Canadians have long been used to the idea that, as the Supreme Court put it in Re Manitoba Language Rights,  1 SCR 721, “[t]he judiciary is the institution charged with the duty of ensuring that the government complies with the Constitution.” (745) In New Zealand, things are very different of course, because the constitution is not entrenched. Parliamentary sovereignty prevails, and the courts’ role is limited accordingly. Although there is a statutory bill of rights, the New Zealand Bill of Rights Act 1990, it specifically provides that courts cannot invalidate or otherwise refuse to apply legislation that is inconsistent with it, and contains no remedial provision analogous to section 24 of the Canadian Charter of Rights and Freedoms.
So it is, or at perhaps was, an open question what, if anything, a court might be able to do when it concludes that a statute is inconsistent with the Bill of Rights Act. Might it go so far as to issue a declaration to that effect, or is it limited to only stating this opinion in the course of its reasons? In Attorney-General v Taylor  NZCA 215, the New Zealand Court of Appeal says that, sometimes at least, a formal declaration can be made, and upholds the very first such declaration issued by a New Zealand court, confirming that the disenfranchisement of all convicted prisoners (and not only of those serving sentences longer than the three-year Parliamentary term) is inconsistent with the Bill of Rights Act’s guarantee of the right to vote. (The Attorney-General was not contesting the substantive point, it is worth noting, but only disputing that the declaration could and should have been made.)
* * *
The first question for the Court was whether authority to make a “declaration of inconsistency” existed at all and, if so, what its source was. The answer, the Court holds, is that superior courts have such an authority as part of their jurisdiction to answer questions of law, and that the Bill of Rights Act supported it. The Court rejects the Attorney-General’s submission that express statutory authorization is required to permit the making of declarations of inconsistency. Just as Parliament’s legislative authority does not derive from positive law but from political fact, so does the judicial authority of the courts. Neither branch owes its authority to the other; rather, “a distribution of the state’s sovereign powers among the branches of government emerged from the political settlement concluded in the decades following the Glorious Revolution of 1688”.  Ultimately, “[i]nconsistency between statutes is a question of interpretation, and hence of law, and it lies within the province of the courts.” 
The Court notes that the Bill of Rights Act itself contemplates the possibility of a judicial assessment of the consistency of other legislation with its provisions, whether its results are stated in the court’s reasons (which the Attorney-General accepted was permissible) or in a formal declaration. Moreover, New Zealand has undertaken to provide domestic remedies for breaches of its obligations under the International Covenant on Civil and Political Rights, the implementation of which is one of the Bill of Rights Act’s stated purposes, so that it should be interpreted in accordance with this undertaking. Besides, in the Human Rights Act 1993, Parliament has already authorized the making of declarations of inconsistency when legislation breaches equality rights. Although the Bill of Rights Act contains no equivalent provision, this “evidences parliamentary acceptance that a court may make declarations about the inconsistency of legislation with rights protected by the Bill of Rights”. 
Second, the Court had to address an intervention by the Speaker of the House of Representatives, who argued that the whole case, or at least the way in which it had proceeded, was an infringement of Parliamentary privilege. In particular, the Speaker was concerned by the reliance, at first instance, on a report prepared by the Attorney-General to alert the House of Representatives of the incompatibility (in the Attorney-General’s opinion) between the blanket disenfranchisement of prisoners and the protection of the right to vote in the Bill of Rights Act. Indeed he sought sought to prevent the use of any “speeches in the House, select committee reports or submissions made to select committees”  to ascertain the consistency of legislation with the Bill of Rights Act, arguing that this would be tantamount to calling Parliamentary proceedings into question contrary to article 9 of the Bill of Rights 1688 and the Parliamentary Privilege Act 2014. The Court rejected these arguments, holding that “a court does not impeach parliamentary proceedings merely by describing parliamentary processes or making a finding about the same subject matter,”  so long as it does “not endorse or criticise Parliament’s treatment of the issues”.  It is permissible, too, to refer to the Attorney-General’s report, although it is important for the courts to come to their own, independent conclusions.
Third, the Court considered the conditions in which declarations of inconsistency should or should not be granted. Such declarations, thought they do not affect anyone’s rights, are part of a “dialogue” (it might have been more accurate to say “conversation”) involving the the different branches of government, which
is not unique to constitutional disputes. It describes the routine work of government, in which Parliament legislates and the executive administers and courts interpret, leading in due course to legislative reform to better meet the community’s evolving needs. 
The only difference is that a declaration of inconsistency is a “more pointed”  than usual expression of a court’s opinion, which carries with it
the reasonable expectation that other branches of government, respecting the judicial function, will respond by reappraising the legislation and making any changes that are thought appropriate. 
