Reconciling Contradictory Binding Precedents: Doctrine, Discretion, and the Limits of Stare Decisis

A practising barrister asked to advise on a point of law does not simply want to know what the relevant rule is. She wants to know which rule applies when the authorities pull in opposite directions, and that is a question the textbooks rarely answer with the honesty the problem demands. Contradictory binding precedents are not theoretical curiosities; they are a recurrent feature of a common law system that has accumulated centuries of doctrine without the benefit of systematic codification. When two decisions of the same court, each binding on the court below, point to incompatible conclusions, the lawyer is left holding two loaded weapons and told to choose one. That choice, how it is made, why it is made, and whether the tools deployed to justify it are intellectually honest, is the central concern of this essay. The argument advanced here is that while English law possesses a sophisticated armoury of formal reconciliation mechanisms, those mechanisms ultimately serve as vessels for a covert judicial discretion that the rhetoric of stare decisis conceals rather than constrains. The formal rules are real, but they are better understood as the grammar of judicial reasoning than as its determinant.

The doctrine of stare decisis rests on values that are both principled and pragmatic. Predictability, equality before the law, and the efficient administration of justice all depend on courts following prior decisions rather than reinventing doctrine with each new case. Cross and Harris captured the essential point with characteristic economy: the authority of a precedent derives not merely from the wisdom of its reasoning but from the institutional position of the court that decided it.¹ Dworkin elevated this idea into a more ambitious theory of adjudication: the image of the chain novel, in which each judge must write the next chapter in a way that renders the whole narrative coherent, illustrates why consistency across time is not a bureaucratic preference but a moral requirement of the enterprise.² On this view, a judge who ignores an inconvenient precedent is not merely breaching convention but committing an act of injustice, treating like cases differently without justification.³ Schauer has offered a complementary account, arguing that the binding force of precedent derives from the weight of the past: a rational agent who recognises the systemic value of consistency will follow even a decision she believes was wrongly decided, precisely because the alternative, each court evaluating every question afresh, is worse for litigants and society alike.⁴ Yet the same system generates its own contradiction. Courts are not infallible, legal contexts change, and appellate decisions are reached by panels of judges who do not always reason from the same premises even when they reach the same result. The very process of precedent-formation is thus the origin of precedent-conflict. Stare decisis demands consistency while simultaneously producing the inconsistency it was designed to prevent.

The primary tools for navigating conflicting authorities are the ratio decidendi, the distinguishing process, and the principle of hierarchical deference. None of them is as clean in practice as the doctrinal literature suggests. Cross and Harris defined the ratio of a case as the ruling of law necessary to the decision on the facts as found, a formulation which obscures the extent to which the identification of the ratio is itself an interpretive act.⁵ Judges routinely read precedents narrowly or broadly depending on whether they wish to follow or evade them. Cross acknowledged as much, noting that there is no meta-rule which specifies how narrowly or broadly the facts of a prior decision must be characterised before its ratio can be extracted.⁶ Twining and Miers developed this further, observing that the identification of the ratio is inseparable from the identification of the legally relevant facts, and that lawyers who agree entirely on the facts of a case may still disagree fundamentally about its ratio because they prioritise different features of the factual matrix.⁷ The duty of care litigation from Donoghue v Stevenson⁸ through Hedley Byrne⁹ to Caparo¹⁰ illustrates the point vividly: each successive decision invoked the authority of its predecessors while generating a ratio so different in scope that earlier and later cases are reconcilable only through the most determined interpretive effort.

Distinguishing on the facts is the most commonly deployed reconciliation mechanism and, for that reason, the most susceptible to strategic deployment. When a court holds that a prior decision is distinguishable because the contract in question contained a particular clause, or the claimant occupied a subtly different legal status, it may genuinely be identifying a material difference. It may equally be performing a feat of forensic contortion to reach a preferred result. Cross and Harris were candid that distinguishing can be a device for departing from a precedent without saying so.¹¹ Duxbury observed that the practice gives courts a degree of flexibility that sits uneasily with the determinacy that stare decisis is supposed to deliver.¹² There is an argument insufficiently explored in the literature that the availability of distinguishing as a reconciliation tool may actually increase the total volume of contradictory authority in the long run, because courts that might otherwise invoke the Practice Statement, and thereby conclusively resolve a conflict, instead distinguish their way around an inconvenient precedent, leaving the underlying tension unresolved for future courts to navigate. Distinguishing postpones the reckoning without eliminating it.

