Some External Constraints and Constitutional Principles

This paper examines five sources of constraint for Parliament, reflecting “constitutional principles”. They are:

  1. Fundamental human rights and values (including those affirmed in the New Zealand Bill of Rights Act 1990)
  2. The principle against retrospective legislation, including the principle against reversing court judgments
  3. The Treaty of Waitangi
  4. Constraints implicit in the doctrine of the separation of powers
  5. Constraints (or at least considerations) flowing from the fact that it is courts that ultimately determine the meaning of legislation and do so in light of their reading of constitutional constraints (those above, as well as others)
A preliminary point: is Parliament constrained at all?

The question whether Parliament is legally constrained is, of course, controversial. Generations of lawyers have been trained in the theory of Professor Dicey of Oxford, writing at the turn of the 20thCentury. Dicey explained that under the British Constitution, largely unwritten just as ours is, Parliament is the supreme law-maker—meaning that no law it enacts can be invalidated by any other institution. Hence there are no explicit legal constraints upon it.

Dicey never advanced this idea to suggest that Parliamentary was licensed to make unjust or tyrannical legislation. His emphasis was different: supremacy flows from the fact that no body above Parliament may rule its laws illegal. What Parliament says is the law. That is why there can be no such thing as legal constraints on parliament.

Dicey emphasised practical (or political) restraints:  the essential decency of elected representatives, their natural desire to be popular and be re-elected, the opportunity of the citizenry to replace them at genuine elections. The wise and principled law-making of elected representatives was the guarantee against oppression and injustice.

Parliamentary supremacy was not the main point, only a corollary of the main point. The main point is that Parliament is the judge of its own justice in law-making.

It could be otherwise. A place could be made in a new New Zealand constitution for express limits on the power of parliament, enforced by the ordinary courts or some other institution. Naturally that would put intense focus on the exact wording of those express limits, and (if this were to be the case) on the legitimacy of their impact being determined by judges.

Whether that would be an improvement over the status quo is (to me at least) a genuinely difficult question.

For the moment the practical constraints on our Parliament remain in its observing certain limits—some not reduced to writing, others affirmed in the Bill of Rights. Those who advise governments, and who develop policy options that may become law, have to pay heed to these constraints. The decency and respect for fundamental values that we expect from our legislators obviously must be found there as well.

1. Fundamental values and the Bill of Rights as a constraint upon Parliament

The story our laws tell

If one were to read our entire statute book and ask “what stories does it tell about New Zealand, what are its deepest themes?”, the answer would, I suggest, be something like this:

We want to improve our condition—our health, our wisdom, our material well-being, our environment. We value the ability for each of us to make our own way in life. We want to be aware of our heritage; we want “sustainability” for future generations. We want to guard ourselves from crime and avoidable accidents. In pursuing these goals we recognise that there are certain things we will not do. We value human life, dignity and freedom. We do not legitimate the taking of a life other than in self-defence. We do not take away liberty without proof of guilt beyond reasonable doubt. We leave much to individual choice in a market economy, one constrained by rules designed to make it fair.

These themes arise in what our legislatures and judges have done, in statutes and through the common law, for centuries. For the most part the judicially-derived themes are protective of our liberty and dignity; the parliamentary themes are not inconsistent but include the more aspirational ones—health, welfare, education, discovery and innovation.

When it became necessary to do so, from about WW2, we began to use the language of rights, more recently “human rights”, to describe some of these deep values to which we subscribe.  (To mark us out form other world views, perhaps, such as Communism.)

But much of what we now call human rights has always inhered in the law. Contract law has promoted commerce, and with that, human dignity, liberty and autonomy. Contract law asked no questions about our reasons and preferences, but enforced our bargains. Only in the 20th century did it become obvious that liberty of contract (at least by goods and services providers) could come at the cost of equality for all. It took statutes to prohibit race and sex discrimination and remove some of our liberty, but in the cause of equality.

Tort law, for its part, protected rights to property, to bodily integrity, to reputation, to be free from the harm carelessly inflicted by others, and so on.

When the UN proclaimed its Universal Declaration of Human Rights, it chose rights that were drawn from the legal systems of the civilised nations of the world. These human rights were great abstractions—of what our civilisation has held to be important. They certainly underpinned our common law and much of our statute law. They are freedom of expression, religion, association, right to a fair trial and so on.

As is well known, eventually there were two covenants on rights, finalised in 1966: The Covenant on Civil and Political Rights and the one on Economic Social and Cultural Rights: the right to health, to education, to welfare etc.

A parlour game for lawyers is to try to imagine a New Zealand statute, any statute, that cannot be sheeted home in some way to some conception of a human right as its animating feature.  So we have Education Acts that give practical effect to the right to an education; a Public Health Act that does the same for health; a Crimes Act that protects life and limb, bodily integrity, as well as property, reputation and privacy. A Dog Registration Act that protects us from harm from uncontrolled dogs. Biosecurity, environmental law, conservation, commercial regulation—you name it, one can come up with a reason why it advances important interests that someone somewhere in the world is wanting, not without reason, to call a human right.

