ROGER PARTRIDGE: Why Courts Cannot Determine the Scope of Their Own Authority

Critics of judicial overreach face an odd challenge. The most sophisticated response is not to defend the decisions – it is to deny that constitutional limits exist at all. If courts made the rules, the argument runs, courts can remake them. Last month’s column, An Inheritance Worth Defending, drew that response, among others.

Four arguments recur. On the surface they are distinct – one concerns the foundations of parliamentary sovereignty, one the proper limits of common law development, one a comparison with Australian constitutional law, and one concerns the lessons to be drawn from two landmark cases on judicial review and constitutional supremacy.

But they share a common premise: that courts are entitled to determine for themselves the scope of their own authority. Each challenge, examined closely, is a variation on that theme. And the theme does not survive scrutiny.

Parliamentary Sovereignty Is Not a Common Law Invention

The most sophisticated objection runs as follows. Courts created parliamentary sovereignty as a common law rule; they can therefore modify or limit it. The argument has academic pedigree. In obiter dicta in Jackson v Attorney-General [2005] UKHL 56, Lords Steyn and Hope went further, suggesting that parliamentary sovereignty is a common law construct created by judges.

Lord Steyn stated that it is “a construct of the common law” and that “the judges created this principle”; Lord Hope agreed that the principle “has been created by the common law.” In Taylor v New Zealand Poultry Board [1984] 1 NZLR 394, Cooke P. mused that some common law rights might “lie so deep” that no Parliament could abrogate them – and that courts might refuse to give effect to legislation that tried.

The implication is stark: the claim that the Supreme Court is exceeding its constitutional role becomes incoherent. A court that sets its own limits has no limits.

But the claim is self-defeating. If parliamentary sovereignty is grounded in the common law, and courts made the common law, then grounding parliamentary sovereignty in the common law simply returns authority to the courts – which is precisely what the argument was trying to establish. It bootstraps judicial authority from itself.

A yet deeper challenge asks: If the courts did not create parliamentary sovereignty, what grounds it?

“It is turtles all the way down,” said one commentator, invoking the classic image for infinite regress: every foundation rests on another, and the regress never ends. If that is right, the claim that any court has overstepped its constitutional role is equally groundless. There is no foundation for it.

Fortunately, however, the tortoise stands on firm ground.

Parliamentary sovereignty emerged not from judicial decree but from centuries of political struggle – the Case of Proclamations (1610), the Civil War, the Bill of Rights 1689 – in which Parliament asserted authority over both Crown and courts. When courts articulated that settlement in their judgments, they were acting as witnesses to a constitutional fact, not its authors.

As Oxford’s Emeritus Professor John Finnis has explained in his magisterial essay, “Judicial Power and the Balance of Our Constitution,” judicial recognition of that settlement was retrospective, not creative. Courts did not invent parliamentary sovereignty – they recognised it.

Jeffrey Goldsworthy demonstrated in The Sovereignty of Parliament and The Myth of the Common Law Constitution that parliamentary sovereignty rests on the convergent practice of courts, Parliament, and Crown treating Parliament’s authority as foundational – not because Parliament declares it so, but because every institution behaves consistently as though it is true.

That convergent practice is what H.L.A. Hart, in The Concept of Law, called a rule of recognition: the foundational norm from which all other legal validity flows. Attempting to derive it from something deeper – the common law, natural law, popular will – does not solve the regress. It is the regress.

But even setting this aside, the argument does not arise in New Zealand. Here, the common law itself exists by parliamentary authorisation. The Supreme Court Ordinance 1841 conferred on New Zealand courts the jurisdiction to apply English common law. The English Laws Act 1858 and the Imperial Laws Application Act 1988 confirmed and continued that foundation.

As Emeritus Professor Peter Watts KC showed in his New Zealand Law Review analysis, courts derive their authority to apply and develop the common law from Parliament’s legislation, and exercise that authority within the limits Parliament has set – not the other way around. Consequently, courts in New Zealand do not stand in a position of co-equal constitutional authorship with Parliament. Their authority to apply and develop the common law is itself conferred by Parliament. That makes the claim that courts may use the common law to limit Parliament conceptually inverted.

The Limits of Common Law Development

A second challenge is more general. Courts have always developed the common law – that is what common law courts do – so why should the Supreme Court’s recent jurisprudence be any different?

But consider a thought experiment. The Supreme Court’s next decision announces that it has reassessed the enforceability of contracts. The doctrine of consideration, it finds, is a historical relic ill-suited to contemporary values of fairness and social solidarity. Henceforth, agreements will be enforceable on broader equitable principles to be developed case by case.

