The debate about New Zealand’s Supreme Court has been framed as a question about the current court – its composition, its appointments, its judicial philosophy. This column frames it differently: as a question about what the current court inherited, and from whom. The methodology that produced Ellis has a father. His name is Robin Brunskill Cooke.
Lord Cooke of Thorndon – Sir Robin Cooke, President of the New Zealand Court of Appeal from 1986 to 1996 – was New Zealand’s most celebrated jurist. The accolades were extraordinary. Chief Justice Elias described him as the country’s “finest [judge].” Geoffrey Palmer called him “the greatest judge that New Zealand has produced.” He sat on appeals to the House of Lords – the only Commonwealth judge in modern times to do so. He received honorary doctorates from Cambridge and Oxford. He held the Order of New Zealand, limited to twenty living New Zealanders.
The reverence is not in dispute. What is in dispute is whether that reverence has permitted a constitutional question to go unasked.
Every revered architect has buildings that outlast them. The buildings do not become sound simply because the architect was eminent. When a building starts to subside, reverence for its designer is not a structural argument. It is an obstacle to inspection.
What Cooke actually said
A case about eggs provides the starting point.
In 1984, a poultry farmer named Taylor was convicted of failing to answer questions put to him by an inspector of the New Zealand Poultry Board. His appeal reached the Court of Appeal. He lost. But the case is not remembered for its outcome. It is remembered for two sentences Cooke wrote in passing:
“I do not think that literal compulsion, by torture for instance, would be within the lawful powers of Parliament. Some common law rights presumably lie so deep that even Parliament could not override them.” (Taylor v New Zealand Poultry Board [1984] 1 NZLR 394, 398.)
Those two sentences assert, in plain terms, that some rights are so fundamental that parliamentary sovereignty does not reach them – and that courts are the institution empowered to identify which ones. The language is that of legal incapacity: Parliament “could not” override them. What courts would do in response, Cooke did not say. Taylor had lost. The statement was unnecessary. It was, in the most precise sense, a planted flag – the culmination of preparatory dicta in at least four earlier decisions.
Where did the authority come from? In Taylor, Cooke did not say. But he did attempt an answer – across two decades of extrajudicial writing – and the answer, examined carefully, does not hold.
In “Fundamentals” (1988), Cooke argued that the constitution rests on two “complementary and lawfully unalterable principles: the operation of a democratic legislature and the operation of independent courts.” Because free democracy itself underpins parliamentary legitimacy, he reasoned, Parliament cannot act in ways fundamentally inconsistent with free democracy – and working out what rights and freedoms are “truly fundamental” is “ultimately an inescapable judicial responsibility.”
The argument has a surface logic. But it contains a fatal flaw: it assigns to courts alone the authority to define the very standard against which parliamentary acts are measured. That is not deductive reasoning from a fixed constitutional premise. It is the conclusion dressed as a premise. Free democracy, on Cooke’s account, means whatever courts say it means.
“Fundamentals” is, in one respect, more careful than Taylor. For the two unalterable pillars – democratic legislature and independent courts – Cooke specified a declaratory remedy: it would be the responsibility of judges “to say so,” and if their judgments were disregarded, to resign rather than serve a state no longer entitled to be called a free democracy. Parliament, on that account, retains the final word.
But the modesty did not hold. His Hamlyn Lecture (1997) described parliamentary sovereignty as “a doctrine evolved by the courts, based on their judgment of political reality” – the rhetorical foundation for the claim that courts might re-evolve it. Once sovereignty is described as court-evolved, the door is rhetorically open to courts re-evolving it.
By “The Myth of Sovereignty” (2005), Cooke was explicit: parliamentary sovereignty “does not survive in-depth analysis.” What replaces it, he suggested, is “interaction, checks and balances, some degree of competition” – but without identifying any rule to determine who prevails when courts and Parliament compete. The implicit answer, as always, is: the courts. Legislation purporting to disestablish the courts would be “constitutionally impracticable and unenforceable.” The trajectory of his writings is not toward greater restraint.
