David Harvey
Retired district court judge
When former Solicitor-General Una Jagose KC told a Law Association breakfast on 2 July 2026 that New Zealand risks “sleepwalking” towards a collapse of the separation of powers, she framed herself as the constitution’s anxious guardian. The framing deserves scrutiny.
For beneath the civic-minded language – the appeals to running clubs and book clubs, the warnings about cynical citizenries – Jagose advanced a set of propositions that, taken seriously, would do more to unsettle New Zealand’s constitutional arrangements than anything the courts’ critics have proposed. Her speech was not a defence of the separation of powers.
It was a defence of one branch’s steady encroachment on another, delivered by one of the most institutionally credentialled voices the encroaching branch could hope for.
The measure of that encroachment has been supplied, over the past two years, by three figures whose credentials are equally difficult to dismiss: Jack Hodder KC, who helped design the Supreme Court he now criticises; Emeritus Professor Peter Watts KC, whose New Zealand Law Review analysis of Ellis v R is the most forensically documented account of what the court has actually done; and Roger Partridge, whose report Who Makes the Law? Reining in the Supreme Court and subsequent “Supreme Court Matters” series have moved the argument from diagnosis to remedy.
Read against their work – and against Chris Trotter’s blunt political analysis of where this contest is heading – Jagose’s speech looks less like reassurance and more like confirmation that the critics have identified something real.
What the critics have actually shown
It is worth being precise about the case Jagose set out to rebut, because her rebuttal engaged a softened version of it.
Hodder’s 2024 paper, One Advocate’s Opinions – The “Least Dangerous Branch”? Predictability and Unease, made two charges.
The first concerned statutory interpretation: that the Supreme Court, through the “principle of legality” and expansive Bill of Rights readings, has developed the habit of navigating around the words parliament actually enacted whenever those words collide with what the judges regard as fundamental rights or Treaty principles – escalating interpretation, in his phrase, into a “constitutional battlefield”.
The second concerned the common law: the court’s repeated assertion that its role is to “develop” the law in response to “changing societal values”. Hodder’s question remains unanswered by anyone, including Jagose: by what logic, training, experience or evidence do courts identify and weigh society’s “values”? Judges have no polling operation, no electorate, no mandate. What they have is their own milieu – and “changing societal values”, in practice, means the values of that milieu.
The Supreme Court Act 2003 committed the new court to the rule of law and the sovereignty of parliament, values connoting stability and predictability. Nothing in it commissioned a journey.
Watts converted unease into demonstration. His analysis of Ellis establishes, through the primary sources, that the jurisdiction of New Zealand’s courts is a statutory grant: the Supreme Court Ordinance 1841 conferred on the colonial courts the jurisdiction of the English common law courts; the English Laws Act 1858 and the Imperial Laws Application Act 1988 confirmed and continued that foundation.
The courts did not create the platform on which they stand: parliament did. Against that background, what the Ellis majority did was extraordinary. In a case with no Māori party, no Māori complainant, and no party contending that tikanga was relevant, the court – at its own instigation, while the Law Commission’s review of tikanga’s place in the law was actually under way – declared tikanga potentially relevant to any issue of common law or statutory interpretation.
It discarded the orthodox recognition framework of Public Trustee v Loasby and its requirement of a genuine Māori connection to the facts, and replaced them with nothing determinate. As Watts puts it, abrogating the continuity of the 1841 Ordinance and the 1988 Act “would involve a constitutional moment for Aotearoa New Zealand” – quintessentially a task for the legislature or a referendum, not the judiciary.
His conclusion is not rhetorical excess but description: Ellis was a revolution, effected by judicial decree, that has “thrown a cloud over the entire body of existing case law and the entire statute book”.
Partridge has drawn out the structural point that makes the individual cases – Fitzgerald, Ellis, Smith v Fonterra, Family First, Edwards – more than a list of grievances.
A court whose authority to apply and develop the common law is conferred by parliament cannot coherently use the common law to limit parliament.
A court that releases itself from obedience to the legislature because the legislature has crossed a line the court itself drew is not protecting the constitution: it is asserting the authority to define it.
And a court that sets its own limits has, by definition, no limits.
That is why Partridge’s remedies – targeted corrective legislation, a “thin” statutory definition of the rule of law, tightened interpretation rules, reform of section 6 of the Bill of Rights Act, appointment criteria emphasising restraint – are not attacks on judicial independence. They are parliament doing what only parliament can legitimately do: settling, openly and accountably, questions the courts have taken to settling for themselves.
