Estoppel

Estoppel is one of those doctrines that law students often find slippery and practitioners frequently under-appreciate. It sits at the intersection of contract, property, equity, and evidence, and it has grown organically over centuries rather than being designed. The result is a doctrine – or really, a family of doctrines – that is powerful, flexible, and occasionally incoherent.

1. What is estoppel, really?

In the equitable estoppels, the central idea is that a party may be prevented from resiling from an assurance, promise, or shared assumption where another has relied on it and it would be unconscionable to permit resiling. The older, more technical estoppels – estoppel by record and estoppel by deed – are better understood as rules about finality, conclusiveness, or evidential consistency, rather than as applications of a single unconscionability principle. English law has deliberately resisted a fully unified theory.

The word itself comes from the Old French estoupail (a bung or stopper), and you’ll still see older judgments talking about a party being “estopped” from asserting something. The idea is of being plugged or blocked.

Coke famously described estoppel as occurring “because a man’s own act or acceptance stoppeth or closeth up his mouth to allege or plead the truth” (Co. Litt. 352a). That captures something important: estoppel is often about preventing inconsistency, not about creating rights from nothing – though, as we’ll see, the modern law has moved well beyond this.

The main species

English law recognises several distinct types of estoppel:

Type	Essence
Estoppel by record (res judicata / issue estoppel)	Cannot relitigate what a court has decided; modern litigation categories include cause-of-action estoppel, issue estoppel, and abuse of process
Estoppel by deed	Bound by factual recitals in your own deed
Estoppel by representation	Bound by a representation of existing fact on which another relied
Promissory estoppel	Bound by a promise not to enforce strict legal rights
Proprietary estoppel	An equity may arise from reliance on an assurance about rights in identified property; the court then decides how to satisfy that equity
Estoppel by convention	Bound by a shared assumption underlying dealings
Estoppel by acquiescence	Often treated as related to, rather than wholly distinct from, the other equitable estoppels, arising where a party stands by while another acts on a mistaken belief

Whether these are really one doctrine or many is itself a long-running debate – more on that below.

2. Historical development

The common law roots

Estoppel by record and estoppel by deed are the oldest forms, rooted in medieval pleading. They were technical and narrow: you were bound by what the record showed, or by what you had put under seal. Coke’s treatment in the early seventeenth century is the classic source.

Estoppel by representation emerged later, crystallising in the nineteenth century. The foundational case is:

Pickard v Sears (1837) 6 Ad & E 469 – Lord Denman CJ laid down that “where one by his words or conduct wilfully causes another to believe the existence of a certain state of things, and induces him to act on that belief … the former is concluded from averring against the latter a different state of things.”

This was refined in Freeman v Cooke (1848) 2 Exch 654, where Parke B clarified that “wilfully” did not require actual intent – it was enough that a reasonable person would take the representation as true and act on it.

A crucial limit emerged in Jorden v Money (1854) 5 HLC 185: estoppel by representation only applies to representations of existing fact, not to promises about future conduct. Mrs Jorden had said she would never enforce a bond against Mr Money; she changed her mind; the House of Lords held she was not estopped because that was a promise, not a representation of fact. This case shaped the law for over a century and created the gap that promissory estoppel would later fill.

The equity side: proprietary estoppel

Meanwhile, equity was developing its own doctrine. The early cases were about expenditure on land under a mistaken belief:

• Dillwyn v Llewelyn (1862) 4 De G F & J 517 – a father informally gave his son land, and the son built a house on it. Lord Westbury granted relief that amounted to perfecting the imperfect gift. The case is cited as an early proprietary estoppel authority rather than as a clean statement of fee-simple entitlement.
• Ramsden v Dyson (1866) LR 1 HL 129 – important for two strands: Lord Cranworth’s standing-by/mistake analysis (in the majority) and Lord Kingsdown’s broader dissent on encouraged expectations. Later authority has often treated Lord Kingsdown’s formulation as particularly influential in the development of proprietary estoppel.
• Willmott v Barber (1880) 15 Ch D 96 – Fry J’s famous “five probanda” tried to systematise proprietary estoppel into a structured test (mistake by claimant, expenditure on faith of it, knowledge by defendant of his own rights, knowledge of claimant’s mistake, encouragement). In the “imperfect gift”-type case these ossified into a straitjacket, but modern courts now treat the probanda as useful evidential signposts rather than discarded law.

