The appeal is dismissed.
Mr Siemer must surrender at the Registry of the High Court at Auckland at 9.00 am on Monday 15 July 2013.
[*] Chambers J died before this judgment was delivered. The remaining Judges have decided under s 30(1) of the Supreme Court Act 2003 to continue the proceeding to judgment.
The “great coercive powers of proceedings for contempt” are common law jurisdiction possessed by courts to punish, including by imprisonment, conduct which risks undermining the administration of justice. Although the circumstances in which contempt may be found vary, a commonly recurring basis is knowing breach of a court order. Such orders may comprise substantive final or interlocutory relief in proceedings or orders ancillary to the exercise of substantive jurisdiction made under the inherent power of the court to control its processes. The common law power to punish for contempt is distinct from the statutory jurisdiction under s 138 of the Criminal Justice Act 1985, also protective of the administration of justice, to punish by a fine of up to $1,000 breach of court orders authorised under that section.
Vincent Ross Siemer appeals the decision of the Court of Appeal dismissing his appeal from an order made by a Full Court of the High Court committing him to prison for six weeks under the common law contempt jurisdiction. The contempt was treated as consisting in breach of a suppression order by publication of a pre-trial judgment of 9 December 2010 made by Winkelmann J in the High Court. The judgment ordered severance of criminal trials in respect of three defendants jointly charged with 15 others and specified trial by judge alone for those defendants whose trials were not severed. The suppression orders were made by Winkelmann J against the world, purporting to bind non-parties.
It was not in dispute that Mr Siemer, who was not a party to the criminal proceedings in which the suppression order was made, published the judgment on his website with knowledge of the order. He also published commentary on the judgment which breached the terms of the order. The order was not only placed as a banner on the judgment to which Mr Siemer’s own article was linked electronically, but he had altered the banner to purport to exempt publication on his website. In the High Court committal proceedings it was acknowledged by Mr Siemer that he knew of the suppression order. He sought, however, to defend the proceedings on the basis that the order was not lawfully made. Whether the Judge had power to make the suppression order against non-parties and whether it was necessary to protect fair trial rights (thereby justifying limitation of Mr Siemer’s right to freedom of expression) remain matters of controversy on the appeal to this Court.
The arguments on this appeal invoke rights recognised by the New Zealand Bill of Rights Act 1990 to freedom of expression, fair trial, and natural justice. The principal and threshold question raised by the appeal is, however, whether the appellant was able to question the legality of the suppression orders in the contempt proceedings. In the High Court and in the Court of Appeal it has been held that such challenge is barred by a “rule” against collateral challenge, at least if the court which made the order had jurisdiction to make an order of the sort made. It was accepted in both Courts and is acknowledged by counsel for the Solicitor-General on appeal to this Court that such “jurisdictional” exception to the general rule against collateral challenge would apply if any common law inherent power to make a suppression order is excluded by s 138 of the Criminal Justice Act or if the common law power to issue suppression orders does not extend to orders against the world but only to orders against parties to the litigation in which the order is made. The first suggested exception turns on the interpretation of s 138. The second requires consideration of whether the view that there is no common law power to make suppression orders against non-parties (taken by the Privy Council in Independent Publishing Co Ltd v Attorney-General of Trinidad and Tobago) should be preferred to the view that there is such power (earlier taken in New Zealand by the Court of Appeal in Taylor v Attorney-General).
In the present case, the Court of Appeal, dismissing the appeal from the Full High Court finding of contempt, held that s 138 of the Criminal Justice Act does not exclude the inherent power of the High Court to make non-party suppression orders. It applied the approach of the Court of Appeal in Taylor in preference to that of the Privy Council in Independent Publishing in holding that the Judge had inherent common law power to make orders against the world prohibiting publication. The Court of Appeal also affirmed the view taken by the High Court that the correctness in law of the suppression orders could not be challenged “collaterally” in the contempt proceedings. As a result, it declined to entertain the further challenges to the order, including the argument that the order constituted an unreasonable restriction of the appellant’s freedom of speech, contrary to s 14 of the New Zealand Bill of Rights Act. On the approach taken by the Court of Appeal, the contempt was complete on publication in knowing breach of an order made with jurisdiction. It held that other grounds of challenge could not be raised in the contempt proceedings but could be put forward only in proceedings which directly challenged the validity of the order and that such direct challenge could not be made following the breach which constituted the contempt.
For the reasons explained in what follows, I am of the view that s 138 of the Criminal Justice Act excludes common law powers to make suppression orders of the type made here. Breach of orders made under s 138 may be punished by fine. It follows that I would allow the appeal on the basis that the sentence imposed was one the High Court had no jurisdiction to impose. Since that is, however, a minority view, it is necessary for me to explain why I also consider that, if the common law power to make the non-publication order here made has not been supplanted by s 138 and extends to orders against non-parties (as I conclude it does for reasons given in  to ), the appellant should succeed in the further arguments that he was able to challenge the legality of the suppression order in the contempt proceedings. While I come to the conclusion that orders against the world may be made to protect the administration of justice, that conclusion is linked to and dependent upon the view I take that someone who is not a party to the proceedings in which the orders are made must have the opportunity, if proceeded against for contempt for breach, to question their legality in the contempt proceedings, including on the grounds that they constitute an unreasonable restriction on the right to freedom of expression.