Would we expect it to be otherwise? We want our governments to do important things in our interests. We have given the label “rights” to important interests that we value very, very much. Modern controversies can become ever more refined, such as the right to freedom on the internet, or to rugby on free to air television.

But my point is that rights, principally social and economic ones, explain why Parliament enacts the laws it does. Rights, principally civil and political ones, explain why it must do so in certain ways, respecting basic values (fair trial, liberty, freedom from discrimination and so on).

It is no surprise then, that our Bill of Rights says in s 2 that the rights and freedoms it contains are “affirmed”, not created. Rights have been around for a very long time in our heritage. And, in New Zealand, as well protected as they are in those states that have had “constitutional” bills of rights.

But the Bill of Rights does not in fact contain all our rights. Some of the most fundamental are omitted—liberty and equality, for example.

But even if not included in the Bill of Rights itself, they are plainly fundamental in New Zealand. A study of our “precedents” reveals this. By “precedents” I mean the legislation that our Parliament has enacted in the past and hence what the citizenry has implicitly called for and approved.

The right to liberty

Many bills of rights in the world confer a right to liberty; ours does not. The US Constitution guarantees liberty in the 5th Amendment and in the 14th Amendment it prevents states from taking away liberty save by due process of law.  So, too, does the Canadian Charter.

But there is no doubt that liberty is a fundamental value in New Zealand.

Our whole apparatus of criminal law and procedure, and rules of evidence, reflects a solicitous concern for physical liberty. That is what explains the criminal process, the presumption of innocence and so on.

But liberty is more than not being locked away. It is the right to live in our own way and make our own choices. Countless examples can be given in our statute book of the way in which legislation goes so far, and no further, implicitly recognising a zone of protected liberty that it is somehow wrong for the state to invade. As the Americans would say, liberty includes the right to make fundamental personal decisions about how one lives one’s life, free of constraint, save when there is some compelling state interest that survives strict scrutiny by a court.

Those in the engine room of government, as advisors and legislators, have plainly been aware of this in the past.

Some examples:

When Parliament enacted the Smoke Free Environments Act 1990, for example, its ambit was protecting people from secondary smoke. It did not impede the basic liberty of persons to use tobacco. Even when it came to prisoners in prison cells, the only possible policies that prison superintendents can make were those that protected employees and others from secondary smoke: the basic liberty of a prisoner to smoke was, on the face of the Act, left untouched.

When Parliament prohibited discrimination in the Human Rights Act it left various categories of otherwise prohibited discrimination unlawful, It did not apply the Act to private clubs, nor to homeowners letting rooms in their own house, nor to churches and religious bodies who ordain and employ pastors and leaders. These exceptions—of going so far and no further—recognised a zone of liberty. A freedom, if you like, to make choices that not everyone will approve of.

The ban on smacking in 2007 was controversial precisely because it seemed an intrusion into the private life of New Zealand families, even if well-intentioned. And it is intriguing that the legislative (and political compromise) seemed to be one that New Zealand is coming to live with. That smacking is “banned”, but then it “sort of isn’t”—first because the political settlement involved inserting what was always going to be the case (that there was a prosecutorial discretion) and second because a referendum induced the Prime Minister to say that the legislation was never intended to criminalise a “light smack”. So in that somewhat stumbling way, we ended up maintaining a zone of privacy, or liberty, in which parents would not lightly be criminalised for smacking, and a showing of greater than trivial harm was required.

On more substantial matters, New Zealand decriminalised homosexual sex well before the USA held that decriminalisation was required by the word “liberty” in their Constitution. And we liberalised abortion law around the same time the Americans did, but it was done by Parliament and its recognition of liberty rights for women, and not by virtue of the word liberty in a Constitution.

Importantly, sticking with abortion, we also recognised the powerful counter value of the unborn child. Just as, in the USA, there is a compelling state interest in the potential life represented by the fetus, one that becomes less “potential” and more “actual” as birth approaches. All these things were factored into our law as well, without a constitution with a right to liberty and a right to life.

What do these examples tell us? That Parliamentary legislation must reckon with fundamental values such as human liberty and human life and that, since our Parliament’s inception, it has in fact sought to do so. Of course, persons may differ over how they weight the value of liberty over the need to prevent harm or protect rights of others. And this is the stuff of politics. But there is, or ought to be, a thumb on the scales for liberty, at least when it is not ranged against the rights of others but just against the interests that government is seeking to pursue. If it is to be outweighed, then a case must be made that it is proportionate to do so. And that is the currency in which policy makers must deal: what are we proposing to do to, or ask of, our citizens? What reasons can justify this level of intrusion?

The best way of testing the operation of this principle is to look at what we have not done, but could have done, by way of legislation. Interventions in families, for example, are generally predicated on the threshold of “harm to the child” and not the seductive “best interests of the child” (even though there are certainly places for that too; as when a court must make a decision affecting a child). But the strong default setting is that we leave the assessment of a child’s best interests to the child’s family or caregivers, and we have not lowered the threshold of harm so as to justify interventions for, say, “mere” bad dietary choices or poor parenting skills. That is a principle of liberty.