Promissory estoppel: Denning’s reformulation.

The gap left by Jorden v Money was filled – controversially – by Lord Denning in:

Central London Property Trust Ltd v High Trees House Ltd [1947] KB 130 – During WWII, a landlord agreed to halve the rent on a block of flats because tenants were hard to find. After the war, the landlord’s receiver claimed the full rent for the post-war period. Denning J held that the landlord could recover full rent going forward, but would have been estopped from claiming the wartime arrears because of the promise to accept reduced rent. The promise was intended to be binding, had been acted on, and it would be inequitable to go back on it.

High Trees drew on Hughes v Metropolitan Railway Co (1877) 2 App Cas 439, where the House of Lords had held that negotiations which led a tenant to believe forfeiture would not be enforced suspended the running of a repair notice. Denning essentially excavated this older principle and repackaged it.

The doctrine was then refined and limited in cases like in cases like Combe v Combe [1951] 2 KB 215 (Denning himself holding promissory estoppel is “a shield, not a sword” – it does not create new causes of action) and Tool Metal Manufacturing Co Ltd v Tungsten Electric Co Ltd [1955] 1 WLR 761 (establishing that the estoppel generally only suspends, rather than extinguishes, rights, subject to reasonable notice).

You can view key promissory estoppel cases here.

Proprietary estoppel modernised

Fry J’s five probanda were eventually loosened. The key shift came in:

Taylors Fashions Ltd v Liverpool Victoria Trustees Co Ltd [1982] QB 133 – Oliver J held that the five probanda were not necessary in every case. The real question was whether it would be unconscionable for the owner to insist on strict legal rights, given what had happened. This broad, unconscionability-based approach has dominated ever since.

3. The key modern cases

Proprietary estoppel

This is where estoppel has seen the most vibrant development, particularly through the Supreme Court.

The three-element test (from Oliver J in Taylors Fashions and repeatedly confirmed):

1. A representation or assurance made to the claimant
2. Reliance on it by the claimant
3. Detriment to the claimant in consequence of that reliance

All three are evaluated in the round, with unconscionability as the unifying thread.

Gillett v Holt [2001] Ch 210 – Robert Walker LJ emphasised that the elements “cannot be treated as subdivided into watertight compartments” and that the fundamental principle is “that equity is concerned to prevent unconscionable conduct.” A farm worker who had been repeatedly assured over decades that he would inherit the farm, and who had organised his life around that, was granted relief when the promises were broken.

Jennings v Rice [2002] EWCA Civ 159 – the central question on remedy: what is the appropriate response to the unconscionability? Robert Walker LJ held the court should consider proportionality between the expectation and the detriment. The claimant does not necessarily get what he was promised; he gets what is necessary to do equity.

Then came two significant House of Lords decisions in quick succession that appeared to pull in different directions:

Cobbe v Yeoman’s Row Management Ltd [2008] UKHL 55 – A property developer spent significant time and money obtaining planning permission on the strength of an oral agreement in principle with the owner, who then demanded renegotiated terms. The House of Lords denied proprietary estoppel. Lord Scott took a notably restrictive view: in a commercial context between experienced parties who knew their agreement was not legally binding, there was no basis for estoppel. The claimant got a quantum meruit instead. Lord Scott went further, suggesting proprietary estoppel should not be used to circumvent the formality requirements of section 2 of the Law of Property (Miscellaneous Provisions) Act 1989.

Thorner v Major [2009] UKHL 18 – David worked unpaid on his cousin Peter’s farm for nearly 30 years, with Peter giving hints (handing him a bonus notice saying “for my death duties”) rather than explicit promises. The House of Lords allowed the estoppel and granted him the farm. Lord Walker and Lord Neuberger emphasised the familial and agricultural context, where oblique assurances were the norm. Assurances need only be “clear enough” in context.

