory in the case of intestacy, the case is widely believed to be decided wrongly and should not be considered as conclusive.