The three certainties

The three certainties are the minimum conceptual conditions for the creation of an express private trust. They are conventionally stated as certainty of intention, certainty of subject matter, and certainty of objects. The formula is associated with Knight v Knight, but the deeper point is not historical taxonomy. A trust is an institution under which the trustee is bound to hold particular property for another or for a permitted purpose. If the court cannot identify a binding intention to create that institution, the property to which the obligation attaches, or the persons for whose benefit the obligation is to be performed, there is no administrable trust.

The topic follows naturally from Week 1. If the trust divides legal title and equitable entitlement, the three certainties ask when that division has occurred. They are not mere technicalities. They mark the boundary between property and hope, between obligation and moral pressure, and between judicial enforcement and private aspiration. English equity has historically been willing to recognise informal, even elliptical, dispositions; but it is not willing to invent a trust where the settlor has not imposed fiduciary obligations, has not identified the property subject to those obligations, or has not provided a workable beneficiary principle.

The Cambridge significance of the topic is twofold. First, it is a foundation topic. Later problems on formalities, constitution, resulting trusts, secret trusts, charitable trusts, powers, purpose trusts, tracing and remedies often turn on an initial question whether a trust was ever created. A weak answer treats the three certainties as a checklist. A strong answer asks what each certainty is doing in the architecture of equitable ownership. Secondly, the topic is unusually rich for essays because the cases expose a tension between respect for settlor autonomy and the demands of legal administration. The courts often say that they construe rather than remake the settlor's words. Yet the difference between Re Adams and Comiskey, or between Re London Wine and Hunter v Moss, shows that classification may depend on institutional assumptions about property, commerce, and judicial workability.

In supervision and in the Tripos, distinguish three questions. The first is whether the language used was intended to create legal obligation rather than expressing motive, confidence, wish, or expectation. The second is whether the property is identified with enough precision that beneficial interests can attach to it. The third is whether the beneficiaries are identified by a test the court can apply. Those questions overlap, but they must not be collapsed. A settlor may clearly intend a trust of uncertain property. A class may be conceptually certain but evidentially hard to prove. A purported trust may fail not because beneficiaries are unknown, but because the settlor has created only a power. Precision in the taxonomy is rewarded because it reveals the remedial consequence: failed express trust, resulting trust, absolute gift, valid power, or valid trust.

The doctrine of the three certainties belongs to the nineteenth-century rationalisation of equitable principle, but its roots are older. The medieval use enabled one person to hold seisin for the benefit of another. Its effectiveness depended upon the Chancellor's willingness to enforce conscience against the legal owner. Once the use developed into the modern trust, equity required some basis on which conscience could be fixed. The trustee must know that he is bound, know what he is bound in respect of, and know for whom he is bound. The modern certainties are therefore an attempt to give institutional form to an equitable jurisdiction originally framed in moral language.

Knight v Knight supplied the canonical formulation. Lord Langdale MR's statement was addressed to testamentary language, a context in which property owners often used precatory expressions: wishes, hopes, recommendations, requests and confidence. In the earlier chancery tradition, such expressions had sometimes been treated as sufficient to impose trusts, especially where a testator gave property to a relative and expressed a wish that others should be provided for. That older approach may be seen as paternalistic and contextual. The court asked whether the donee's conscience ought to be bound in the circumstances. During the nineteenth century, however, the law became more reluctant to convert moral exhortation into proprietary obligation. This shift is visible in Lambe v Eames and Re Adams, where precatory words were not enough.

The change was not merely linguistic. It reflects the general nineteenth-century movement from conscience to administration. Trusts had become major devices for family settlement, commercial holding, investment and testamentary planning. A doctrine based on loose moral implication could unsettle title and frustrate alienability. Certainty of intention protects donees from involuntary fiduciary office and protects third parties by indicating when proprietary equitable interests have arisen. Certainty of subject matter allows the court to identify the fund and shares. Certainty of objects ensures that the trust can be supervised and, in the case of fixed trusts, that beneficial entitlement can be distributed.

The twentieth century then complicated the picture. The old insistence on a complete list of beneficiaries for all non-charitable trusts was relaxed for discretionary trusts in McPhail v Doulton. Lord Wilberforce's approach aligned discretionary trusts with powers by asking whether it can be said of any given person that he is or is not within the class. That decision was driven partly by the practical needs of modern pension and employee benefit trusts, where large fluctuating classes are common. The law therefore moved from strict enumerability towards conceptual administrability.