The tension between Cobbe and Thorner is essentially about context: commercial parties dealing at arm’s length are expected to nail things down in writing; family members making life choices are not. Lord Neuberger explicitly narrowed Cobbe in Thorner, saying Lord Scott’s observations on formality were not necessary to the decision.

The most recent major decision is:

Guest v Guest [2022] UKSC 27 – A son worked on the family farm for decades on the understanding he would inherit a substantial share. Relations broke down; he left; the parents changed their wills. The Supreme Court (3-2) held that the court first identifies the unconscionability, and there is then an assumption (not a presumption) that the simplest way to remedy it is to enforce the promised expectation. The court must still test proportionality, consider detriment as a cross-check, and discount where appropriate for matters such as accelerated receipt. Lord Briggs for the majority framed unconscionability as the touchstone throughout. The minority (Lord Leggatt) would have started with the detriment (essentially a reliance-based measure). This disagreement reflects a deep theoretical split about what proprietary estoppel is for: protecting expectations, or compensating reliance? It has not been resolved.

Promissory estoppel

The doctrine has been much less active in modern English law – perhaps because consideration doctrine has been relaxed (see Williams v Roffey Bros & Nicholls (Contractors) Ltd [1991] 1 QB 1, finding “practical benefit” sufficient consideration), reducing the pressure on promissory estoppel.

Still, key points:

• Collier v P & MJ Wright (Holdings) Ltd [2007] EWCA Civ 1329 – the Court of Appeal did not overrule Foakes v Beer (1884) 9 App Cas 605. It held that, on the facts, the debtor had a real prospect of establishing promissory estoppel — enough to set aside a statutory demand. Its broader implication (that part payment plus reliance may make it inequitable to claim the balance) remains controversial and has not been confirmed at higher authority.
• Baird Textile Holdings Ltd v Marks & Spencer plc [2001] EWCA Civ 274 – confirmed that promissory estoppel cannot found a cause of action in English law.

Estoppel by convention

Amalgamated Investment & Property Co Ltd v Texas Commerce International Bank Ltd [1982] QB 84 – where parties have acted on a shared assumption (here, that a guarantee covered a particular loan), neither can resile if it would be unjust. This is a mutual-assumption doctrine, not a representation doctrine.

Republic of India v India Steamship Co Ltd (No 2) (“The Indian Endurance”) [1998] AC 878 – Lord Steyn restated the doctrine: the assumption must be shared, communicated (expressly or by conduct), and it must be unjust to go back on it.

Tinkler v HMRC [2021] UKSC 39 – important modern restatement. HMRC sent an enquiry notice to the taxpayer’s old address; the taxpayer’s agent acted as though it had been validly served. The Supreme Court held that estoppel by convention requires a shared assumption that is objectively manifested between the parties – something must “cross the line” between them. The assumption may be one of fact or law, but separate internal assumptions are not enough; the party raising the estoppel must show the other party shared, or at least acquiesced in, the assumption in a way that crossed the line between them.

Estoppel by representation

Still governed substantially by the 19th-century principles, but note:

• The representation must be clear and unambiguous (Low v Bouverie [1891] 3 Ch 82).
• It must be of existing fact (Jorden v Money), though the boundary between fact, law, and mixed fact and law has softened since Kleinwort Benson Ltd v Lincoln City Council [1999] 2 AC 349 reduced the force of older “law/fact” distinctions in restitution. Where the relevant point is really a shared assumption of law, estoppel by convention is usually the cleaner analysis.

Issue estoppel / res judicata / abuse of process

This is a distinct stream, important in litigation:

• Henderson v Henderson (1843) 3 Hare 100 – parties must bring their whole case; cannot litigate piecemeal.
• Arnold v National Westminster Bank plc [1991] 2 AC 93 – issue estoppel can sometimes be relaxed where new material emerges that could not have been discovered with reasonable diligence.
• Johnson v Gore Wood & Co [2002] 2 AC 1 – Lord Bingham reframed the Henderson rule as a broad abuse-of-process doctrine requiring a merits-based judgment, not a rigid bar.