Where distinguishing fails because the facts are genuinely indistinguishable, the Court of Appeal is ordinarily bound by its own previous decisions. The exceptions to that rule were catalogued in Young v Bristol Aeroplane Co¹³ and they remain the framework within which contradictory Court of Appeal authorities are navigated. Lord Greene MR identified three circumstances in which the court could decline to follow its own prior decision: where the prior decision conflicts with a subsequent House of Lords ruling, where two of the court’s own decisions are irreconcilable and it must choose between them, and where the prior decision was made per incuriam.¹⁴ The second exception is directly relevant to the problem at hand. When two Court of Appeal decisions genuinely contradict each other, Young permits a later panel to select the one it regards as correct. The formal framing presents this as a duty to prefer the better-reasoned authority, but that framing is somewhat misleading: the court is not applying a rule that determines the outcome but exercising a substantive judgment about legal merit. Tiverton Estates Ltd v Wearwell Ltd illustrated this precisely, with Scarman LJ acknowledging that in choosing between two conflicting Court of Appeal decisions the court was inevitably making fresh law, whatever doctrinal language it employed.¹⁵

The per incuriam doctrine deserves sustained scrutiny because it is simultaneously the most intellectually interesting and the most unprincipled of the Young exceptions. A decision made per incuriam, through carelessness, is one reached in ignorance of a relevant statutory provision or binding authority which, had it been considered, would have led to a different result. The doctrine’s rationale is impeccable: it would be absurd to bind future courts to a decision reached in ignorance of the very materials that governed the issue. Its application is another matter. In Morelle v Wakeling,¹⁶ Sir Raymond Evershed MR held that per incuriam was limited to cases where a decision had been given in ignorance or forgetfulness of some inconsistent statutory provision or of some authority binding on the court concerned, and that it could not be extended to mere failures of reasoning, however egregious.¹⁷ That formulation is principled, but it generates its own difficulty: it means that a Court of Appeal decision which reaches a demonstrably wrong result through flawed analysis remains binding, provided all the relevant authorities were placed before the court. The wrongness of the reasoning is not, on the Morelle approach, a basis for escape.

Williams v Fawcett¹⁸ demonstrated the strain that a rigidly narrow per incuriam doctrine generates. The Court of Appeal there held that a series of its own decisions on the procedural requirements for committal applications were per incuriam because they had consistently misread the relevant rules. Sir John Donaldson MR acknowledged the conceptual awkwardness:¹⁹ a decision reached without the benefit of full argument might nonetheless have been decided the same way had all materials been available, but where the court could be confident that the error was fundamental and the consequences for litigants severe, a more flexible approach was warranted. The later decision in R v Simpson took a different line, with Judge LJ emphasising that Williams was an exceptional case justified by its peculiar procedural context and should not be taken as authorising a general departure from the Morelle framework.²⁰ The result is that Morelle and Williams are themselves in tension, producing the meta-level irony that the doctrine designed to address irreconcilable precedents has generated its own irreconcilable precedents. Marshall, surveying comparative precedent doctrine, concluded that per incuriam represents the common law’s most explicit acknowledgment that formal bindingness and substantive correctness sometimes pull apart, but that the system has no principled mechanism for deciding which should prevail.²¹ Burrows similarly observed that the criteria for per incuriam remain elastic and contested.²² This is one of those areas where intellectual honesty compels the admission that the law is genuinely unclear in a way that no amount of doctrinal analysis will resolve.

Above the Court of Appeal, the position is structurally different. The Practice Statement of 1966²³ freed the House of Lords from its former obligation to follow its own decisions, replacing the self-imposed rigidity of London Street Tramways²⁴ with a discretion to depart ‘when it appears right to do so.’ Lord Gardiner LC’s accompanying statement was notable for its explicit acknowledgement that too rigid adherence to precedent may lead to injustice in a particular case and also unduly restrict the proper development of the law.²⁵ What the Practice Statement created was not a licence for free development of the law but a carefully managed safety valve, and the restraint with which the Lords exercised it is itself revealing. In Herrington v British Railways Board,²⁶ the House departed from the occupiers’ liability rule in Robert Addie & Sons (Collieries) Ltd v Dumbreck,²⁷ not by invoking the Practice Statement expressly but by finding that changed social conditions had fundamentally altered the underlying assumptions of the earlier decision. In Miliangos v George Frank (Textiles) Ltd,²⁸ the House used the Practice Statement to permit judgment in foreign currency, overruling a rule that had become commercially untenable in the wake of the Bretton Woods collapse. In R v Howe,²⁹ the Lords departed from Director of Public Prosecutions for Northern Ireland v Lynch³⁰ on the availability of duress in murder, invoking principle over pragmatism. What is striking about this sample is not merely the restraint in frequency but the consistency of justification: each departure was grounded in changed social circumstances, commercial absurdity, or fundamental principle rather than judicial preference alone.