In the ECHR there is the “right to a private life” into which they have bundled questions like euthanasia, same sex marriage, and the right to practice homosexual sex.

In the US and Canada, as I say, the word liberty does that work. We have no word at all, if we are looking into the Bill of Rights for one. But the basic idea that we have these liberties is as well entrenched in our land as in others. We may lack the ex post facto review that comes with written constitutions empowering courts. But we do not lack the basic idea that there are constraints on Parliament: that it should go so far and no further.

There is a tremendous amount of agreement about the core of these rights in every society but it is at the margins that we disagree—over such things as abortion, same sex marriage and assisted suicide or euthanasia. Still, these things are amenable to principled discussion and debate in legislatures as matters of policy, and may perhaps be best resolved there rather than in winner-take-all litigation such as one gets in the US and Canada.

It is interesting to reflect how conservative our Bill of Rights drafters were: they left liberty and equality out, almost certainly because of the concern that it would drag us into the North American debates over exactly what these meant. Our Bill of Rights was animated by the “process theory” of constitutionalism:  It was important to protect voting rights and free speech; thereafter the people should be free to have such laws as our Parliament declared.

It is interesting to reflect that in 2008 when the ACT party promoted a Regulatory Responsibility Bill that gained a lot of support from, at least, the Business Round Table, this would have effectively introduced a parallel “Bill of Rights” containing a right to liberty and a right to property (that is, doing the work that a Bill of Rights on the US or European model might have done, but which ours does not). That movement, for a RRB, explicitly reflected disillusionment about elected representatives and questioned whether they were reaching the right sorts of conclusions on what liberty and property rights meant. The RRB promoters though our law would be better served if the courts were empowered to give their opinions on whether legislation infringed those values.

It is true that these very broad and vague rights leave a lot to the application process.  But that does not mean that we should not always be vigilant in properly analysing and labelling our proposals for laws: We need to ask: do they invade personal liberty, the right to make fundamental choices about how we live? If so, can they be justified under the heightened standard that such laws ought to require?

The Bill of Rights

The Bill of Rights goes well beyond liberty (or perhaps just contains more specific iterations of liberty: to speak, believe, assemble, move, and so on). Most of our fundamental rights are affirmed in the Bill of Rights. Recall that this was initially intended to be a supreme law Bill of Rights that would have empowered courts to invalidate legislation after it was enacted, if they concluded that the legislation unreasonably infringed a protected right or freedom.

But, as enacted, the Bill of Rights now says explicitly (in s 4) that no court may refuse to apply, nor hold invalid or in any way ineffective, a legislative provision by reason of its inconsistency with the Bill of Rights.

So far as the law-making process is concerned, the salient points about the Bill of Rights are these:

Its principal mission, said Geoffrey Palmer in the White Paper introduction, was to serve as a set of “navigation lights for the whole process of Government to observe”.  While judicial power over legislation was reduced to mere interpretation under s 6, not invalidating it, the basic aim of controlling government remained. The section 7 process was inserted into the Bill of Rights as enacted; it had not been in the White Paper proposal.

Section 7 requires that the Attorney-General make a report to Parliament if he or she considers any provision of a bill to be inconsistent with any of the rights or freedoms contained in the Bill of Rights. Plainly that was designed to minimise the chances of infringing legislation, whether passed unwittingly or deliberately. Someone was assigned the job of spotting inconsistencies. And that was the Attorney-General, a member of the government.

The s 7 procedure is a well-known procedure within the Ministries and Departments. You will know that opinions (called “vets”) are given by the MOJ team, or by Crown Law in the case of Justice bills (and often there is collaboration between the two sites of expertise).

There have been 58 s 7 reports over the nearly 22 years of the Bill of Rights; 28 for Government and 30 for non-government bills. That is a larger number than I would have predicted 22 years ago (I predicted 0). What is the cause of this?

I think there is a lack in the expected sense of “enormity” in passing a bill that breaches the Bill of Rights. This lack perhaps developed because the first Government that had to reckon with the Bill of Rights was a Government that had lampooned it and voted against it on enactment just a few weeks before the 1990 election. It was saying that it would likely repeal it when elected.

But it didn’t. They quite responsibly followed its processes. But nor did they vest it with great significance. The first Attorney-General was at pains to point out, for example, when advising that a bill breached the Bill of Rights in his opinion, that this did not mean that Parliament could not pass it. And so there emerged the idea that breaching the Bill of Rights was not impossible, not really such a bad thing. This is a phenomenon still being manifested, quite oddly, when Attorneys-General signal a breach under s 7 and then go on to vote for the bill themselves. Odd when you consider the implications of s 4 in the context of s 5. That (in the words of s 5) we have a law on our books that is unreasonable to the point that it is not demonstrably justified in a free and democratic society.