4. Areas of Difficulty and Points of Contention

This is the long-running theoretical question. Lord Denning famously thought there should be a unified doctrine. In Amalgamated Investment, he said the different estoppels “have been moulded into a single doctrine.”

The House of Lords has firmly rejected unification, most clearly in Republic of India v India Steamship Co Ltd [1998] AC 878, where Lord Steyn said that to treat all estoppels as aspects of a single overarching principle “is a beguiling theory, but it is contrary to authority.”

Australia took a very different path. In Waltons Stores (Interstate) Ltd v Maher (1988) 164 CLR 387, the High Court of Australia essentially merged promissory and proprietary estoppel into a single unified equitable estoppel that can found a cause of action. This is one of the starkest divergences between English and Australian common law.

Combe v Combe says promissory estoppel cannot found a cause of action in English law. Proprietary estoppel plainly can. Why the difference?

The conventional answer is that proprietary estoppel is about property rights (which equity has always protected substantively), while promissory estoppel is about promises (where the doctrine of consideration draws the line between enforceable and unenforceable). But this is not entirely satisfying, and the Australian merger shows it is not the only possible answer.

The problem becomes acute where an estoppel arises in a quasi-commercial context that doesn’t involve land — Cobbe is the leading attempt to police this boundary, but the line remains porous.

This is one of several live issues in proprietary estoppel, alongside the clarity required of the assurance, the limits of the doctrine in commercial contexts, and its interaction with statutory formalities.

• The expectation approach: give the claimant what they were promised (subject to proportionality). This is what Guest v Guest (majority) endorses.
• The reliance approach: compensate the detriment suffered. This is Lord Leggatt’s dissent in Guest and is sometimes said to be the true historical basis of the doctrine.

The distinction matters enormously. A claimant promised a £3 million farm who worked for £50,000 less than market wages over 20 years gets very different outcomes on each approach. Guest v Guest treats expectation as the natural starting point (subject to proportionality and cross-checks), but the argument is not over – it is a deep conceptual disagreement about what estoppel is trying to do. The remedy may ultimately be expectation, reliance, or some intermediate award, depending on what is proportionate to cure the unconscionability.

Critics (including some judges) have worried that “unconscionability” is doing too much work in modern estoppel – that it risks becoming a formula for “palm-tree justice” (a pejorative label for discretionary decision-making untethered from principle).

The counter-view is that unconscionability is a genuinely operative principle that demands attention to context: it asks whether, given all the circumstances, it would be unjust for the defendant to resile. This is particularly important where assurances are informal, as they often are in family contexts.

Section 2 of the Law of Property (Miscellaneous Provisions) Act 1989 requires contracts for the sale of land to be in writing. Can estoppel circumvent this?

• Yaxley v Gotts [2000] Ch 162 said yes, at least where the claim could also be framed as a constructive trust.
• Cobbe raised doubts about this (Lord Scott).
• Thorner stepped back from Lord Scott’s view, treating it as non-essential dicta.
• The position is fact-sensitive. Yaxley succeeded partly because the facts also supported a constructive trust, expressly preserved by section 2(5). Cobbe shows the opposite: the court will not dress up an incomplete commercial land bargain as proprietary estoppel merely to evade section 2, particularly where sophisticated parties knew no binding contract existed.

How clear must it be? Thorner accepted oblique hints in context. But the representation must still be capable of being understood as an assurance about a specific property (or sufficiently identifiable property). Thorner v Major itself discussed what happens if the boundaries of the farm change over time – the assurance can attach to the property as it is when the matter comes to be resolved.

Tinkler tightened the doctrine by requiring something to cross the line between parties. But it remains genuinely useful in commercial contexts where parties operate for years on a shared understanding of a document or arrangement. The current live question is how assertive its use can be – can it effectively rewrite clear contract terms? The tendency is to say no: estoppel by convention will not be allowed to contradict clear express agreement where doing so would undermine the contractual bargain.