It is not necessary to decide whether similar challenge can be made by way of a defence to an order made under s 138(2)(a) or (b). That matter is not currently before us. It turns principally on construction of the legislation creating the statutory offence, including the relatively limited penalty under it which may be contrasted with the open-ended power to punish at common law.
I accept entirely that orders of the court must be obeyed. And it will almost inevitably be abuse of process justifying refusal to entertain a defensive challenge for a party to litigation to seek to defend proceedings for committal for contempt for breach of adjudicated orders by attempting to re-litigate the decision to make the order. Nor can it be a defence in proceedings for contempt that the person breaching the order himself believes it to be unlawful and void. He is not entitled to act on his own judgment as to its validity, which may well be wrong. But it entails no retreat from the position that court orders must be obeyed to take the view that someone who is not a party to an order may raise its legality in proceedings to commit him for contempt for disobedience.
That is not the position taken by the majority in the present appeal. They consider that non-jurisdictional challenge cannot be raised in proceedings to commit someone in breach for contempt. Such challenge must rather be made pre-breach either in proceedings for judicial review (a procedure available only in relation to orders made by an inferior court) or by application to the court which made the orders (either the High Court or an inferior court) for reconsideration and variation of the orders. In respect of “jurisdictional” challenges, the Judges in the majority reserve their position on whether such challenge can be entertained in the contempt proceedings, while indicating that in any event they would reject the two jurisdictional challenges conceded by the Crown to be available to the defendant in the contempt proceedings: exclusion of the inherent power by s 138 and the claim that there was no inherent power to make orders against non-parties.
The suppressed judgment was one of a series of interlocutory pre-trial rulings made by Winkelmann J in what was a complicated and major case involving a large number of accused. The case had been the subject of considerable publicity and public anxiety. Previous rulings in the proceedings seem to have all been subject to non-publication orders.
The orders in issue were contained in a banner at the top of the judgment of 9 December 2010 which read:
THIS JUDGMENT IS NOT TO BE PUBLISHED (INCLUDING ANY COMMENTARY, SUMMARY OR DESCRIPTION OF IT) IN NEWS MEDIA OR ON INTERNET OR OTHER PUBLICLY ACCESSIBLE DATABASE OR OTHERWISE DISSEMINATED TO THE PUBLIC UNTIL FINAL DISPOSITION OF TRIAL OR FURTHER ORDER OF THE COURT. PUBLICATION IN LAW REPORT OR LAW DIGEST IS PERMITTED.
No reason for imposing this restriction on publication was given in the body of the judgment itself, which made no reference to the suppression orders. That was consistent with the approach the Judge had taken in earlier pre-trial rulings in the same proceedings. In an exception to this approach, in a judgment delivered on 23 April 2010, the Judge in her reasons said simply, “I make the standard orders suppressing publication of the contents of this judgment.”
It is clear from the context, as the Full Court held, that the suppression orders were made to protect the right of the accused to fair trial. The Full Court considered that the reason that the orders were made was “self-evident, namely that publication of some or all of the material contained in the pre-trial ruling may prejudice the right of any accused to a fair trial” because it could “contain material that will be inadmissible at trial and which should not be publicised to the potential pool of jurors”. Such orders are commonly made for that purpose in pre-trial judgments dealing with the admissibility of evidence and the procedure to be followed at trial. As is consistent with the purpose of protection of fair trial, the orders in the present case were expressed to continue only “until final disposition of trial or further order of the Court”. Mr Siemer’s website commentary itself acknowledged that the reason for suppression orders in the earlier pre-trial rulings had been stated by Winkelmann J as being “to ensure the jury pool is not prejudiced by pre-trial information”.
The orders made prevented publication not only of the reasons why judge-alone trial instead of trial by jury was ordered in the case of some of the accused and why severance was granted to three of the accused (which entailed consideration of the evidence, the Crown allegations, and the likely duration of the trial), but the fact that judge-alone trial and severance orders had been made at all. These were matters Mr Siemer described in a subsequent publication on 18 December 2010 (in which he reported that he had received notification that the Solicitor-General regarded his earlier publication as a breach of the suppression orders) as matters of “genuine public interest”. The Crown subsequently sought a variation on the basis that the orders were wider than was necessary to protect fair trial and that more limited suppression orders would be necessary only if the accused intended to appeal the mode of trial rulings made by Winkelmann J. Any relaxation of the suppression orders was however opposed by the accused. Winkelmann J thought it “prudent” in those circumstances to continue suppression except as to “that part of the judgment which declares the result”. Her minute recording the ruling referred to the nature of the trial and “the extent of publication in connection with it occurring outside the mainstream media”. The suppression orders were accordingly varied to permit publication of the result of the judgment of 9 December 2010, with effect from 21 December 2010.