The judicial reluctance to invoke the Practice Statement too freely reflects an awareness, articulated by Paterson, that the legitimacy of the court depends on its being perceived as constrained by law rather than simply making it.³¹ Raz made the connected point that the authority of law is practically undermined if subjects cannot rely on its stability: a court that overrules its own decisions too readily destroys the very basis on which its authority rests.³² The United Kingdom Supreme Court, which succeeded the Appellate Committee following the Constitutional Reform Act 2005,³³ has continued this tradition of cautious engagement. In Austin v Southwark LBC, Lord Hope confirmed that the Practice Statement remains available to the Supreme Court and that the considerations governing its use are unchanged.³⁴ Yet the institutional transformation wrought by the 2005 Act carries implications for the Practice Statement’s future deployment that have not been adequately theorised. The Law Lords sat as a legislative chamber, their judgments shaped by an awareness of the constitutional proximity of Parliament and the consequent expectation of legislative deference. The Justices of the Supreme Court occupy a building physically and symbolically separate from Westminster, their role publicly framed as constitutional adjudication rather than legislative assistance. Lord Reed’s judgment in Miller reflects precisely this heightened self-consciousness about the court’s constitutional function.³⁵ If the Justices increasingly understand themselves as the guardians of constitutional principle rather than the refiners of private law doctrine, they may prove more willing to use the Practice Statement to resolve persistent precedent conflicts in public and constitutional law, where the systemic stakes of leaving irreconcilable authorities unresolved are arguably higher. The restraint appropriate to a chamber of the legislature is not obviously appropriate to a fully independent constitutional court. Whether that logic will eventually manifest in a more interventionist approach to overruling remains to be seen, but it is a structural pressure the current doctrine does not acknowledge.

The Human Rights Act 1998 introduces a structural complication that the domestic law of precedent was not designed to handle. Section 3 requires legislation to be read compatibly with Convention rights ‘so far as it is possible to do so,’³⁶ while section 6 makes it unlawful for a public authority, including a court, to act incompatibly with those rights.³⁷ The consequence is that a domestic precedent which predates the HRA, or which failed to engage adequately with Convention rights, may conflict directly with a clear line of Strasbourg authority. The question of how that conflict is to be resolved is not a question the Act resolves with any clarity, and the courts have had to work out the answer through a series of decisions that are themselves only partially reconcilable. The HRA does not formally place Strasbourg authority above domestic precedent in the hierarchy of sources; it creates instead a duty of compatible interpretation that can, in effect, render earlier domestic authorities obsolete without formally overruling them. This is precedent conflict resolved by stealth rather than by rule.

Pinnock v Manchester City Council³⁸ is the clearest illustration of the pressure the HRA places on domestic precedent. The Supreme Court there held that where Strasbourg case law, taken as a whole, established a clear and constant line of authority on a Convention right, domestic courts were required to follow it even where this meant departing from a prior domestic decision. Lord Neuberger’s formulation was carefully qualified:³⁹ the duty to follow Strasbourg applied only where the authority was genuinely clear and consistent, not where the ECtHR’s own jurisprudence was internally divided. The structural implication is nonetheless significant. The HRA created a second normative hierarchy alongside the domestic precedent system, and where those hierarchies conflict, the domestic authority yields. Ghaidan v Godin-Mendoza⁴⁰ confirmed that Convention-compatible interpretation can require courts to read statutory provisions in ways that depart significantly from their natural meaning, and implicitly from the interpretive assumptions on which earlier domestic authorities had proceeded. Duke v Reliance Systems Ltd,⁴¹ where Sir John Donaldson MR held that the courts were not entitled to distort statutory meaning to achieve compatibility with a directive, now reads as a product of a pre-HRA world. The HRA has not merely added a mechanism for resolving precedent conflict; it has restructured the underlying hierarchy within which that conflict takes place.