“Fundamentals” is the high-water mark of modesty: not judicial supremacy, but constitutional guardianship – a court that declares, warns, and ultimately yields. But even in that form, the problem remains. A court that claims authority to declare when “free democracy” has been violated must also claim authority to define what free democracy requires. A court that claims authority to declare when "free democracy" has been violated must also claim authority to define what free democracy requires. That is a judgment on the constitutional validity of Parliament’s acts by reference to a standard the court itself supplies. The canary does not merely detect the gas. It decides what counts as poison.
Jeffrey Goldsworthy – Emeritus Professor at Monash and author of the definitive Commonwealth work on parliamentary sovereignty – engaged Cooke’s “deep rights” thesis in the same pages in “Is Parliament Sovereign? Recent Challenges to the Doctrine of Parliamentary Sovereignty” (2005). Goldsworthy identified the circularity precisely: “note the peculiarity of the notion that the courts are currently obligated to obey Parliament’s statutes, because it is sovereign, but can release themselves from the obligation, because its sovereignty is their creation and subject to their control... Any ‘obligation’ would be illusory.”
And on the argument from the extreme cases argument – the torture hypothetical, the redhead – Goldsworthy was equally direct: it does not follow that parliamentary sovereignty ought to be abandoned and judges authorised to invalidate outrageously unjust statutes, because “judges are almost certain to interpret such an authority too broadly.” The answer to parliamentary tyranny is not judicial supremacy. It is political resistance – and ultimately democratic constitutional change, brought about “openly and democratically, not surreptitiously through judicial decision-making.”
Cooke’s position presents an inescapable dilemma. Either it was serious – in which case it is a claim to judicial authority resting on a circular argument: courts defining the standard against which Parliament is measured, then certifying their own power to apply it. Or it was not serious – mere rhetoric, a hypothetical gesture toward unimaginable extremity. In that case, its continued reverential citation across four decades is itself evidence of something: that hagiography can protect a constitutionally radical posture from the scrutiny it demands. There is no comfortable middle ground.
Former Australian High Court Justice Michael Kirby – a self-described “kindred spirit” of Cooke and otherwise aligned with him on human rights and judicial method – identified the problem with precision: “By challenging the power of Parliament with notions of ‘deep rights’ the judge challenges the democratic character of the system of which the judiciary is part and which sustains the judiciary’s own legitimacy.” You cannot invoke democratic legitimacy to justify a judicial power that exists outside any democratic mandate.
The private law parallel
Cooke’s constitutional method had a private law mirror. Bowen v Paramount Builders [1977] 1 NZLR 394 is the emblem.
A homeowner discovered defective foundations years after purchase. The subsequent purchaser sued the builder in negligence despite having no contractual relationship with the defendant. The purchaser had taken no steps to obtain an assignment of contractual rights his predecessor may have had. The Court of Appeal, nonetheless, allowed the claim. The reasoning seemed incremental, but the conceptual slide was fundamental. The loss claimed was purely economic: repair costs, diminished value, the difference between what the purchaser paid and what they received. No physical harm to any other property. Financial disappointment dressed as property damage.
The contract/tort boundary – the line between obligations voluntarily assumed and duties judicially imposed – was precisely the boundary that distinguished Parliament’s domain from the court’s. Parliament could design a liability scheme for defective buildings, consulting widely, calibrating rules, accepting democratic accountability for the consequences. The Court of Appeal could only impose liability and move on. England retreated from the same path in Murphy v Brentwood [1991]. Australia’s High Court had never followed it.
The consequences for New Zealand took a generation to document: a leaky buildings crisis that produced something in the order of a billion dollars of liability, with councils as involuntary insurers and ratepayers bearing the cost.
Cooke did not yield. He argued the retreat extrajudicially in “An Impossible Distinction” (1991) 107 LQR 46, contending that the line the House of Lords had drawn was analytically incoherent – and three years later gave that view judicial effect, delivering the leading judgment of the New Zealand Court of Appeal in Invercargill City Council v Hamlin [1994] 3 NZLR 513. Cooke’s court declined to follow Murphy, treating the Bowen line as “New Zealand common law.” The Privy Council affirmed in 1996, citing “local conditions.”