Jagose’s answer, and why it fails
Jagose’s response to all this proceeded by relocation and reassurance. She relocated the constitutional danger from the bench to the Beehive – parliamentary urgency, thinning consultation, “procedural workarounds” – and reassured her audience that the tikanga jurisprudence is “just the ordinary work of a common law legal system”, continuous with a Privy Council observation from 1901. Neither move survives contact with the critics’ actual arguments.
The urgency point is not wrong, but it is beside the point. Overuse of urgency is bad legislative practice, and Jagose was right to say no administration has clean hands.
But urgency is exercised within parliament’s undoubted procedural authority, is visible, is politically contestable, and is punishable at the ballot box.
It is, in other words, subject to exactly the accountability mechanisms that judicial lawmaking escapes. To equate a procedural vice that voters can remedy with a jurisdictional revolution that voters cannot is to flatten the very distinction on which the separation of powers depends.
The continuity claim fares worse, because it answers a question nobody asked. No serious critic – not Watts, not Hodder, not Partridge – denies that tikanga had a recognised place in New Zealand law before 2022. It did: in customary property matters and issues intrinsically connected with Māori affairs, admitted through the Loasby tests, with a real Māori connection to the facts.
That is precisely the settled, confined position that Ellis discarded. Citing the 1901 Privy Council to normalise Ellis is like citing the law of easements to normalise the abolition of land title: the existence of the old, bounded doctrine is not evidence for the legitimacy of its boundless successor. Watts’s charge is about method and mandate – a sweeping change, in an unsuitable case, uninvited by the parties, pre-empting a democratic process already in train.
“Familiar common law method” it was not. The common law’s genius is incremental analogy from case to case; Ellis was proclamation.
The Treaty question: the speech’s revealing moment
But the passage of the speech that matters most – the one that should be read carefully by anyone tempted to take the reassurance at face value – is Jagose’s rhetorical question: “Can parliament conclusively define, without Māori, the meaning of that Treaty that occupies the foundational place in our constitution?... Do you see limits to constitutional legitimacy there? Because I do.”
Consider what this proposition actually entails. Māori New Zealanders are not “without” parliament: they are in it – as voters on the general and Māori rolls, as members across every party, as ministers, as a dedicated set of electorates that exist nowhere else in the Westminster world.
When parliament legislates, Māori legislate, on precisely the same terms as every other citizen: through equal representation.
To say that parliament cannot “conclusively define, without Māori” the meaning of the Treaty is therefore not to say that Māori must have a voice – they have one, constitutionally guaranteed.
It is to say that the voice Māori exercise as electors and representatives is insufficient, and that some additional, extra-parliamentary Māori concurrence is required before the elected legislature’s word is final.
Follow the thought to its destination and the question becomes unavoidable: who holds this veto, and who chose them?
If it is the courts, then unelected judges have acquired the power to disallow legislation – the very power the Bill of Rights 1689 stripped from the Crown and that no New Zealand statute has ever conferred on any court.
If it is the Waitangi Tribunal, or iwi leadership bodies, then a corporately defined constituency, accountable to no general electorate, sits above the House of Representatives on the most fundamental question in the nation’s constitutional life.
Either way, the principle of political equality — one person, one vote, each vote of equal weight, the majority’s law binding until the majority changes it — has been abandoned, and abandoned not by referendum or entrenched enactment but by assertion from a podium.
Jagose sees “limits to constitutional legitimacy” in parliament defining the Treaty. Most New Zealanders, one suspects, would see a rather starker limit to constitutional legitimacy in the alternative she implies.
This is where Chris Trotter’s intervention, published in LawNews on 12 July, supplies what the doctrinal debate has lacked: a political sociology of the dispute.
Trotter – no fellow-traveller of The New Zealand Initiative – reads Jagose’s speech as, in effect, “a call for New Zealand’s politicians to straighten-up and fly right”, and identifies the contest as one “between the aspirations and values of the educated middle class and the beliefs of those occupying less-exalted positions in the socio-economic hierarchy”.