The problem is illustrated at its extreme by the Prostitution Reform Bill (2003). Obviously it was contentious. It legalised brothels. An issue then arose about the local government’s by law making power to regulate advertising signs on brothels. By way of an SOP the Government introduced a new clause about bylaws that explicitly overrode the Bill of Rights: [1]

Despite subsection (1), a bylaw may be made under section 12 even if, contrary to section 155(3) of the Local Government Act 2002, it is inconsistent with the New Zealand Bill of Rights Act 1990.

That is the first and still the only instance of an explicit overriding of the Bill of Rights. This cannot remotely be argued to be consistent with the Bill of Rights. Yet it was done uncontroversial in the cause of restricting commercial expression about an activity that was being made legal. It set a dangerous precedent. There are, of course, few supporters for sexually explicit advertising signs for brothels, or even non-sexually explicit signs for that matter. But, again, the whole point of a Bill of Rights is to set out rights, allied with a requirement that limits upon them be justified. It is people with unpopular causes that most need a Bill of Rights. It seems to me, therefore, that councils ought to have been left the task of drawing up bylaws that accomplished advertising restrictions in a Bill of Rights-consistent manner. After all, the proportionality test readily permits objective such as preserving neighbourhood amenities, and a degree of restriction on advertisers’ freedom of expression is plainly reasonable. What is not reasonable is excluding local government from the very need to act reasonably.

Another manifestation of the s 7 problem in the early days was this:

The Bill of Rights seemed to be regarded, sometimes at least, as some sort of final hurdle to cross, after the real work of determining the appropriate policy to pursue. Then, with that policy decided, if the bill failed to clear the hurdle, that was no great problem because s 4 meant it could be enacted nonetheless.

We saw this in a Ministry of Health paper on banning “direct to consumer advertising” of medicines. The paper included this observation: because government is able to introduce legislation inconsistent with the Bill of Rights, the “emphasis must be on whether [the] proposal [to ban advertising] is justified from a policy perspective, not whether the proposal is consistent with the Bill of Rights Act 1990”.

That, of course, turns the whole Bill of Rights on its head. It is not as if there can be a realm of  “policy” that excludes the Bill of Rights. The whole point is that the Bill of Rights is the standardfor policy, as it is for law and conduct. It is the navigation lights. Contrary to the quoted statement, then, the emphasis must be on whether policy is consistent with the Bill of Rights. An infringing policy ought not knowingly to be suggested for enshrinement in legislation.

To be clear: the reasons that make any particular policy worthwhile to pursue may well make it reasonable to limit rights. That is what s 5 is for: to justify reasonable limits on rights that are demonstrably justified on a free and democratic society. But the point is that the structured and disciplined approach envisaged by s 5 must be taken. What is the s 5 standard:  (identifying objectives for a law and assessing the rationality and proportionality of the means in light of the objectives) is required to be taken so as to ensure that policies conform to the Bill of Rights. Sidelining the Bill of Rights altogether in order to pursue a policy misses its whole point.

A section 5 analysis is likely to involve the assembly of evidence. Section 5 is about a culture of justification.

The true message of s 4 is very different. It is not a licence to Parliament to pass infringing legislation. Rather, it speaks to the possibility that a court may find a provision in an enactment to be inconsistent with rights, even though Parliament believed it was. Without s 4 the question would have arisen “which of the two conflicting statutes applied”, the Bill of Rights or the one before the court? Section 4 was inserted to answer that question. Its message is that the Bill of Rights is subordinated. But recall s 4 is only a message to courts.

The message of the Bill of Rights to Parliament is that Parliament should legislate consistently with it: see s 3. The purpose behind s 7 is to facilitate that, by requiring the Attorney-General to advise if he or she believes that a bill is inconsistent. Of course, a s 7 report is, in the end, just the view of the Attorney-General. It would be quite legitimate for Parliament itself, or sufficient members of it, to disagree, and enact a bill that the majority of its members believe to be consistent. It is not only the views of the Attorney-General or judges that count. The Bill of Rights is in fact well balanced, in that regard. It leaves each branch to make its own judgment.  The Attorney-General gives a view, the House can take a different view, and the judges may be invited to give their view also if litigation is brought. But the ultimate message of s 4 of the Bill of Rights is that it is Parliament’s view that prevails.)

In giving that account of how the Bill of Rights is intended to operate you will note that I envisage Members can quite legitimately disagree on the merits with the Attorney-General’s opinion. They can enact legislation on the basis that it is in their view quite consistent with the Bill of Rights. That does happen. What seems wrong to me is when Members do not engage with it, or enact legislation even admitting that it is inconsistent with the Bill of Rights. That has happened also.

It seems to me that the former approach is entirely legitimate, if done honestly and conscientiously. After all, the standard that Parliament has set itself is to legislate consistently with the Bill of Rights—not with what the Attorney-General says about the Bill of Rights. Members must be free to reach independent decisions.