Such “pointed” expressions of judicial opinion should not be made lightly; a statement in the court’s reasons for judgment is sometimes, and even “ordinarily”  preferable. But it is sometimes necessary to go further. However, the courts are to apply fairly strict criteria for standing (at least when compared with the Canadian “open bar” approach), and to ensure that there exists a real adversarial dispute and that they have the relevant evidence available to them before pronouncing on the rights-consistency of legislation.
Fourth and last, the Court asks itself whether a declaration should have been granted in this case. It concludes that because “[t]he undiscriminating limitation … on so central a right demanded justification [and] [n]one was forthcoming” a declaration of inconsistency “was the appropriate way both to convey the Court’s firm opinion that the legislation needs reconsidering and to vindicate the right”. 
* * *
To Canadian readers this all might seem like pretty tame stuff. And indeed there is no mistaking the notes of caution in the Court’s discussion, above all in its statement that “indications” rather than formal declarations of inconsistency should “ordinarily” suffice. What “ordinarily” will mean in practice remains, of course, to be seen, but at least for now the Court seems to think the step of granting a formal remedy ― even one that could produce no more than a purely symbolic effect ― is a serious, even an exceptional one.
Yet I think it would be a mistake to make light of the Court’s decision and, perhaps more importantly, of its reasoning. Although its conclusions are cautious, it still reflects a confident view of the judiciary’s constitutional position as a branch of government that is, in its own sphere, not Parliament’s subordinate, but its equal. It is worth noting that the primary ground on which the Court rests the authority to make declarations of inconsistency is not an implication from the text or nature of the Bill of Rights Act (as it had done in Simpson v Attorney-General  3 NZLR 667 (CA), a.k.a. Baigent’s Case, where it held that damages were available for breaches of the Act by the executive). Rather, the authority to make declarations of inconsistency is said to come from the judiciary’s own inherent powers, which the Court goes out of its way to say are not the product of any legislation but of the constitutional order of things (my phrase, not the Court’s). Similarly, the Court resists the Speaker’s attempt to restrict judicial discussion of Bill of Rights Act issues, even as it cautions that judges must be seen to interfere with the deliberations of Parliament.
Indeed, this case can be seen as a clash between two competing constitutional visions. One, advanced both by the Attorney-General and the Speaker, sets Parliament, protected by its sovereignty and privilege, above the other branches of government, whose first concern must be to avoid disrespecting or challenging it. The other, which the Court adopts, treats the branches as (almost) co-equal: “each is sovereign within its sphere of authority in the sense that it may act without the permission or authority of the others”. To be sure, Parliament is first among equals because it can make law, and thereby oust judicial power (though New Zealand judges, as their British counterparts, have on occasion mused about the limits of that authority) or, in other cases, royal prerogative. But at least until it does so equality, not subordination, is the rule. It is a respectful equality, but respect goes both ways: not only must the courts exercise restraint and show comity on appropriate occasions, but Parliament too ought to engage in constitutional dialogue, and go so far as to reconsider its enactments, when called upon to do so by the courts.
Yet I am quite skeptical about the potential for constitutional dialogue between the judiciary and Parliament, on which the Taylor Court rests such hopes. We know that in Canada the “dialogue” has turned out to be quite one-sided, with the Supreme Court telling Parliament what it had, and what it could not, do. As the majority put in Sauvé v. Canada (Chief Electoral Officer), 2002 SCC 68,  3 SCR 519, another prisoner disenfranchisement case,
the fact that the challenged denial of the right to vote followed judicial rejection of an even more comprehensive denial, does not mean that the Court should defer to Parliament as part of a “dialogue”. Parliament must ensure that whatever law it passes, at whatever stage of the process, conforms to the Constitution. The healthy and important promotion of a dialogue between the legislature and the courts should not be debased to a rule of “if at first you don’t succeed, try, try again.” 
The power dynamics in New Zealand are, of course, the opposite of those in Canada. It is Parliament, not the judiciary, that gets to have the last word in a constitutional conversation. But I do not expect it to be any more open to persuasion than the Supreme Court of Canada. I would love to be proven wrong on this, but I’d be quite surprised if ― assuming there is no change of government at the forthcoming election ― New Zealand’s Parliament chose to “reconsider and vindicate the right” to vote as the Bill of Rights Act, which it was happy to ignore on this issue, requires it to do.
* * *
Subject to an intervention by the Supreme Court, the courts of New Zealand do, then, have the ability to formally declare legislation to be inconsistent with the New Zealand Bill of Rights Act, despite the Act not authorizing them to do so. This authority rests on a conception of the constitution in which the branches of government are almost, if not quite, equal, rather than Parliament lording it over the courts (and the executive). Yet there is reason for skepticism about the vision of respectful dialogue between Parliament and the courts that this relative equality is supposed to foster. Someone gets to have the last word, and it seems likely enough that, in New Zealand as in Canada, it will be the only that will count.
NOTE: See also the comments by Andrew Geddis, on Pundit, and Edward Willis, on his Great Government blog.