After the judgment of the Full Court sentencing Mr Siemer to six weeks’ imprisonment was delivered on 4 July 2011 and after his appeal to the Court of Appeal had been filed, Mr Siemer tried on 25 August 2011 to apply for review of the suppression order in the High Court. The application seems to have been prompted by the view expressed in the Full Court decision that such procedure was available and should have been followed. The application was rejected by the Registrar, perhaps because the attempt to challenge the order directly at that late stage amounted in substance to a challenge to the finding of contempt, although the basis for rejection is not clear. Before then, but also following the determination of the contempt proceedings by the Full Court, another person, Ms Bright, had sought leave in the criminal proceedings themselves to bring an originating application to vary the suppression order. That application too had been rejected by the Registrar after Brewer J, to whom it had been referred, directed the Registry on 26 July 2011 “not to receive an originating application in this matter”. The minute in which the directions are given records:
Open justice, fair trial, freedom of expression, rights to natural justice and judicial responsibility
The arguments raised by the appeal are made against a background of human rights and general principles of justice. Those of principal significance and which bear both on the validity of the suppression orders made and the proper approach to the jurisdiction to punish for contempt include the principle of open justice, and rights to fair trial, freedom of expression, and natural justice.
The principle of open justice, long-recognised as essential to the rule of law and now acknowledged in respect of criminal procedure by s 25 of the New Zealand Bill of Rights Act, requires court proceedings to be conducted in public. Article 14 of the International Covenant on Civil and Political Rights, on which s 25 of the New Zealand Bill of Rights Act is based, permits exclusion of the press and the public for a number of reasons, including “to the extent strictly necessary in the opinion of the court in special circumstances where publicity would prejudice the interests of justice”. But it requires “any judgement rendered in a criminal case or in a suit at law [to] be made public except where the interest of juvenile persons otherwise requires or the proceedings concern matrimonial disputes or the guardianship of children”. Section 138 of the Criminal Justice Act, discussed at  to , is legislation protective of the principle of open justice (and in this respect less tolerant of exclusion of the public than the pre-existing common law it reforms). It enables the imposition of restrictions for limited purposes identified by the statute and in circumstances where strictly necessary for those purposes. The general rule adopted in s 138 is that “every sitting of any court dealing with any proceedings in respect of an offence shall be open to the public”.
A principal responsibility of the courts in securing the proper administration of justice is protection of the right to fair trial, itself also now affirmed by s 25 of the New Zealand Bill of Rights Act in fulfilment of the obligation under art 14 of the International Covenant on Civil and Political Rights. The requirement to ensure that trials are fair may be met by use of statutory powers but may require recourse to the ancillary inherent powers of the court to control its procedures and secure the proper administration of justice.
The right to fair trial is not qualified in the International Covenant on Civil and Political Rights. It is an absolute right, not a relative right which must be balanced against other rights and interests recognised by law. Where fair trial is risked through pre-trial processes being publicised, there may be justification for deferring publication and restricting freedom of speech, at least for the pre-trial period.
The right of “[e]veryone” under s 14 of the New Zealand Bill of Rights Act to freedom of expression, “including the freedom to seek, receive, and impart information and opinions of any kind in any form”, includes the right to impart information about court proceedings. Freedom of expression, unlike the right to fair trial, is a right that is qualified under the International Covenant on Civil and Political Rights. The justifications identified under the Covenant for restrictions on free speech include measures to protect the rights or reputations of others and the protection of national security, public order (ordre public), public health, or morals. Since the New Zealand Bill of Rights Act must be observed by the judiciary, a suppression order (whether under a statutory provision or under the inherent power of the court, where the inherent power is not excluded by statute) may lawfully be made by a court for fair trial purposes only where publication would create a real risk that the course of justice would be impeded or prejudiced. Rights to fair trial and freedom of expression are important requirements of law against which the legality of non-publication orders fall to be assessed.
In addition, the rights to natural justice affirmed in s 27 of the New Zealand Bill of Rights Act are also relevant to the questions raised by this appeal as to the power of a court to make orders against the world and the ability of an accused to attack the validity of such an order as a defence in contempt proceedings based on breach of the order. Section 27 makes it clear that every person whose “rights, obligations, or interests protected or recognised by law” are affected is entitled to the observance of the principles of natural justice. Moreover, if such rights, obligations, or interests are affected by the determination of any tribunal, such a person has the right “to apply, in accordance with law, for judicial review of that determination”.
The human rights to public hearing, freedom of expression, and natural justice may be subject to “such reasonable limits prescribed by law as can be demonstrably justified in a free and democratic society” under s 5 of the New Zealand Bill of Rights Act. General Comment No 34 of the Human Rights Committee of the United Nations, in referring to the restrictions which may legitimately be placed on freedom of speech, makes the point that the requirement of prescription “by law” “may include laws of parliamentary privilege and laws of contempt of court”. In order to be characterised as a law, a rule must be formulated with sufficient precision to enable an individual to regulate his or her conduct accordingly and it must be made accessible to the public. A limitation imposed by court order which is not “justified in a free and democratic society” under s 5 of the New Zealand Bill of Rights Act, is unlawful. Judges in exercising judicial authority are bound to observe the rights and freedoms contained in the New Zealand Bill of Rights Act. A limit which is not necessary in the particular case to meet a proper objective, such as to secure fair trial rights, may not lawfully be imposed by a court, even if its power to make orders of the sort in an appropriate case is not in doubt.