Would that be legitimate? After all, courts created the law of contract. On the critics’ logic – courts may unmake it too. Yet every lawyer reading this would recoil. The question is why.

Not because courts cannot change the law. They can and do. But because some changes are of a kind and magnitude that place them outside what courts are institutionally authorised to make. The law of contract is not merely a judicial preference revisable by the next bench. Its revision belongs to Parliament.

The values embedded in the common law – certainty, voluntary obligation, protection of persons and property, the confinement of judicial discretion – were not invented by judges. They accumulated through centuries of consistent practice. They are what the common law is, not a preference judges are free to revise.

As Oxford’s John Finnis has explained, the declaratory theory – the principle that courts find and apply the law as it already is, rather than create it anew – is not the fairy tale Lord Reid famously dismissed. Properly understood, it is a disciplined constitutional responsibility.

Courts are not free to update those values to reflect contemporary preferences. Their task is narrower: to identify what the law already requires, correcting only errors inconsistent with the law as a whole – an internal standard of coherence with existing doctrine, not a licence for judicial renovation.

Two cases examined in an earlier column in this series illustrate how that standard operates in practice – and where courts have crossed it. Donoghue v Stevenson [1932] AC 562 falls on the right side of the line: legitimate method, albeit imperfectly executed. Lord Atkin’s handling of some precedents and his extension of the duty across both English and Scots law without adequate justification attracted criticism for years. For all that, Lord Atkin reasoned upward from existing duty categories to the principle already implicit in them – the neighbour principle was deduced, not conjured up.

Bowen v Paramount Builders (Hamilton) Ltd [1977] 1 NZLR 394 falls on the wrong side: crossing a boundary the existing cases had consistently respected. As Watts demonstrated in “Taxonomy in Private Law – Furor in Text and Subtext,” the Bowen line deployed the Donoghue tort to protect a wholly different interest – bare economic loss – substituting judicial policy for a principled limit.

R v Ellis [2022] NZSC 114, discussed in detail below, represents an altogether different register – not imperfect execution, nor crossing a doctrinal boundary, but what Watts shas called a revolution.

Even Lord Bingham, who took a broader view of judicial lawmaking than Finnis, identified five categories of question that belong to Parliament in his essay “The Judge as Lawmaker: An English Perspective”: where citizens have ordered their affairs in reliance on settled law; where reform requires detailed legislative work beyond judicial competence; where the issue involves contested social policy; where Parliament is actively engaged; and where the matter is far removed from ordinary judicial experience. The classical liberal values embedded in the common law – autonomy, voluntary obligation, protection of persons and property – were not invented by judges but drawn out of a tradition that had always embodied them. Once principles acquire centuries of consistent application, changes of that magnitude require democratic deliberation only Parliament can provide.

The distinction between these three categories – imperfect execution, crossing a doctrinal boundary, and overturning the jurisdictional settlement itself – is not a theoretical refinement. It is the difference between judicial development that remains answerable to the law and judicial action that places itself beyond it.

Ellis Is Not Mabo – It Is the Revolution Mabo Was Not

Perhaps the most pointed objection draws on Mabo v Queensland (No 2) (1992) 175 CLR 1. Just as the High Court of Australia recognised Indigenous land rights in that landmark decision, the argument runs, the Supreme Court was entitled – even obliged – to give tikanga its proper place in New Zealand’s common law in Ellis. Some even treat the two cases as essentially alike.

But the comparison does not survive scrutiny – and that is so even granting Mabo anything Ellis’s defenders may claim. In Mabo, Brennan J, delivering the majority decision, insisted the Court was “not free to adopt rules that accord with contemporary notions of justice and human rights if their adoption would fracture the skeleton of principle which gives the body of our law its shape and internal consistency.”

Mabo attracted serious black-letter criticism – Dawson J dissenting, and Dyson Heydon AC QC later identifying it as a case where courts entered terrain better handled by Parliament.

But whatever one makes of Mabo’s methodology, what Brennan J said the Court was doing was recognising a narrow, historically grounded doctrine – native title – governing rights in land that pre-existed colonisation. The High Court asked what the common law, properly understood, already required. It did not reach outside that framework to install a new general source of law.

Ellis made no such claim to restraint. Tikanga was already part of New Zealand’s common law in certain circumstances – cases involving customary property rights and matters intrinsically connected to Māori affairs, where courts had always required a specific Māori connection to the facts. Nobody seriously disputed that.