Professor Peter Watts KC named the move in his seminal New Zealand Law Review article, “Taxonomy in Private Law–Furor in Text and Subtext.” “Sir Robin Cooke’s rhetoric,” Watts wrote, “does not, however, dent the distinction between damaging another person's property and a complaint about the quality of building work that damages nothing but itself. Indeed, even to talk of ‘damage’ in the latter situation can mislead; the product is defective from the start.”
Cooke’s objection was not that the distinction could not be drawn, but that he disliked the outcomes it produced. That is policy preference dressed as analytical impossibility – the kind of instrumental reasoning that turns judges into legislators.
The sparkler and the nuclear bomb
The standard response to this analysis – offered, in good faith, by many who admired Cooke and still do – is contextual. The 1970s and 1980s were a specific moment. Parliament was effectively captured by the Muldoon government operating without meaningful checks. The Privy Council connection created lag and detachment from New Zealand conditions. There was no Bill of Rights. Cooke, on this account, was filling a vacuum left by dysfunctional democratic institutions.
This argument is sincerely held and not without foundation. The historical context was genuinely unusual. The structural pressures on the Court of Appeal were real.
But the argument justifies outcomes, not methodology. A decision can be defensible on its merits while lacking constitutional warrant in its method. And Cooke never framed his approach as a temporary measure. He asserted permanent constitutional principle – “lawfully unalterable” structures, the “myth” of sovereignty, deep rights that endure regardless of parliamentary will. The contextual defence is constructed by others on his behalf; he claimed not a contingent licence but an inherent power.
Nor does the argument survive its own logic: if judicial supremacy over Parliament is justified because Parliament cannot be trusted, that claim requires a judicial power no Parliament ever authorised. Then Deputy Prime Minister later Attorney General – Michael Cullen identified the circularity: for the courts to “find” a higher law modifying Parliament’s constitutional status would “amount to constitutional change by stealth.”
None of which means Cooke’s era produced nothing of value. His contributions to administrative law, to procedural reform, to the clarity of legal reasoning in dozens of areas – these are real and enduring. The point is not that Cooke was wrong about everything. It is that his methodology was without recognisable constitutional warrant, that its consequences were bounded in his hands by his own instincts and the informal constraints of a small legal community, and that the building he designed is still standing – now occupied by judges who have discarded the restraints he at least nominally observed.
The sparkler has become a nuclear bomb.
Ellis and the revolution Cooke made possible
Ellis v R (Continuance) [2022] NZSC 114 must be understood for what it was. Watts – hardly a polemicist – described it as “revolutionary.” That word was earned.
The Ellis appeal was mundane: should a criminal appeal continue after the appellant’s death? Neither Ellis nor the complainants were Māori. No party had raised tikanga. The case could have been resolved on orthodox principles. Instead, the majority – at its own initiative – seized on this unremarkable procedural question to attempt to remake New Zealand’s legal foundations.
Justice Glazebrook’s leading judgment dismissed the traditional framework for recognising tikanga within the common law on the ground that it rested on “the assumption of the superiority of Western values and a view that the common law inherited from the United Kingdom should be presumptively dominant.”
But that dominance was not a view. It was a statutory fact. As I argued in Who Makes the Law?, the common law was presumptively dominant not because of any cultural assumption but because Parliament had repeatedly said so – in the legislation that founded the courts, in the English Laws Act 1858 and its successor, the Imperial Laws Application Act 1988. The Supreme Court was not correcting a bias. It was overriding Acts of Parliament without acknowledging that it was doing so.
Having swept aside those statutory foundations, the Court then declined to replace them with anything coherent. Justice Glazebrook stated she would not “attempt a reformulation of the test.” Chief Justice Winkelmann described tikanga as relevant to “an infinite variety of factual circumstances.” Justice Williams conceded that courts “cannot authoritatively declare [tikanga] for general purposes.” A source of law had been installed without a rule of recognition, without a framework for application, and without any mechanism for resolving conflict with existing doctrine. Watts put it precisely: the decision had thrown “a cloud over the entire body of existing case law and the entire statute book.”