The most important claim in this analysis, and the one that the formal doctrine most conspicuously conceals, is that the choice between competing authorities is never merely a technical selection between two legal rules. It is a substantive policy decision about which version of the law should govern future conduct, and the mechanisms of reconciliation serve primarily to legitimise rather than to determine that decision. Bell observed that legal arguments derive their acceptability not from logical entailment alone but from their coherence with the broader normative framework within which the legal community operates.⁴² That observation carries particular force when the choice between competing precedents has obvious policy weight. The conflict in R v R⁴³ between earlier authority that implied marital immunity in rape and the developing understanding of individual autonomy was not resolved by neutral doctrinal analysis: the House of Lords chose a reading of the criminal law that reflected a particular view of what the rights of married women required. The formal mechanism deployed, that earlier authorities were no longer good law given changed social circumstances, was legally available. But it was the value judgment, not the mechanism, that determined the outcome. The mechanism was selected because it was available and appropriate to the conclusion; the conclusion was not reached because the mechanism demanded it. Finnis was right that legal positivism, at the point of applying general rules to contested cases, consistently underestimates the extent to which moral reasoning enters the analysis.⁴⁴ This matters for democratic legitimacy in a way the reconciliation literature rarely confronts directly. When courts select between contradictory precedents by deploying per incuriam or distinguishing, they are making law. The question of whether unelected judges should make law in contested normative domains is not resolved, but merely obscured, by describing what they are doing as reconciliation. Beatson argued that the boundary between judicial development of the common law and judicial usurpation of the legislative function is inherently unstable and cannot be policed by doctrinal rules alone.⁴⁵ The mechanisms of precedent reconciliation sit precisely at that boundary, and they are the less visible for it.

Lee, writing in the Oxford Journal of Legal Studies, has argued that the persistence of conflict in the common law is not a failure of the system but a feature of it: irreconcilable authorities signal the presence of a genuine legal controversy that no formal resolution can dissolve, and the honest response is to acknowledge rather than conceal the choice being made.⁴⁶ Allan, by contrast, argued in the Law Quarterly Review that the reconciliation mechanisms of precedent, far from constraining discretion, actually licence it, providing judges with a vocabulary of doctrinal respectability that substitutes for the transparent reasoning that rule of law values require.⁴⁷ The tension between these positions has no clean resolution. Lee is right that formal acknowledgment of conflict would at least be honest; Allan is right that honesty without constraint is not obviously an improvement. What the formal analysis reveals is that the reconciliation of contradictory precedents involves three distinct operations, only the first of which is genuinely rule-governed: identifying whether a conflict exists, selecting the applicable mechanism, and exercising judgment about which authority should prevail. Cross and Harris were candid that the third operation cannot be reduced to rules.⁴⁸ Dworkin would say the judge reaches for the principle that best justifies the existing legal materials taken as a whole.⁴⁹ The less idealistic view is that the judge reaches for the outcome she regards as correct, then deploys whichever mechanism most naturally supports it.

The Dobbs comparison, which the literature treats as a cautionary tale about the instability of explicit overruling, actually illuminates something more fundamental about the choice between legal systems.⁵⁰ When the United States Supreme Court overruled Roe v Wade explicitly, the result was not merely legal change but a crisis of institutional legitimacy: the court was seen to have made a political choice, and the formal apparatus of precedent, which had been invoked to justify Roe for fifty years, was equally invoked to demolish it. The English preference for covert adjustment, for resolving precedent conflict through distinguishing, selective per incuriam, and HRA-mediated supersession rather than explicit overruling, can be defended on precisely this ground. A system that adjusts incrementally, that allows the law to evolve in response to changed circumstances without announcing that it is doing so, preserves the appearance of continuity even when the substance has shifted. Fuller argued that the rule of law requires laws to be applied consistently over time, but he also recognised that a legal system incapable of development is not more legitimate for its rigidity.⁵¹ Craig’s distinction between formal and substantive conceptions of the rule of law is directly relevant: a purely formal account demands procedural consistency; a substantive account demands that the law produce just outcomes even when consistency requires modification.⁵² The English approach to precedent conflict, covert in its operation but principled in its justifications, is better understood as a substantive rather than a formal rule of law commitment. The doctrinal mechanisms are the scaffolding, not the building. Acknowledging that is not an admission of failure but an honest account of how a mature legal system handles the irreducible tension between stability and justice. The common law’s deepest claim to legitimacy has never rested on its consistency alone; it has rested on its capacity to be consistently, if quietly, right.

Reference(S):