But that said, there are cases in which the Attorney-General’s advice, set out with legal analysis and reasons, is frankly more compelling than members’ assertions that they assess the balance between rights and social interests differently. Such assertions can be easy to make. If made too easily the Bill of Rights will fail in its mission to protect the rights of unpopular minorities when threatened by popular causes. Convicted offenders against children are probably society’s most unpopular people, along with suspected murderers, but the Bill of Rights is intended to protect them too.

So the message is to carefully consider the Bill of Rights when developing policy and promoting legislation. The Bill of Rights demands justification under s 5: that if rights are to be impaired, the limits upon them must be no more than is reasonable and demonstrably justified in a free and democratic society. These are concepts that have their counterpart on the constitutional jurisprudence of most western nations. It is a question of balancing the reasons for a law against the impact that it makes on a protected right or freedom.

There should never be, in policy development and law-making, an occasion for relying on s 4 of the Bill of Rights to “protect” inconsistent legislation. To repeat, s 4 is there simply to mark the fact that a judicial opinion as to a measure’s inconsistency (after trying to interpret it consistently) cannot lead to a refusal by a court to apply the law. That protects parliament’s own views on the measure’s consistency.  For the Bill of Rights to work in that way, the onus is on legislators and policy makers to responsibly engage with the Bill of Rights and its methodology. They must keep within the “navigation lights” it charts out. They must strive to act consistent legislation.

2. Retrospective legislation and interference with existing or completed court proceedings

The starting point is that legislation is supposed to be prospective. A law ought generally to apply from the time it comes into force. It is often said that a new law ought not to affect “existing rights”.

This reflects a general understanding that it would be unfair for the law to attach, to events that have already happened, consequences that differ from those arising under the law at that earlier time.

The Interpretation Act 1999 sets out this basic principle in s 7: “An enactment does not have retrospective effect.” But, as Burrows and Carter point out in their excellent discussion of this topic, the Interpretation Act elsewhere recognises (in s 4(1)(b)) that this general principle will not apply where the “context of the enactment requires a different interpretation”. [That is, where the Act under interpretation is in fact intended to apply retrospectively.]

So that leaves us with the question for policy makers and law drafters: when is it appropriate to make a law that has retrospective effect?

The answer starts with a recognition that there is in fact a continuum between purely prospective law on the one hand, and retrospective law on the other. Even a purely prospective law might operate in a manner that impacts upon persons differently according to choices they made in the past (eg laws abolishing school zones might affect land values). The legislative judgment in that example—that school zones ought to be changed or abolished—would most probably leave no room for protecting the “rights” of persons who had made purchases relying on the zones. That is, it would be hard to protect their interests in a regime with no zones at all. The reasons in favour of a new regime are also a reason for that modest degree of retrospectivity. These are matters for political judgment.

So the principle against retrospectivity is not a self-executing one. Varying degrees of retrospectivity may in fact be called for, according to the demands of the subject matter. e depends on why it is being done. A typical context is when litigation has exposed a “defect”, from Government’s point of view, in legislation. A court might have reached a view of what an enactment means that is at odds with the view taken by Government. Or there might be proceedings in foot in which such an outcome, previously unanticipated, now seems likely.

In this situation, a starting point is that in most situations it cannot possibly be illegitimate for Parliament to change the law for the future. The law cannot always be frozen in the form in which it is pronounced by a court. So the critical question becomes not whether the law can be changed but what to do about “vested interests”. Plainly, the party that won the case has such an interest. Other litigants intending to make the same point but whose cases have been filed but not yet heard may also have such an interest. Then there are those who could bring such a case, but have not yet done so, to think about as well.

The whole question has been the subject of a LAC paper (Report No 9, June 1996. available at http://www2.justice.govt.nz/lac/pubs/1995/legislation/chapter_6.html(external link)

As well, of course, as being addressed by Burrows and Carter.

Essentially, to protect existing rights under the law there is often a need for transitional arrangements in a new law. Some allowance can and often should be made for persons who have relied on the earlier law, by say, filing patent applications or court proceedings, or taking some other steps. And, of course, it is the same when legislation overturns a court judgment. Those who have won their cases, and whose win has prompted the Government’s desire to change the law, have a strong prima facie case that they should be allowed to keep their litigation victory.

But it must depend on the facts. There is a continuum. At one end a change in the law with retrospective effect is plainly unjust, as when acts already completed are made criminal (and were manifestly not so at the time).

At the other end of the spectrum are cases where a law is given retrospective effect so as to affirm the position that all would reasonably have assumed to be the case, but which concluded (or even threatened or pending) litigation has demonstrated not to be the case. A New Zealand example is where it is discovered that a District Court had not been designated as a place at which jury trials could be conducted, yet many trials had in fact been held there.

In a useful article on the subject of retrospective legislation, Sampford and Palmer “Retroactive Legislation in Australia: Looking Back at the 1980s” (1994) 23 Federal Law Review 217 discuss the factors that re relevant to determining the acceptability of retrospective legislation.