This is not Bowen writ large. It is Bowen taken to its constitutional terminus. The method is identical: find the existing rule objectionable, declare the relevant distinctions artificial, and substitute judicial preference for parliamentary choice. What changed is the scale. Bowen reshaped liability in negligence for defective buildings. Ellis purported to restructure the foundations of New Zealand’s legal order.
The lineage is traceable. Bowen established that courts could impose obligations Parliament had not authorised, on parties who had not assumed them, in service of judicial perceptions of good policy. Taylor asserted that courts could identify rights beyond Parliament’s reach, without constitutional foundation. Fundamentals converted those isolated positions into a programmatic claim: two unalterable principles, judiciary and Parliament as complementary partners.
The current Supreme Court inherited all three – and applied them without even Cooke’s nominal acknowledgment that parliamentary sovereignty exists as a meaningful limiting principle.
Justice Winkelmann describes the constitution as “constantly being reshaped” by court decisions – Cooke’s claim in modern dress. Justice Glazebrook cited the Taylor dictum approvingly in her own extra-judicial writing. Justice Williams describes “law-making as nation-building” – the Cooke method, stated without qualification.
The blueprints are on the wall of the court they inherited.
What the hagiography has cost
The legal profession has paid a price for its reverence.
The emperor’s new clothes were pointed out, more than once, and by serious people. Goldsworthy had done so, contemporaneously. James Allan, Professor of Law at Otago before leaving for Queensland, pressed the same case. Professor John Smillie, then Allan’s Otago colleague, said Cooke’s Taylor claim had “no [] foundation in legal or democratic principle.”
None cut through. The problem was never that Cooke’s methodology lacked critics. The problem was that the critics could not compete with the reverence. The festschrift format, the honorary doctorates, the annual Robin Cooke Lecture, the Order of New Zealand: all of it made constitutional criticism feel, to those within the profession who might have raised it, like attacking a national monument.
The mainstream of New Zealand legal culture treated his methodology as something to celebrate and extend, not examine. What was there for lawyers not to like, when the methodology gave them political power? And when the legal realist and decolonisation frameworks arrived, they provided Cooke’s successors with academic cover for what he had practised on instinct alone – a double insulation: the reputation foreclosed constitutional scrutiny, and the jurisprudential apparatus gave his heirs respectability.
That deference has consequences. Ellis. Fitzgerald, in which the Court effectively declined to apply clear statutory language because it disapproved of the result. Climate Clinic, in which the Court read a 1991 mining statute as requiring consideration of climate change impacts that Parliament never mandated. Each is the Bowen move in a different doctrinal disguise: find the existing rule objectionable, declare it insufficient, and substitute judicial preference.
It has produced a jurisprudence in which “values” have replaced “rules,” “balancing” has replaced “law,” and the question the Court actually asks – “What outcome serves contemporary justice as the judges understand it?” – is no longer the question Parliament was answering when it enacted the statute.
Jack Hodder KC, in his landmark paper for the Legal Research Foundation, asked the question the hagiography foreclosed: “By what logic or experience, and by what criteria, do the courts identify and weigh inconsistent ‘values’?” It was a question for Cooke as much as it is for the current court. It is always unanswered.
The indictment
The indictment is this.
Cooke was careful about how far he went. His instincts were conservative enough, and the legal community small enough, to provide informal guardrails that kept the consequences bounded. But he planted a constitutional flag without constitutional authority. He assigned to courts alone the power to define what free democracy requires. That is not a constitutional foundation. It is a circular claim, however modestly dressed.
Cooke’s methodology substituted judicial values for legal rules, dressed policy preference as analytical necessity and bequeathed to his successors a licence without limits – and without the instincts that had kept it in check.
The troublesome work of the current Supreme Court is his product. It has discarded even his concession that Parliament retained the final word. It rewrites statutory language, installs new sources of law without parliamentary mandate and reads into statutes obligations Parliament never imposed.
The building outlasts the architect. And when the building starts to collapse, you cannot repair it by celebrating the genius of its designer. You have to look at the foundations. They were defective from the start.
To read the article on the LawNews website, click here.