As is further explained at , the requirement that limitations be “prescribed by law” supports a power to make orders, rather than leaving conduct to be subject to penalty for contempt according to whether it is determined ex post facto to be calculated to interfere with the proper administration of justice (the basis of common law proceedings for contempt). The terms of s 5 of the New Zealand Bill of Rights Act are therefore significant in the reasons I consider that the common law as it was expressed to be in cases such as Attorney-General v Leveller Magazine and Independent Publishing is properly adapted in modern conditions to allow, consistently with the approach followed in New Zealand since Taylor, that the inherent powers of the court to protect the administration of justice extend to orders against non-parties where such orders are necessary. The consequential adaptation of the common law I would however recognise as necessary in relation to subsequent contempt proceedings for breach of such orders is that non-party orders must be able to be challenged for unlawfulness by way of defence in contempt proceedings. This conclusion is explained in  to .
Orders against the world
Although both rest on the authority of the court, there are differences between orders made in the course of litigation which are binding on the parties to that litigation and orders directed to the general public. These differences were pointed out by Woodhouse J in Taylor:
There is an obvious difference between the situation of strangers in relation to an order made at large and disobedience to a Court order resulting from litigation by one of the parties or by some person intermeddling. Not only is there usually a right to appeal but direct disobedience of the order inter partes could deprive the other party of the means of enforcing his rights. So proceedings brought for contempt are intended to support both the authority of the Court and the adjudicated claims of the party affected.
In Solicitor-General v Siemer, contempt proceedings were brought against Mr Siemer for his breach of an order which had been made by way of substantive relief in proceedings to which he was a party. They were proceedings in which Mr Siemer had rights of appeal against the order. His breach of it not only challenged the authority of the Court but deprived another party to the litigation of adjudicated rights. That was the background against which I joined with McGrath J in this Court in saying:
Effective administration of justice under our constitution requires that the orders of the courts are obeyed unless properly challenged or set aside. Public confidence in the administration of the law, also necessary for its effective administration, requires that there is a strong expectation that those who ignore court orders are quickly brought to account. Achieving these aims is part of the objective of the law of contempt.
Orders against the world are unusual. They are effectively judicial legislation. In considering the availability of challenge in contempt proceedings in  to  below I take the view that the reasoning in Boddington v Brish Transport Police applies equally to court orders in proceedings to which the defendant was not a party as it does to by-laws or other subordinate legislation. Just as it has been held contrary to the rule of law to condemn someone for infringement of a by-law without affording him effective opportunity to challenge to its validity in the enforcement proceedings, I consider it contrary to the rule of law to condemn a non-party for contempt for breach of a court order without affording him the opportunity to challenge it in the contempt proceedings.
It is necessary to express disagreement with the view that permitting the legality of a court order to be challenged in proceedings for contempt based on breach of the order is to countenance non-observance of court authority and undermine the constitutional position of the courts. Permitting such challenge does not mean that the order is something that may be observed or not. It does not mean that people are entitled to act on their own assessment of validity. It means that conduct contrary to the order is undertaken under peril of contempt proceedings. I do not understand the efficacy of such jeopardy to be different from other proceedings for enforcement. The fact that it is necessary to have recourse to enforcement does not mean that the rule infringed is something that may be observed or not. The risk that the administration of justice will be brought into disrepute by the punishment of someone for breach of an invalid order strikes me as just as real as the risk that the administration of justice will be brought into disrepute by the punishment of someone for breach of an invalid by-law. The former is perhaps more harmful to the standing of the court because in respect of such orders the court is both law-maker and enforcer. A rule which countenances such injustice is not calculated to promote respect for the rule of law, the basis on which the requirement to observe court orders rests.
Breach of court orders and the common law contempt jurisdiction
The focus of the case throughout has been on Winkelmann J’s power to make the suppression orders prohibiting publication of her judgment of 9 December 2010. The critical issues have been whether there is at common law a power to make such an order against the world and, if such a power does exist at common law, whether it has been ousted by s 138 of the Criminal Justice Act. I have some doubts as to whether this approach puts matters the right way around. It may be more illuminating to start with the natures of, respectively, the common law contempt jurisdiction, which gives rise to the present appeal, and the statutory penalties provided for breach of orders made under s 138.
The statutory penalty provided by s 138(7) is imposed for “breach of any order made under paragraph (a) or paragraph (b) of subsection (2) of this section”. The offence created by s 138(7) is a summary one, with a relatively minor penalty of a fine of up to $1,000. Breach of an order made under s 138(2)(c) (excluding members of the public) “may be dealt with as contempt of court” under s 138(8), perhaps because such breach is likely to amount to a contempt in the face of the court, justifying summary response. The reference to punishment for contempt is an indication that s 138 does not affect the contempt jurisdiction of the courts, even if it ousts the inherent power of the courts to make orders of the kind authorised by s 138(2) and provides a statutory penalty for their breach which may preclude recourse to the common law contempt jurisdiction for vindication of the court’s authority.