Even Jagose’s homely examples betray it: the running club and the book club are not where the median voter spends a scarce evening. The jurists enrolled in what Trotter calls the decolonisation and indigenisation projects genuinely cannot imagine that legislation dear to professional-class hearts might strike other citizens as illegitimate; and so, when electoral fortunes reverse – as they did in 2023 – the new parliament’s statutes are received not as democracy working but as an emergency justifying judicial correction. “Their rights run too deep,” the reasoning goes, “for a House of Representatives temporarily occupied by the wrong people to cancel.”
Trotter also identifies the intellectual ancestor of this reasoning: Lord Cooke’s famous aside in Taylor v New Zealand Poultry Board that some common law rights “presumably lie so deep that even parliament could not override them”.
Cooke chose torture as his example precisely because no decent person would quibble – and that is what makes the dictum so treacherous. The extreme case is the recruiting sergeant – the actual deployments are always more mundane.
Once “rights that lie too deep” is accepted as a judicially enforceable category, its content is whatever a majority of a five-judge bench says it is, and the flashpoints will not be torture statutes.
They will be, as Trotter predicts, legislation reversing the courts’ own tikanga innovations, or measures like the Legislation (Definitions of Woman and Man) Amendment Bill – contested social questions on which the professional class and the wider electorate diverge most sharply.
Partridge’s circularity objection and Trotter’s class analysis are the same insight in different registers: “deep constitutional norms”, unattached to any enacted text, resolve in practice into the norms of the people doing the divining.
The prudential warning the judiciary should heed
Trotter’s conclusion is the one the courts’ partisans least want to hear, and it is the political complement to Watts’s legal analysis: in any genuine struggle between parliament and the courts there can be only one winner, and it is not the courts.
The judiciary commands neither purse nor sword nor mandate. Its authority is, as Hodder observed, a function of its legitimacy, and its legitimacy depends on remaining within the areas where institutional competence is real and democratic accountability is not expected.
Jagose herself supplied the evidence, though she did not draw the inference. The same Helen Clark Foundation survey she cited for public distrust of government found – as her critics were quick to note – that only 43 per cent of New Zealanders believe the courts make fair, impartial decisions on the evidence, falling to 31 per cent among the financially struggling and 35 per cent among the under-30s.
A judiciary perceived to be pursuing the values of one social stratum polls like an institution of that stratum. The surest protection for judicial independence has never been the suppression of criticism or the sacralisation of the bench. It is judicial restraint.
Nor is the endgame merely reputational. A parliament persuaded that the courts have appointed themselves its supervisor has remedies, all of them lawful and all of them blunt: corrective statutes, redefined jurisdiction, and restructured appointments.
Every escalation by the bench legitimises the counter-escalation, and each round leaves the courts weaker and more politicised – the American road Partridge has repeatedly warned against.
Trotter reaches for the older precedent: an anointed English king who tried to overawe his legislators and lost his head. The courts will lose nothing so dramatic. They will simply lose – and constitutional government will be poorer for the contest having been forced.
Conclusion
Jagose asked her audience to talk about the constitution beyond the profession’s echo chamber. On this, she and Partridge agree, and both are right. But the conversation she proposes to take to the running clubs rests on a premise the running clubs are entitled to reject: that the limits on their parliament may be discerned and enforced by people they never elected, by reference to norms nowhere enacted, in the service of values they may not share.
The orthodox settlement – parliament makes the law, courts faithfully apply it, and the remedy for bad law is the ballot box – is not a relic. It is the only arrangement yet devised in which every citizen, Māori and non-Māori, urban professional and provincial tradesman, counts for exactly one.
Hodder, Watts and Partridge have shown, with a rigour their critics have not matched, how far the Supreme Court has drifted from that settlement.
Jagose’s speech, for all its polish, shows why the drift will not correct itself. The correction is parliament’s to make, it should be made calmly and soon – and, as Trotter reminds us, if it comes to a struggle, made it will be.
It is fair to record the contrary view.
Jagose and those who share her position would say that judicial review of the lawfulness of public power is itself part of the Westminster inheritance; that the common law has always developed, and that recognising tikanga is development of the same kind; that the Treaty’s constitutional significance cannot be reduced to ordinary majoritarian politics given the circumstances of its signing; and that talk of “reining in” courts risks the genuinely authoritarian drift she described.
Those arguments deserve engagement rather than dismissal. But engagement is what they have now received – from Hodder, Watts, Partridge and, in his own register, Trotter – and on the central questions of mandate, method and democratic equality, the answers have not been forthcoming.
This article was originally published by A Halfling’s View.