These factors are “justice” and “reliance”. Arguments in terms of justice weigh heavily against retroactive criminal penalisation. But the retroactive validation of technical invalidities (say in an officer’s appointment, or a court’s designation) may often serve justice and may be a benign form of retroactivity.  And this might be so even when it serves to prevent highly technical appeals against findings of substantive guilt. There have been instances of this sort of retroactivity in New Zealand, for example where a technical error in the Summary Proceedings Act meant that District Court judges had been hearing certain offences even though they were not listed in the schedule that permitted such hearings.

So justice must be assessed not only from the point of the view of the person affected but also from the wider public interest.

Reliance, for its part, also works both ways. For example, in the case of law that validates regulations held to be invalid, it will often be the case that government relied on the validity of the enactment and would have validated it earlier if it had known of the invalidity.

The real question becomes whether the litigant who has exposed a “defect” that Parliament wishes to correct should be able to keep his or her litigation victory. There seems to be a strong principle in favour of this—of allowing the victory to be kept. But this is subject to the consideration that sometimes the reason for wanting to change the law for the future also serves as a reason for needing to reverse the result of the litigation.

Two examples to illustrate: In Lesa v Attorney-General which found that tens of thousands of Western Samoan citizens also had New Zealand citizenship and so could come to live in New Zealand , it was possible to reverse that by legislation, while preserving the victory of the handful of people affected by the judgment.

But where the judgment itself produces a state of affairs that could not be preserved if the new amending law is to make any sense or have any utility, then the litigation victory probably cannot be preserved. That is, it may be the specific ruling that Parliament wishes to overturn, as well as future possibilities. Sampford gives the example of reversing an unexpected interpretation of a customs law that allowed dangerous weapons to be imported; the reasons in favour of changing the law for the future also supported changing it for the litigants concerned. It was not a victory that justice entitled them to keep.

Litigation that interferes with pending court proceedings (as opposed to concluded ones) is amenable to the same sort of inquiry.

The Clyde Dam legislation in 1986 might be seen in this category; as the then Government saw it, at least, changing the law governing the granting of water rights as had just been declared by the Planning Tribunal was not enough. It judged that the national interest required a reversal of the result in the case and, more, the granting of water rights. But the legislation provided that the parties to the litigation were entitled at least to their costs. In the end these are political controversies.

The fact that there can be no hard principle against legislating to end court cases is revealed in the case of the Sealords Maori Fisheries Settlement in 1992. The legislation that imposed the settlement deal on all Maori (admittedly with the consent of most Maori) had the effect of terminating then pending legal proceedings brought by many dissenting tribes. This was adjudged to be a benign and necessary by-product of a settlement that most agreed to be a just and fair one, achieved after proper consultation. (Subsequently the Human Rights Committee of the United Nations, hearing the complaint of a dissenting tribe brought under the OP procedure, declined to find that New Zealand had breached any ICCPR right in this regard.)

So in every case there is a need for evaluation of what is required in the interests of justice, and the extent to which there has been reliance that should not be upset.

Where it is possible to allow court victories to be kept the clear practice is to do so. Typically this is where a small number win a victory that could be followed by more litigants. The law can be changed to preclude the latter group, while preserving the victory of the former.

3. The Treaty of Waitangi

As we all know, the Treaty of Waitangi is a foundational document in the sense that it made the colonisation of New Zealand possible.

But from the English perspective, it was not designed to be the law of New Zealand, nor even to be the basis for law in New Zealand (in the sense of being a constitution).

Rather, the Colonial Office view seems to have been that the Treaty recorded the Maori signatories agreement to the cession of sovereignty. Articles 2 and 3 then essentially affirmed what the effect would be once English common law applied: that Maori would still own what they owned (essentially, land, so long as they wished to retain it) and Maori and European would be treated equally. Seen in this way, English law delivered the Treaty promises in a legal system of English-style institutions and English law concepts.

As we now know, it was not that simple—not least because the Maori version of the Treaty reads somewhat differently. As it happens, those differences capture well the conflicting aspirations that were inherent in the transaction at Waitangi and elsewhere—not just technical conflicts in the wording of the document but in the very phenomenon of one state asserting power in the territory of another, powerful, people. The same tensions are in play in Canada, the US and Australia, even when there is no analogous treaty. But the Treaty of Waitangi in Maori captures the essence of the conflict. First, because Article 1 reads (arguably) as a delegated right for the Crown to govern. Second, because article 2 seems to state a continuing Maori sovereignty over all that which Maori hold to be important (not just tangibles and things that can be owned in English law).

We have seen this controversy played out over the last three decades and of course its happening as we speak today in relation to the question of water rights.

But what does this mean for our topic today? Is the Treaty of Waitangi an external, constitutional constraint on Parliament?

Not in a technical legal sense: and of course this is what the PM is voicing when he says that the Waitangi Tribunal need not be followed on water ownership.

But I find it useful to identify three levels at which Treaty of Waitangi issues are aired as a matter of law:

First, the Treaty as a constitutional imperative. What does it mean for how we design our systems of government? Realistically, our New Zealand Constitution has reflected the English one, even down to the point as saying the treaties are not operative in the legal system unless made so by an Act of parliament.