There is high authority that the common law proceedings for contempt of court, undertaken against Mr Siemer by the Solicitor-General in the present case, require proof of conduct which risks the proper administration of justice. On this approach, contempt may not be shown simply by proof of knowing breach of a court order, without more. Court orders and other steps taken to protect the interests of justice may be evidence both of the risk to the administration of justice in publication and knowledge of risk on the part of the person against whom proceedings are brought. But their evasion or breach is not in itself always contempt.
In Leveller, Lord Scarman, in making the point that the law of contempt had been bedevilled by technicalities throughout its history, referred to one such technicality as being the question whether a court can make an order “binding on persons who are neither witnesses nor parties in the proceedings before the court”. This was, he thought, the wrong question:
It is a misconception of the nature of the criminal offence of contempt to regard it as being an offence because it is the breach of a binding order. The offence is interference, with knowledge of the court’s proceedings, with the course or administration of justice ....
In the same case, Lord Edmund-Davies made it clear that “something more than disobedience of the court’s direction needs to be established”:
That something more is that the publication must be of such a nature as to threaten the administration of justice either in the particular case in relation to which the prohibition was pronounced or in relation to cases which may be brought in the future.
And Lord Russell described contempt of court as “the improper interference with the due administration of justice [which] need not involve disobedience to an order binding upon the alleged contemnor”.
Although in Independent Publishing the Privy Council held that the common law did not permit a suppression order to be made against a non-party, it concluded that the publication in that case was contempt because it risked the proper administration of justice. But the mistake made by the Judge was described by Lord Brown as being not only in having believed his non-publication orders to have been validly made, but also in believing “their breach ipso jure to constitute a contempt”.
If common law proceedings for contempt are not made out simply by establishing knowing breach of a non-party court order but require the court to be satisfied that the person proceeded against has wilfully risked the proper administration of justice (a question that, in cases of prohibition on publication, requires consideration of whether the prohibition was necessary), then whether the court had inherent power to make the order against the world may be something of a distraction. As I discuss below, I consider that such an order may well have evidentiary and precautionary value (as Lord Scarman in Leveller thought the case in relation to the order there made, while the other members of the House of Lords thought a warning would be appropriate) which is sufficient justification for its making even if it is subject to challenge. I would not however treat knowing breach of a court order directed to the world as necessarily amounting to contempt.
It is the effect of the publication (in a publication case) that must be assessed for its impact on the administration of justice. And if restraint of publication is not reasonably necessary to avoid risk to the administration of justice, it cannot amount to contempt even if it was prohibited by court order or advised against by court warning. If the common law contempt jurisdiction requires proof of risk to the administration of justice in the publication (as I think is commensurate both with the penalties able to be imposed and the nature of the proceedings, which amount effectively to an exceptional common law offence), there is no question of the validity of prohibition of publication being treated as collateral to the contempt proceedings. It is central to the contempt.
In the case of breach of orders made inter partes, the power of the court to prevent abuse of its processes may well be properly invoked to prevent challenge in contempt proceedings to orders which should have been challenged in the proceedings in which they were made. I would not want to be categorical when such a case is not before us. It may be that there are cases where it is in the interests of justice that the validity of an order made inter partes should be able to be challenged in proceedings for committal for contempt for its breach. But they are likely to be rare.
Whether the statutory offence under s 138(7) of breaching a court order made under s 138(2) may be defended by showing that the order was not reasonably necessary is similarly not a matter that is before us. In favour of an interpretation that the offence is complete on breach with knowledge of the order are the summary nature of the offence and the limited penalty able to be imposed. On the other hand, the offence may trench on the rights of freedom of expression and natural justice already discussed. And the arguments discussed at  to  in relation to the need to permit effective opportunity to challenge such orders where the person proceeded against is not a party to the proceedings in which the order is made resonate in connection with the statutory offence also, if a little more faintly because of its relatively minor nature.
For these reasons, I do not think it can be conclusive of the contempt proceedings either that there is knowing breach of a court order or that an order against the world was available at common law. I explain in some more detail in what follows why I consider that challenge in the contempt proceedings cannot be foreclosed as merely “collateral” in circumstances where it cannot be characterised as an abuse of process. And I explain further why I think that orders banning publication may usefully be made even if they are defeasible in the common law proceedings for contempt to enforce them. But first it is necessary to deal with the argument that s 138 of the Criminal Justice Act excludes the common law contempt jurisdiction in relation to publication by setting up an exclusive regime where it applies for prohibiting publication of court proceedings and for punishing breach of the orders it authorises, ousting both any inherent jurisdiction to make suppression orders of the sort provided by the section and common law proceedings for contempt upon breach.
Criminal Justice Act 1985
Powers to exclude members of the public and forbid publication of reports or accounts of evidence and submissions given at sittings of courts dealing with criminal offences were contained in s 138 of the Criminal Justice Act at the time Winkelmann J made the order suppressing publication of her judgment. They could be exercised only where the court considered that such orders were required in the interests of justice or in other limited circumstances, not in issue here. Such powers modified the general rule stated in s 138(1) that court sittings dealing with criminal offences were required to be open to the public.