5. Modern developments and continuing tensions

The post-Guest landscape for proprietary estoppel

Guest v Guest has stabilised the remedial approach but not resolved the conceptual divide. Lower courts are still working out:

• How to value expectation where the promised asset was never going to pass immediately (e.g., promises of inheritance). Guest suggests that acceleration of the interest may require a discount.
• How to handle cases where the relationship has broken down irretrievably – sometimes a clean-break money payment will be preferred to a property transfer.
• The interaction with family provision claims under the Inheritance (Provision for Family and Dependants) Act 1975.

Estoppel and commercial dealings

There is a clear modern trend of courts being cautious about estoppel in commercial contexts, particularly where sophisticated parties have lawyers and could have made binding agreements. Cobbe remains influential here even after Thorner.

The Australian divergence

Australia’s unified equitable estoppel (post-Waltons Stores, Commonwealth v Verwayen (1990) 170 CLR 394) continues to be a live point of comparison. English textbooks increasingly acknowledge that the Australian approach is coherent and workable, even if English law has chosen a more fragmented path.

Contractual estoppel

A relatively recent development is the recognition of contractual estoppel – where parties contractually agree on a state of affairs (e.g., “the Claimant acknowledges it has not relied on any representation”), they can be bound by that even if it is not factually true. Peekay Intermark Ltd v Australia and New Zealand Banking Group Ltd [2006] EWCA Civ 386 and Springwell Navigation Corp v JP Morgan Chase Bank [2010] EWCA Civ 1221 are the leading cases. This is controversial – it arguably isn’t really estoppel at all, but a form of contractual term, and it sits in some tension with the Misrepresentation Act 1967 and UCTA 1977. There is ongoing debate about how far it should be controlled.

A key qualification is now First Tower Trustees Ltd v CDS (Superstores International) Ltd [2018] EWCA Civ 1396: contractual estoppel cannot be used to sidestep section 3 of the Misrepresentation Act 1967 where the clause is, in substance, a non-reliance or exclusion clause. Reasonableness is assessed by reference to UCTA principles.

Estoppel in public law

In public law, estoppel has largely been displaced by the doctrine of legitimate expectation (R v North and East Devon Health Authority, ex parte Coughlan [2001] QB 213), particularly where the issue concerns administrative powers or planning control. R (Reprotech (Pebsham) Ltd) v East Sussex County Council [2002] UKHL 8 treats private-law estoppel as only an analogy in that context, because public-law remedies must take account of the wider public interest. Public bodies can, however, still be affected by ordinary private-law estoppels when acting in a private-law capacity, subject to statutory powers and duties. Practitioners crossing from private law need to be alive to this shift.

6. A practical framework

If you are approaching an estoppel issue cold, these are the questions to ask, in order:

1. Which type of estoppel is in play? Don’t conflate them. The requirements differ.
2. Is there a clear representation, assurance, promise, or shared assumption? What is its content?
3. Was it directed to (or reasonably understood by) the claimant? Was it communicated?
4. Did the claimant actually rely on it? Reliance is often presumed where detriment is substantial, but it is a factual question.
5. Is there detriment? Not necessarily financial – forgoing opportunities, life choices, uncompensated work can all count.
6. Is the context commercial or familial? This affects how strict the courts will be.
7. Would it be unconscionable to resile? This is the unifying touchstone in the equitable estoppels.
8. What is the appropriate remedy? Consider expectation, reliance, and proportionality. In proprietary estoppel, remember Jennings v Rice and Guest v Guest.
9. Are there statutory or formality obstacles? Section 2 of the 1989 Act, wills formalities, etc.
10. Is estoppel being used as a shield or a sword? This constrains which doctrines are available.

Final thought:

Estoppel rewards careful thinking. The temptation is to reach for it whenever something feels unfair, but the doctrine is disciplined: it requires identifiable assurances or assumptions, real reliance, real detriment, and – crucially – a context where it would genuinely be unjust for the other party to go back on what they said or did. The modern trend, especially since Guest v Guest, is toward principled flexibility: unconscionability remains the touchstone, but proportionality, context, and the expectation/reliance balance all matter.