The result is a doctrine with historical layers. Precatory words and testamentary construction belong to a world of family provision. Subject-matter cases such as Re London Wine and Hunter v Moss respond to commercial and financial assets. Objects cases move from family settlement to pension funds and modern discretionary arrangements. The common thread is not a mechanical formula but the court's concern that equitable obligations must be enforceable without requiring the court to perform a wholly discretionary or legislative function.

The starting point is the tripartite formula: intention, subject matter, objects. It is indispensable, but dangerous if treated as self-executing. Each certainty answers a different institutional question.

Certainty of intention asks whether the alleged settlor intended to impose a trust obligation. The word trust is neither necessary nor conclusive. Equity looks to substance: whether, objectively construed, the language and circumstances show that the recipient is to be bound to hold property for another, rather than merely being morally encouraged or absolutely benefited. A commercial segregation of customer money may create a trust without the word trust being used; conversely, the word trust may be ineffective if the surrounding arrangement is inconsistent with fiduciary holding. Intention is assessed objectively, not by undisclosed subjective hope. In Paul v Constance, repeated statements that money was as much the claimant's as the defendant's, together with their handling of the account, sufficed to show an intention that the claimant should share beneficially. In Re Kayford, a company opening a separate account for customer prepayments was held to have created a trust, because the arrangement objectively manifested an intention to keep the money from the company's general assets.

Testamentary language is especially sensitive. Expressions such as in full confidence, hoping, wishing or desiring may or may not create a trust. The modern tendency is reluctance to impose a trust from precatory language alone. Re Adams is the leading example: a gift to a wife in full confidence that she would do what was right as to children was construed as an absolute gift. Comiskey, by contrast, found a trust because the will, read as a whole, contained a gift over and a direction that the nieces should take after the widow's death. The exam point is that there is no magic word. Ask whether the donee is given beneficial freedom or is made an office-holder under an equitable obligation.

The cases should be learned not as isolated propositions, but as a sequence in which equity refines what it means for a trust to be judicially enforceable.

Knight v Knight is the organising authority. Lord Langdale MR's tripartite formulation remains the standard language of the subject. Its importance lies in identifying the conditions for converting a disposition into an enforceable trust. It does not, however, solve hard cases. Most litigation concerns borderline language or borderline property rather than complete absence of one of the three elements.

Re Adams and Comiskey form a classic pair on intention. In Re Adams, a gift to a widow expressed confidence that she would do what was right by the children. The Court of Appeal refused to treat that confidence as a trust. The widow received beneficially; the children had no equitable proprietary entitlement. In Comiskey, the House of Lords construed the will as imposing obligations because the instrument contained stronger directional language and a gift over to nieces. The lesson is that intention depends on the whole instrument, not on a dictionary of precatory expressions. Cambridge examiners often reward candidates who resist the crude proposition that words of confidence can never create trusts.

Palmer v Simmonds and Re Golay's Will Trusts show the subject-matter requirement in two modes. Palmer fails because the fund itself is uncertain: the bulk of an estate is not an identifiable quantum. Re Golay succeeds because the standard of a reasonable income can be applied by a court. These cases are useful because they show that certainty is not the same as arithmetical precision. Equity does not require mechanical exactitude where ordinary judicial assessment is possible.

Re London Wine is the leading modern authority on tangible fungibles. Customers who had paid for wine could not claim particular bottles because none had been appropriated to them. Without segregation, the subject matter of each trust was uncertain. The case is central to insolvency contexts: recognising trusts over unsegregated assets removes value from the general pool of creditors, so courts demand careful identification.

Hunter v Moss is the counterweight. A declaration of trust of 50 shares out of a larger holding of identical shares in the same company was valid. The Court of Appeal distinguished cases involving chattels. Its reasoning has been criticised, but the decision is entrenched in English law. The practical principle is that identical intangible shares may be treated differently from physical assets requiring segregation.

McPhail v Doulton is the landmark on objects. The House of Lords abandoned the complete list test for discretionary trusts and adopted the any given postulant test. This made large modern discretionary trusts viable. Re Baden (No 2), the sequel, exposed the unresolved question of how much uncertainty the test tolerates. The three judgments remain essential because they separate conceptual uncertainty, evidential uncertainty and administrative difficulty. Together these authorities show that the three certainties are not a rigid Victorian relic. They are adjusted by the courts to the kind of trust, the kind of property, and the kind of obligation in issue.

See practice questions section below.

See further reading section below.