Section 138 was in the following terms:
The statutory powers to suppress publication of reports or accounts of evidence and submissions where required in the interests of justice (such as for fair trial purposes) are authorised against the world. Orders “of any kind described in subsection (2)” (forbidding publication of evidence or submissions, forbidding publication of the name or identification of a witness, and excluding the public from the court) are expressed by s 138(5) to be “in substitution for any such powers that a court may have had under any inherent jurisdiction or any rule of law.” As a result, “no court [has] power to make any order of any such kind except in accordance with this section or any other enactment”.
I doubt whether the reference to “evidence” in s 138(2)(a) is confined to evidence actually heard at a sitting of the court. I think it more likely that the subsection authorises a prohibition on reporting evidence “adduced” by being “brought forward for consideration” at such a hearing, as is the usual meaning of “adduce”. That interpretation would permit prohibition on publication of evidence considered for the purposes of admission at a hearing under s 344A of the Crimes Act 1961. On that basis, I cannot agree with the majority view at  that it does not cover “evidence which one party wishes to adduce but which is held to be inadmissible” or “a defendant’s prior criminal record” (if referred to at a sitting of the court). But, indeed, even if wrong in that view, I consider that the ability to prohibit publication of the submissions made (which will refer to any such evidence if a ruling has been sought on it or if it is referred to in support of a bail application, for example) covers any reference to evidence proposed to be called at trial referred to in the submissions. And if there is no such reference, the terms of s 138(6) permit the court to decline to state any facts on which it has relied in its determination. It must also be remembered that the exclusion of the inherent powers is in ample terms. Section 138(5) makes it clear that the powers contained in subs (2) prevent recourse to the inherent jurisdiction in respect of orders “of any such kind”.
To the extent that a pre-trial judgment contains an account of evidence or submissions which need to be suppressed until verdict in the interests of fair trial, s 138(2) provides statutory authority for a prohibition on publication. Reports or accounts of evidence (including that ruled inadmissible) and submissions cover the substantial ground in respect of material which, if disclosed prematurely, could risk fair trial. It is hard to think of information before the court prejudicial to fair trial which could not be so characterised. If publication of a court determination might risk fair trial or otherwise prejudice the administration of justice, the suppression authorised by s 138(6) permits the court to decline to state in public the facts, reasons or considerations taken into account in the determination. That authority does not extend to the result determined, in what may be seen as a legislative judgment protective of open justice. Because the basis on which the suppression orders were made here is not explained, it is not possible to conclude that announcement of the result in accordance with s 138(6) was itself seen by the Judge as prejudicial to fair trial. The Crown at the time seems to have been of the view that prohibition of publication of the result went further than was necessary and in this Court counsel for the Solicitor-General acknowledged as much. The Judge herself varied the suppression order to permit reporting of the outcome of the decision in her judgment of 21 December 2010, suggesting that the original order may have been seen as unnecessary. But, in any event, I am of the view that suppression of the determination could not have been available under any inherent power because such power is inconsistent with the terms of s 138(6). Such inconsistency excludes the inherent power as effectively as an express provision such as s 138(5).
The terms of s 138(2)(a) are apt to cover an account contained in a judgment as well as reports and accounts able to be made by those who attend the sitting of the court. I am unable to accept that the language of s 138(2)(a) is properly understood to refer only to reports or accounts of the evidence and submissions by those other than the judge. Reference in a judgment to evidence referred to or submissions made is equally an account or report of such evidence and submissions. There is no basis to read down the provisions of s 138(2) to exclude such account or report because it appears in a judgment. If an order made under s 138(2)(a) relates to evidence or submissions also contained in a judgment, publication of an account of the evidence or submissions is equally prohibited by the order. Were it not, secondary reporting would always undermine the order. The terms of s 138(2)(a) in my view cover all reports and accounts.
The power to exclude from the court all but those exempted from exclusion under s 138(2)(c), does not apply to “the announcement of the verdict or decision of the court (including a decision to commit the defendant for trial or sentence) and the passing of sentence”. Verdicts, decisions, and sentences are required by s 138(6) to “take place in public” in every case. I do not think it is possible to read down this subsection to exclude from it pre-trial determinations made in the proceedings or to confine it to the “important stages” of a criminal proceeding (however they are to be identified and distinguished from determinations which are “incidental”). The section applies, as s 138(1) makes clear, to “every sitting of any court dealing with any proceedings in respect of an offence”. I see no reason why, “[s]ubject to .... any other enactment”, that should not extend to bail variations or adjournments. (Middle-banding determinations, referred to by the majority, are determined “on the papers” under s 184Q of the Summary Proceedings Act 1957, in force at the relevant time.)
Section 138(6) applies in all cases and “[n]otwithstanding that an order is made under subsection 2(c) of this section”. The subject matter of s 138, as described in subs (1), is “every sitting of any court dealing with any proceedings in respect of an offence”. Exclusion orders can be made under s 138(2)(c) in respect of “the whole or any part of the proceedings”. The sense of the provisions read together is that decisions in respect of any part of the proceedings must as a general rule take place in public, subject only to the power of the court under s 138(6), if “satisfied that exceptional circumstances so require”, to “decline to state in public all or any of the facts, reasons, or other considerations that it has taken into account in reaching its decision or verdict or in determining the sentence passed by it on any defendant”.
