Terms of art: why resulting, implied and constructive trusts?
This article looks at the historical development of the well-known exception, in the latter part of subsection (5), to the general requirement of Section 2(1) of the Law of Property (Miscellaneous Provisions) Act 1989 ("the 1989 Act") that "a contract for the sale or other disposition of an interest in land can only be made in writing" in one document incorporating all the express terms, that:
"... nothing in this section affects the creation or operation of resulting, implied or constructive trusts."
The article has been written using the Justis database as a primary research tool. Justis encompasses all the primary sources of citable authority (the official ICLR reports and the English reports as well as statutes) and is, from my experience, one of the most cost-effective and easy-to-use online resources. It is referenced in a few instances to enable better understanding of the process of the research in question.
The expressions "resulting trust", "implied trust" and "constructive trust" are now commonly understood to be used as terms of art.Although textbook writers have long complained about what the 1981 Edition of Hanbury & Maudsley's Modern Equity I had at university called the "overlap in classification", academic writers and the courts have rather tended to accept, in principle at least, that resulting, implied and constructive trusts must in some way be conceptually distinct. Justis searches have been used to track the usage of these phrases (and common variants) over the time they have gained currency and to examine whether that view is justified. While I do not propose to set out all the searches carried out using Justis in the course of researching this article, where I do, the terms entered appear in bold type between square brackets.This is because quotation marks are a search tool, indicating a search for an exact phrase. When a search has been conducted the results can be put in date order by clicking the top of the column marked "Year": one click will put the results in reverse chronological order; a second brings up the oldest first.
Looking at the origins of Section 2 of the 1989 Act is neither a whimsical enquiry nor a purely academic exercise; for the practising lawyer having the means to place a provision in its proper context can lead to an argument's succeeding when it might have failed and to a client's securing a remedy which might otherwise have been refused. The ambit of section 2 is, moreover, still a matter of debate; see, for example, McLaughlin v. Duffill ([2010] 1 Ch 1; [2008] EWCA Civ 1627), a case to which I shall return, or Cobbe v. Yeoman’s Row Management ([2008] UKHL 55; [2008] 1 WLR 1752), in which Lord Scott asked how, when "Section 2 of the 1989 Act declares to be void any agreement for the acquisition of an interest in land that does not comply with the requisite formalities" and subsection (5) only makes an exception for resulting, implied or constructive trusts, proprietary estoppel could escape its strictures. While he found it unnecessary to answer the question, he observed before leaving the topic:
"The proposition that an owner of land can be estopped from asserting that an agreement is void for want of compliance with the requirements of section 2 is, in my opinion, unacceptable. The assertion is no more than the statute provides. Equity can surely not contradict the statute."
The origins of Section 2(1) are found in the Statute of Frauds 1677 (or, to give it its original name, An Act for prevention of Frauds and Perjuryes). Section 4 is most often seen as its most direct precursor and, so far as is material, was in the following terms:
"And bee it further enacted by the authoritie aforesaid That from and after the said fower and twentyeth day of June noe Action shall be brought ... or to charge any person upon any or upon any Contract or Sale of Lands Tenements or Hereditaments or any Interest in or concerning them ... unlesse the Agreement upon which such Action shall be brought or some Memorandum or Note thereof shall be in Writeing and signed by the partie to be charged therewith or some other person thereunto by him lawfully authorized."
Section 4 did not relate exclusively to contracts for sale or disposition of an interest in land and nor was it the only provision in the Statute of Frauds affecting such matters. Section 4 also imposed similar formalities upon
- special promises by an Executor or Administrator to "answere damage out of his own Estate"
- special promises to answer for the "debt default or miscarriages of another person"
- agreements made upon consideration of marriage
- agreements not to be performed within a year of being made,
while the primary provisions affecting contracts for sale or disposition of an interest in land contained in other sections were that:
- parol leases and interests of freehold took effect only as estates at will (Section 1), other than some leases of less than 3 years (Section 2, an exception which remains today under s.2(5) of the 1989 Act)
- leases or estates of freehold, copyhold or terms of years certain could only be granted or surrendered by deed or "Writeing signed" (Section 3)
- devises of lands had to be in writing and signed and attested by three or four witnesses or they were "utterly void and of none effect" (Section 5) but, once executed, were only revocable in certain prescribed ways (Section 6),
- declarations or creations of trusts or confidences of land (Section 7) and any assignment thereof (Section 9) had to be in proved by writing signed by the party able to declare such trust or by his will or were void
The precursor to the exception in subsection (5) of 1989 Act is Section 8 of the Statute of Frauds:
"Provided alwayes That where any Conveyance shall bee made of any Lands or Tenements by which a Trust or Confidence shall or may arise or result by the Implication or Construction of Law or bee transferred or extinguished by an act or operation of Law then and in every such Case such Trust or Confidence shall be of the like force and effect as the same would have beene if this Statute had not beene made. Any thing herein before contained to the contrary notwithstanding."
Although the phrase "resulting, implied or constructive trusts" does not appear (that exact formulation was not used until it appeared in s.15 of the Law of Property (Amendment) Act 1924) the verb "result" and nounal derivatives of "imply" and "construct" do appear in close proximity.
The phrase "by the Implication or Construction of Law" appears to be used as one of general description rather than something referring to terms of art and searching on Justis bears this out. Effective searching involves identifying the key words and phrases that are particularly suited to being the subject of a search: [trust "implication of law"], for example, will yield reports in which both the word "trust" and the phrase "implication of law" appear, while [resulting implied constructive trust*], will identify where "resulting", "implied", "constructive" and any word beginning "trust" (so "trust" or "trusts") have appeared together. Where a hit produces a case from the English Reports and other very old reports on Justis, such as the State Trials, one has to open the PDF and search within it for where the terms occur, otherwise, a searchable version of the report appears when the result is first opened and there is a "terms" tab which enables one to scroll through occurrences of the words for which one has searched (along with court-ready printable PDFs that replicate the bound volumes).