We are seeing this constitutional question, of course, in the current “Review” that is to be undertaken by a committee chaired by Sir Tipene O’Regan and John Burrows. There the question will be, because we have a Treaty, what should the Constitution be like?

Second, the Treaty as a political imperative. Given the Treaty of Waitangi and the fact of this being a colony in which an indigenous race was present, what sorts of policies should be pursued, what sorts of laws should be passed in pursuance of those policies? This is where the action is. It is deciding what the Treaty means for us now. In this field, the use of the concept of “the principles of the Treaty” has liberated the Treaty so that it can be applied to modern circumstances that almost certainly were not envisaged.

In recent years our legislation has taken one of several pathways on Treaty related issues.

In 1986 the State-owned Enterprises Act 1986 paved the way with the Delphic phrase “Nothing in this Act shall permit the Crown to act inconsistently with the principles of the Treaty of Waitangi”. And in the famous Maori Council case, the Crown was held to act inconsistently when it transferred land to SOEs without a mechanism to secure it back to give redress to Maori in Treaty claims.

The phrase “principles of the Treaty” was used in other legislation  in various ways—not always as a positive constraint on Crown action as it was in the SOE Act. Sometimes it was expressed as a mandatory consideration to which regard must be had, but not as a substantive limit on “outcomes”.

Other Treaty clause styles followed, and Matthew Palmer for his part urged that it was better to determine more precisely what legislation ought to be like in order to have regard to Maori interests, rather than using the vague and general phrase about “Treaty principles”.

Indeed, we had seen this as early as 1989 when the Education Act did not use the general “Treaty” phrase, but instead gave concrete iterations of what Treaty-recognising was taken to mean in education: making Kura kaupapa schools a possibility, inserting a compulsory Charter provision about Maori culture into all school charters, for example (although in one respect—requiring tertiary councils to “acknowledge the principle of the Treaty of Waitangi”—it followed the 1987 model).

RMA goes a bit further in incorporating more specifically the range of Maori interests to be recognised in planning applications and the like.

The message, I believe is this: don’t lob a treaty clause in, but work out what the Maori dimension requires, if it requires anything at all. I recall there were political difficulties in placing a Treaty clause in the 2000 Public Health Act, the view being that health was not a context in which race differences ought to be salient, or at least that, if they were, it was not a Treaty imperative so much as a multicultural one.

For my particular topic today the salient point is this: is the Treaty of Waitangi a restraint on Parliament’s law making power?

I think the answer to that is similar to the one given as to fundamental principles such as liberty: it is not a legal constraint that courts can enforce but it is a legitimate source of imperatives to be taken into account in the development of policy. But in the end Parliament is the judge of whether it has done enough. The court’s role will be one of interpretation of the legislation, against the backdrop of rights that, through often intense political negotiation, Maori are held to have. That was the pattern in the Ngati Apa case about the Seabed and Foreshore—it puts the focus on the next level of inquiry about the Treaty of Waitangi: the Treaty in the Courts.

There the Treaty has not been regarded as a fetter on the legislature, but it has generated principles of interpretation and process. It is noticeable how process-oriented the Treaty principles have become: that they are about good faith and consultation in matters affecting Maori and do not speak to specific outcomes.

It was significant that in the aftermath of the Sealord litigation—when it went to the Human Rights Committee of the UN who had to decide whether the Treaty rights of dissenting Maori tribes were denied (contrary to article 27 of ICCPR, cf s 20 of the BORA, guaranteeing the right of an ethnic minority to enjoy its—in this case, fishing—culture). Referring to its jurisprudence form cases involving Nordic indigenous peoples, the Committee ruled in favor of New Zealand precisely because the proposal to make the Sealord Settlement had been the subject of a prolonged and effective consultation process, and that the Waitangi Tribunal had given its generally supportive view as to the substance of the settlement.

That raises the issue of the place of the Waitangi Tribunal in the scheme of things. It takes its place at the political level: it is not a court that makes binding rulings as we know, and so is a place for a considered and authoritative opinion as to the merits of matters that are being settled politically and ultimately, usually, by legislation.

By the time Treaty of Waitangi matters get to the courts, the law-making process ought to narrow the issues down (but probably won’t if the broad and vague Treaty principles phrase is used as opposed to a legislated statement of what the Treaty actually ought to mean for the legislation in issue).

In the Courts, the options are to regard Treaty principles as a relevant, even if unstated,  consideration when statutory decision-makers exercise their powers in a manner qaffecting Maori (school closure cases often take this approach), or to prefer Treaty consistent meanings of vague phrases (few cases are in this category in fact, but the Huakina case in 1986 was a formative one when it held that the phrase “public interest” necessitates a separate consideration of Maori values).

Obviously legislation has to be drafted against the background that these sorts of interpretive approached will be taken.

But most importantly, the evolved practice in New Zealand is to determine politically (through consultation if not contestation) what the Maori dimension of any policy proposal ought to be, and to seek to enact that into legislation.

4. Separation of powers

A number of the points already made can be restated as an aspect of the separation of powers.