It may be that the practice of the courts has not been sufficiently respectful of the provisions of s 138 and its emphasis on open justice. That is perhaps indicated in the present case by the reference in the earlier pre-trial ruling to “the standard orders” and the absence of specific reasons in respect of the present order. The Judge is not to be singled out for criticism in this. As the majority reasons indicate, the approach she adopted seems general practice. If so, it does not seem to me to meet the open justice requirements of s 138. More importantly, I am of the view that any inherent power to make the suppression orders made in respect of the judgment of 9 December 2010 was excluded by s 138. I would allow the appeal on this basis.
Contempt and suppression orders
Section 138 is not concerned with the jurisdiction of a court to punish for contempt. The reference to the contempt jurisdiction in s 138(8) indicates that the jurisdiction (whether to punish for contempt in the face of the court or by proceedings for contempt) continues except where the extent of the contempt is confined to breach of an order made under s 138(2)(a) or (b). In such a case the statutory penalty is inconsistent with a free-standing power to punish for contempt. Whether contempt proceedings may nevertheless lie for publication which is shown by the prosecutor in itself to undermine the administration of justice in the particular case for reasons which are not co-extensive with breach of a court order is not in issue in the present case. Here, the sole basis for the finding of contempt is the breach of court order. Because, however, the majority view in this Court is that the common law contempt proceedings were not excluded by s 138 (contrary to the view I take), it is necessary for me to indicate why I consider that in proceedings for contempt based on breach of an order made against the world (and which may be contrasted with an order made in proceedings to which the defendant was a party or contempt in the face of the court), the person against whom the proceedings are brought can challenge the order in the contempt proceedings.
In what follows I explain my conclusion that there is no abuse and therefore there was no basis for refusing to determine the defences the appellant sought to raise. I would not limit challenges to the legality of court orders in contempt proceedings to the questions of whether there is common law power to make suppression orders against the world to protect fair trial or whether the inherent jurisdiction is excluded by statute. Such limited scrutiny is based on unworkable distinctions between errors within and without jurisdiction, from which the common law (at least in New Zealand and the United Kingdom) has been relieved since the decision of the House of Lords in Anisminic Ltd v Foreign Compensation Commission, as Lord Irvine explained in Boddington.
Nor is such approach consistent with the obligations imposed on New Zealand courts under the New Zealand Bill of Rights Act when restrictions on rights are involved. When the House of Lords took a similarly narrow approach (limited to whether an injunction restrictive of freedom of speech was within the scope of a general power possessed by the court), it was rejected as inadequate protection for human rights by the European Court of Human Rights. The European Court of Human Rights required consideration of whether the interference was “necessary having regard to the facts and circumstances prevailing in the specific case”. Similar scrutiny (directed under the New Zealand Bill of Rights Act to whether the restriction is demonstrably justified as “reasonable”) is required in respect of the rights and freedoms protected under the New Zealand Bill of Rights Act.
The appellant was entitled to a determination that the order was lawfully made before being committed for contempt. It would not be lawfully made if an unreasonable interference with the right to freedom of expression or not reasonably necessary to prevent the risk of prejudice to the fair trial rights of the accused. It seems to me wrong in principle and pointlessly formalistic to insist on a sequence by which recourse must first be had to an inconvenient and uncertain remedy (as the informal review suggested to have been available to the appellant here seems to me to be, as discussed further at  to ) rather than to permit defensive challenge in the proceedings for contempt. I would apply here the reasoning of the House of Lords in Boddington. I do not accept there is anything in the nature of a court order which requires a different approach, for reasons further explained below. And what is properly characterised as “collateral” and what as “direct” is almost as doubtful as what is properly characterised as “jurisdictional”. Indeed, until comparatively recently “collateral challenge was a primary means of impugning judicial and administrative action”. Permitting defensive challenge in contempt proceedings if the order breached is made in material error of law seems to me to be necessary to give effect to the rule of law.
Before explaining these conclusions, it is necessary to deal with the question that has occupied most time in the lower courts and in the present appeal: whether at common law there is power to make an order prohibiting publication of evidence by persons who are not parties to the proceedings in which the order is made in order to protect fair trial.
Orders against non-parties
In Leveller, some members of the House of Lords expressed doubt about the decision of the New Zealand Court of Appeal in Taylor, which held that there was inherent common law power to make an order binding on non-parties suppressing publication of information referred to at trial. More recently, in Independent Publishing, the Privy Council held that there is no such common law power and that only the legislature can confer such powers to bind third parties. Notwithstanding the absence of such power, the Privy Council in Independent Publishing considered that the actions of the defendant amounted to contempt because they risked the proper administration of justice. It was held in both Leveller and Independent Publishing that showing breach of a court order would be insufficient to constitute contempt. Rather, it was necessary to show that the conduct undermined the administration of justice.
In Independent Publishing, the Privy Council considered a court might warn of the risk that publication could prejudice the administration of justice and could amount to contempt. But whether the risk eventuated would turn on the court being satisfied that the publication was prejudicial to the administration of justice.