1. R Cross and JW Harris, Precedent in English Law (4th edn, Clarendon Press 1991) 3. ↩
2. R Dworkin, Law’s Empire (Harvard University Press 1986) 228–238. ↩
3. ibid 229. ↩
4. Frederick Schauer, ‘Precedent’ (1987) 39 Stanford Law Review 571, 595. ↩
5. Cross and Harris (n 1) 72. ↩
6. Sir Rupert Cross, ‘The Ratio Decidendi’ (1959) 22 MLR 75, 77. ↩
7. William Twining and David Miers, How to Do Things with Rules (5th edn, Cambridge University Press 2010) 314. ↩
8. Donoghue v Stevenson [1932] AC 562 (HL). ↩
9. Hedley Byrne & Co Ltd v Heller & Partners Ltd [1964] AC 465 (HL). ↩
10. Caparo Industries plc v Dickman [1990] 2 AC 605 (HL). ↩
11. Cross and Harris (n 1) 39–40. ↩
12. Neil Duxbury, The Nature and Authority of Precedent (Cambridge University Press 2008) 113. ↩
13. Young v Bristol Aeroplane Co [1944] KB 718 (CA). ↩
14. Young v Bristol Aeroplane Co (n 13) 729–730 (Lord Greene MR). ↩
15. Tiverton Estates Ltd v Wearwell Ltd [1975] Ch 146 (CA) 172 (Scarman LJ). ↩
16. Morelle v Wakeling [1955] 2 QB 379 (CA). ↩
17. ibid 406 (Sir Raymond Evershed MR). ↩
18. Williams v Fawcett [1986] QB 604 (CA). ↩
19. ibid 615 (Sir John Donaldson MR). ↩
20. R v Simpson [2004] QB 118 (CA) [27] (Judge LJ). ↩
21. Geoffrey Marshall, ‘What Is Binding in a Precedent’ in D Neil MacCormick and Robert S Summers (eds), Interpreting Precedents: A Comparative Study (Dartmouth 1997) 503, 516. ↩
22. Andrew Burrows, ‘The Relationship between Common Law and Statute in the Law of Obligations’ (2012) 128 LQR 232, 241. ↩
23. Practice Statement (Judicial Precedent) [1966] 1 WLR 1234 (HL). ↩
24. London Street Tramways Co Ltd v London County Council [1898] AC 375 (HL) 380 (Earl of Halsbury LC). ↩
25. Practice Statement (n 23) 1234 (Lord Gardiner LC). ↩
26. Herrington v British Railways Board [1972] AC 877 (HL). ↩
27. Robert Addie & Sons (Collieries) Ltd v Dumbreck [1929] AC 358 (HL). ↩
28. Miliangos v George Frank (Textiles) Ltd [1976] AC 443 (HL). ↩
29. R v Howe [1987] AC 417 (HL). ↩
30. Director of Public Prosecutions for Northern Ireland v Lynch [1975] AC 653 (HL). ↩
31. Alan Paterson, The Law Lords (Macmillan 1982) 156. ↩
32. Joseph Raz, The Authority of Law (2nd edn, Oxford University Press 2009) 182–183. ↩
33. Constitutional Reform Act 2005, ss 23–60. ↩
34. Austin v Southwark London Borough Council [2010] UKSC 28 [25] (Lord Hope). ↩
35. R (Miller) v Secretary of State for Exiting the European Union [2017] UKSC 5 [246] (Lord Reed). ↩
36. Human Rights Act 1998, s 3. ↩
37. Human Rights Act 1998, s 6. ↩
38. Pinnock v Manchester City Council [2010] UKSC 45. ↩
39. ibid [48] (Lord Neuberger). ↩
40. Ghaidan v Godin-Mendoza [2004] UKHL 30 [30] (Lord Nicholls). ↩
41. Duke v Reliance Systems Ltd [1988] QB 108 (CA) 113 (Sir John Donaldson MR). ↩
42. John Bell, ‘The Acceptability of Legal Arguments’ in Neil MacCormick and Peter Birks (eds), The Legal Mind: Essays for Tony Honoré (Clarendon Press 1986) 45, 58. ↩
43. R v R [1992] 1 AC 599 (HL). ↩
44. John Finnis, ‘On the Incoherence of Legal Positivism’ (2000) 75 Notre Dame Law Review 1597, 1608. ↩
45. Jack Beatson, ‘The Role of Statute in the Development of Common Law Doctrine’ (2001) 117 LQR 247, 265. ↩
46. James Lee, ‘A Defense of Conflict in the Common Law’ (2016) 36 OJLS 327, 344. ↩
47. TRS Allan, ‘Precedent and Principle: Constitutional Morality and the Development of the Common Law’ (1987) 103 LQR 596, 612. ↩
48. Cross and Harris (n 1) 102. ↩
49. Dworkin (n 2) 239. ↩
50. Dobbs v Jackson Women’s Health Organization 597 US 215 (2022) (US Supreme Court), overruling Roe v Wade 410 US 113 (1973). ↩
51. Lon Fuller, The Morality of Law (revised edn, Yale University Press 1969) 79. ↩
52. Paul Craig, ‘Formal and Substantive Conceptions of the Rule of Law: An Analytical Framework’ [1997] PL 467, 484. ↩