The point here is that legislatures have the function of enacting laws, while courts have the function of adjudicating in civil and criminal disputes and must, where Parliament’s law is involved, apply it (but must of course first interpret it).

So, then, pernicious retroactive legislation, of the kind that criminalises actions that have already occurred, is really tantamount to judging, not legislating. When legislation speaks to the past and to things that have already happened, then the class of perpetrators is a fixed and quite possibly a known class. To say that what this class did was a crime is very close to a legislature passing itself off as a court and passing judgment upon them. It is not absolutely the same: a court would still have to inquire if the person before them is the one that did the act now made criminal. But it is close, and part of the injustice that is generally felt about retroactive legislation is precisely the fact that it is Parliament rendering a judgment on what has happened, rather then legislating for the future when anything might happen.

A similar constraint exists for courts. They are there to adjudicate disputes. Where this involves the need to interpret statutes, they must recognise the need to stay within the language used by parliament and not to make the enactment mean something it cannot possibly mean. There is a line, in other words, between interpretation and legislation. This line was the subject of debate inR v Hansen, when it was urged upon the Supreme Court that the Bill of Rights legitimated the ascribing of meanings to words that went beyond the apparently intended meaning, and indeed required ascribing consistent meanings even if they were strained.

The Court resisted that, and the separation of powers motif is a part of the explanation.

Similarly, when the Court of Appeal in the Sealords litigation was asked by the dissenting Maori tribes to injunct the Government from even introducing the Fisheries Settlement Bill into the House, the court ruled that it could not do this. It had to respect Parliament’s right to freedom in what it considered passing: for a court to prevent a bill being introduced, when a bill might be passed into law, was inconsistent with the separation of powers, and with each branch of government being left alone to do its allotted job.

Clauses that seek to limit or preclude access to courts for judicial review (of the lawfulness of public actions) may similarly offend the separation of powers principle.

5. The need to reckon with the fact that courts will ultimately construe statutes

In the end statutes are just words on paper. It falls to others to decide what they mean. First and foremost, it falls to the executive branch in its mission of carrying out and enforcing the laws. Then it falls also to citizens and their lawyers. Courts become involved only when there is a dispute but, then, their word is authoritative (subject of course to a legislative countermanding).

Plainly the making of quality legislation must take account of judicial approaches to interpretation.

In the remaining time, let me deal with just one interpretive phenomenon—the reading down of general powers to accommodate fundamental rights.

In the United Kingdom the so-called “principle of legality” holds that general powers must be read subject to fundamental human rights, save where these are clearly overridden. This is essentially a recognition of a basic feature of human communication—that much of what we say is pregnant with the necessary qualifications that reconcile out utterances with deep human values. So the command “come immediately” when uttered by an employer to employee will not (in any context where this matters) readily be taken to mean “and ensure you do not stop to rescue a drowning child on your way”.

So it is that the United Kingdom courts have held that a Prisoner Superintendent’s statutory  to make rules for prisons could not be construed to empower even fundamental rights-infringing applications of that power.

Moving to the New Zealand setting, our Court of Appeal has held without difficulty that a regulation-making power in the Prisons Act could not authorise a rights-infringing regulation. At that level, then, the Bill of Rights operates just as it does in countries where it is supreme law. It can invalidate secondary and tertiary legislation. Only where a statute clearly empowers the making of rights infringing rules or regulations, or decisions, will it be held to oust the Bill of Rights. And instances of this are likely to be very rare, as one would expect and hope.

What are the implications for law making? First, there is likely to be no need to seek to oust this principle. The idea that a power is to extend to even rights infringing applications (that is, contrary to that which is demonstrably justified in a free and democratic society) is obviously unattractive.

But second, I have seen some evidence (in Bill of Rights opinions in the s 7 context) of the point being inverted. That is, it is sometimes said of a very broad power (say of search and seizure) that it can be regarded as consistent with the BORA because of the interpretive principle that will lead a court to regard it as authorising only reasonable searches. That is true enough, but the difficult with this style of drafting and reasoning is that the statute book will then contain apparently untrammelled powers. While a court might invalidate unreasonable uses after the event, it is better if the statute descends to detail of what is required for reasonable searches to occur. That is, if it outlines matters of process and grounds.

The aim is not just for a law that can be interpreted by expert judges to rights consistent results. It is also to have a law that will be, in fact interpreted and applied by those in the field, who need the ore specific guidance of detailed provisions.

So if you find yourself answering a criticism of a particular clause by saying, “but a court would never interpret it that way”, it may be appropriate to think again. You may be right about what a court would do, yet it may still be the best option to include more detail into the law, so as to guide its users to reasonable outcomes.


[1]   Recall that the Attorney-General’s obligation to report on inconsistent provisions arises once only, upon introduction of the Bill. This means that when inconsistent provisions are added at committee stages or by way of SOPs there is no express requirement for a report by the Attorney-General. There have been repeated calls for a reform of process to ensure that s 7 reports are made in these situations, but the Attorney-General has so far not been receptive.

This page was last modified on