What the Ellis majority did was categorically different from the majority in Mabo: it declared tikanga an independent and general source of law, applicable to any issue of common law or statutory interpretation regardless of any Māori connection to the parties or facts.

As Watts demonstrates in his New Zealand Law Review analysis, the majority did not merely extend tikanga’s existing role – it installed a parallel legal order alongside the common law, without any rule of recognition, without any Māori connection requirement, and without parliamentary authorisation.

Watts calls Ellis a revolution. The description is precise. Mabo at least claimed the constraint of the common law’s existing framework and a defined subject-matter. Ellis claimed neither – no subject-matter limits, no rule of recognition, no democratic mandate. If New Zealand wishes to establish a bijural legal system incorporating tikanga alongside the common law, that is a legitimate constitutional choice. But it is Parliament’s choice to make, not the courts’.

Drawing the Wrong Lessons: Anisminic and Marbury

Two further objections invoke landmark cases as authority for expanded judicial power. Each draws the wrong lesson from the case it cites.

Anisminic Ltd v Foreign Compensation Commission [1969] 2 AC 147 is invoked as authority for a broad judicial power to supervise and override Parliament. The objection misreads the case – and in doing so, misreads what made it an unexceptional decision.

Lord Reid asked what Parliament actually meant when it provided that a determination “shall not be called in question in any court of law.” He concluded that a determination infected by jurisdictional error was not a “determination” within the meaning of the clause at all. That is orthodox construction. Like Donoghue v Stevenson, Anisminic reasoned from within the existing legal framework to a conclusion already latent in it. Far from supporting judicial supremacy, it is a model of the interpretive discipline this series has been advocating.

The cautionary tale is what later courts made of it – and here the parallel with Bowen is apposite. Lord Reid’s carefully premised conclusion was generalised into a near-constitutional entitlement that Parliament cannot remove by ouster clause. That severs the conclusion from the premises that made it sound. The self-contradiction is direct: Lord Reid’s reasoning rested on Parliament’s presumed intent not to exclude review; the expanded version makes that presumption irrebuttable even against Parliament’s expressed intent. A doctrine that defeats Parliament’s expressed intent on the ground of Parliament’s presumed intent has abandoned the premise it built on. It is not a development of Anisminic. It is its inversion.

Marbury v Madison 5 US 137 (1803) – the great case in which Chief Justice Marshall established the power of American courts to strike down legislation – is invoked as a further precedent for expanded judicial authority. If Marshall CJ could claim that power, why should New Zealand courts not exercise equivalent authority?The answer should be obvious. Marbury depended entirely on a specific constitutional premise: the existence of a written, entrenched constitution expressing supreme law that Congress was bound to observe. Judicial review under Marbury is not a power courts claim for themselves; it is a power the Constitution confers on them, because that Constitution is the higher law they are bound to apply.

Without such an instrument, there is nothing for courts to enforce against Parliament – and no basis for substituting judicial judgment for parliamentary will. Remove that premise, and Marshall’s reasoning has nothing to attach to. New Zealand has no such instrument. The New Zealand Bill of Rights Act 1990 is explicitly non-supreme – section 4 preserves Parliament’s power to enact inconsistent legislation.

The Common Law Is Not Infinitely Plastic

Each of these challenges, examined closely, is a version of the same claim: that courts may determine for themselves the scope of their own authority. That claim, if accepted, would leave nothing stable. It would license courts to dismantle every settled principle on the ground that judges shaped it and may therefore reshape it.

When courts substitute their policy judgments for Parliament’s, citizens lose the ability to read the law and plan accordingly. Contracts become unpredictable. Investments go unmade. Democratic accountability disappears – the institution that made the law cannot be voted out. The building consent crisis – ratepayers bearing over a billion dollars in liability for a judicial policy choice Parliament never made – is only the most documented example of what follows. And courts, deciding cases on the facts before them, with no submissions process and no capacity to canvass wider consequences, are poorly equipped for the social policy choices they are increasingly making. These are not theoretical concerns. They are the documented consequences of the decisions this series has been examining.

Constitutional change of the kind these challenges implicitly advocate is legitimate. But it requires a democratic choice – made openly, by those accountable for it, through institutions the public can hold to account. What is not legitimate is constitutional change by judicial increment, case by case, without mandate and without the possibility of correction. The inheritance is worth defending. So is the understanding of why.

This column was first published in LawNews, the journal of The Law Association, on 13 April 2024. Roger Partridge writes at Plain Thinking.