And what is the result of searching around permutations and combinations of the key words and phrases? The first is that although the earliest English Reports go back to 1220, references to resulting, implied and constructive trusts do not begin to appear with any regularity until after 1700. Early appearances show that their usage was not specialised save to distinguish trusts arising by operation of law from express trusts, as one of the earliest references found in Coke's commentary on Chudleigh's case, Dillon v Freine (76 E.R. 261 (1595)), Coke 113b demonstrates:
"Trust estates may either be created expressly .... or they may arise from the evident intention of the parties, in which latter case they are called resulting trusts or trusts by implication";
and see also Lord Grey v. Lady Grey (36 E.R. 742; (1677) 2 Swans 594) ("constructive trust"). That remained the position until "resulting trust" began to gain currency as a description of a trust arising in favour of the donor, or his heirs, in the case of a failed or ineffective disposal of all or part of the beneficial interest in property. See Fitzgerald v. Thomas Lord Fauconberge ((1729) Fitzg 207; 94 E.R. 722):
"No rule is more certain than that if a man makes a conveyance in trust for such persons and such estates as he shall appoint, and he makes no appointment, the resulting trust must be to him and his heirs";
and Burgess v. Wheat (1746-1779) 1 Black W 123
"Where you have passed the estate without consideration, there in modern language an use results, or a trust results; because it is unequitable that a man should have an interest in the estate, when he has paid no consideration for it."
although it continued to be used generically as well. The terms "implied trust" and "constructive trust", however continued to be used interchangeably and in the generic sense simply of not being express trusts well after 1900, see for example, Henry v. Hammond [1913] 2 K.B. 515.
Can anything further be gleaned about the formulation of the wording now found in s.5 of the 1989 Act? While it may be somewhat speculative it would appear that a formulation first found in the headnote to Espinasse v. Lowe (1764) 3 E.R. 223 (HL), 7 Bro PC 345, echoing a passage of argument at 355 –
"Constructive trusts, or trusts resulting by implication of law, are not within the statute of frauds and perjuries; by an express exception in the statute itself. They arise from the apparent nature of the transaction" –
was taken up by Nineteenth Century Parliamentary draughtsmen. In the Transfer of Trust Estates Act 1830 s.15 one finds almost identical wording:
"And be it further enacted, That the several Provisions herein-before contained shall extend to every other Case of a constructive Trust, or Trust arising or resulting by Implication of Law...";
which is reproduced in the Estates Vest in Heirs, etc., of Mortgages Act, 1838:
"...the several Provisions therein-before contained should extend to every Case of a constructive Trust, or Trust arising or resulting by Implication of Law ..."
The specialised meaning of "resulting trust" which had developed appears to have been recognised in the Finance Act 1917 s.35(2), even as the purely generic usage of "implied trust" (now substituted for "[trust] by implication of law") and "constructive trust" continued:
"This section ....shall not apply to any trustee under an implied or constructive trust, except a resulting trust arising on the determination or failure of an express trust."
It was then but a short step to s.15 of the Law of Property (Amendment) Act 1924, which introduced the wording of s.2(5) of the 1989 Act on to the Statute Book. There followed, of course, the Law of Property Act 1925, s.40, which superseded s.4 of the Statute of the Statute of Frauds in relation to contracts for the sale of land. There was no exception of the type now found in s.2(5) of the 1989 Act, however; the exception went into subsection (2) of the much wider s.53:
"(1) Subject to the provision hereinafter contained with respect to the creation of interests in land by parol—
(a) no interest in land can be created or disposed of except by writing signed by the person creating or conveying the same, or by his agent thereunto lawfully authorised in writing, or by will, or by operation of law;
(b) a declaration of trust respecting any land or any interest therein must be manifested and proved by some writing signed by some person who is able to declare such trust or by his will;
(c) a disposition of an equitable interest or trust subsisting at the time of the disposition, must be in writing signed by the person disposing of the same, or by his agent thereunto lawfully authorised in writing or by will."
It is doubtful given the words "or by operation of law", which appear at the end of s.53(1)(a), that s.53(2) was needed at all, but just as, if one looks back to the original wording of s.8 of the Statute of Frauds, and sees that there was no reason for the use of so complex a formulation, other than the draughtsman's desire to guard against unforeseen nuances of whichever single word or phrase he might have chosen to have deployed, every lawyer who has ever tacked the extra "or by any other means whatsoever" on to the end of a draft undertaking or injunction (i.e. all of us) will sympathise with the desire not to leave anything to chance by using too few words.
The dangers of using too many is too often lost on us: by 1926, while noting that "Lewin on Trusts, 12th ed., p. 124n., ... states "The terms Implied Trusts, Trusts by Operation of Law, and Constructive Trusts, appear from the books to be almost synonymous expressions", Astbury J. in In Re Llanover Settled Estates ([1926] Ch. 626) was already trying to discern a distinction. When s.40 of the 1925 Act was replaced by Section 2 of the 1989 Act, the draughtsman borrowed subsection (2) of s.53, just in case he needed to, hence, in part, Lord Scott's existential angst about the potential application of s.2 to estoppel. The other cause was the extent to which constructive trusts have now been used to perfect oral agreements about shares in land, most notably between cohabitees, and in that context s.2(5) appears to take on a logic that is illusory and would never have occurred to the parliamentary draughtsman of the 1920s, who framed the exception. Historical analysis helps us to see that.
I finish with McLaughlin v. Duffill because in that case the Court of Appeal recognised the very different effect of s.2 of the 1989 Act and s.53 of the 1925 Act.