It should make little difference to the result (subject to any question of onus of proof) whether the correct approach is that contempt is established because an order shown to be lawfully made is knowingly breached or because the act of publication relied on as contempt is conduct which prejudices the proper administration of justice. What would, however, be irreconcilable in the two approaches is if no challenge to the lawfulness of an order could be entertained simply because it is an order of the court and contempt consists solely in its knowing breach, irrespective of its validity.
Even if there is power to make an order against the world (as the Court of Appeal held in Taylor), the subsequent enactment of the New Zealand Bill of Rights Act seems to me to require assessment in the contempt proceedings of the necessity for the order made similar to that required if the Independent Publishing approach were adopted. In both cases contempt will not be established if any prohibition on publication (implicit or explicit) was not reasonable. In particular, what is reasonable will entail consideration of whether suppressing publication is a limitation on freedom of expression that is justifiable in a free and democratic society.
For the reasons to be developed below, I consider that the approach taken by Richmond J in Taylor was too austere and should have allowed for challenge to the legality of the order made. That conclusion is important in the view I take that orders against the world, if not excluded by statute, may be made in New Zealand at common law to protect trial fairness.
Such power is long-established in New Zealand and has some support in early English authority (although it is subject to long-standing criticism, as the Privy Council in Independent Publishing noted). More importantly, recourse to such orders seems to me to be highly desirable where fair trial is at risk from premature disclosure of information. The Privy Council in Independent Publishing acknowledged that a warning to the press would be helpful. An order made on the face of the judgment might be thought to be even more helpful.
It reduces the risk of prejudice to fair trials if what the court considers prejudicial is identified in an order readily accessible to everyone who might be affected. In modern conditions, the risk of prejudice by premature disclosure of information before trial is amplified by statutory procedures for ruling on the admissibility of disputed evidence before trial and the ease of modern dissemination of information, including of judgments and rulings. In pre-trial determinations such rulings often need to describe disputed information both to justify the exercise of judicial authority and to permit effective review or appeal for error. Electronically available judgments and increased emphasis on open justice (which favours admission of the news media and the public to court processes, even where interim protections need to be maintained for fair trial purposes) mean that disputed information which may be highly prejudicial to fair trial is readily available.
Orders remove uncertainty and risk for those potentially liable. And, because restrictions on publication are an interference with freedom to impart information, it enables any restriction which may be justified in a free and democratic society to be “prescribed by law”, as s 5 of the New Zealand Bill of Rights Act requires.
So I would follow the approach taken in Taylor in recognising that courts may make orders prohibiting publication (or otherwise protective of the administration of justice) against the world and despite the doubts expressed in the United Kingdom and Australia. But that acceptance is on the basis that breach of such orders does not itself constitute contempt unless the orders are otherwise lawful. Effectively, that means that it is necessary to conclude that the publication itself constitutes a contempt, as the common law has required.
The order breached may be challenged in contempt proceedings unless such challenge would constitute an abuse of process
I do not accept that breach of such orders by a non-party is to be treated for the purposes of committal for contempt as if the breach were undertaken by a party to the proceedings. I consider that the distinction Woodhouse J drew attention to in Taylor between orders made against the world and those against parties to proceedings (referred to in ) is critical in considering whether publication in breach of suppression orders constitutes contempt.
A person proceeded against for contempt at common law may defend the proceedings on the basis that his actions did not have any tendency to undermine the proper administration of justice. Is the ability to run such a defence removed when the contempt concerned consists of knowing breach of a court order? The Solicitor-General argues that knowing breach of a court order constitutes a contempt of itself and that the validity of the order cannot be set up as a defence in the contempt proceedings because of a rule against collateral challenge, unless the Court had no jurisdiction to make an order of the sort made.
I am unable to agree that any such rule is part of New Zealand law. Apart from the statements made in Taylor (where only Richmond J is clear on the matter), there is no New Zealand authority that denies a defendant to contempt proceedings a right to be heard on the lawfulness of the order breached. I consider that it is consistent with fundamental principles of natural justice that a defendant proceeded against for contempt in breach of a court order may raise material error of law as a defence unless the claim amounts to an abuse of process.
Depending on the circumstances of the case, it may be an abuse of process for a party to proceedings in which the order was made to seek to defend contempt proceedings by challenging the legality of the order if such challenge could earlier have been made by him in the proceedings in which the order was made or by appeal. That was the position in Siemer v Solicitor-General. Such result does not come about because of any rigid “rule” against collateral challenge (that is to say, challenge in proceedings not constituted for the purpose of setting aside the order).
An application for judicial review (or other procedure for direct challenge) is “not a straitjacket which must be put on before rights can be asserted”. And since no crime is committed through the infringement of an invalid rule, the order may be challenged in the court in which the offence is tried, unless such challenge would be an abuse of process. As was said by Lord Somervell in Director of Public Prosecutions v Head:
Is a man to be sent to prison on the basis that an order is a good order when the court knows it would be set aside if proper proceedings were taken? I doubt it.
There is “no rule that lends validity to invalid acts” and exclusion of any consideration of the legality of the underlying order is “too austere and indeed too authoritarian to be compatible with the traditions of the common law”, as Lord Steyn made clear in Boddington.
Boddington (a case generally credited as having brought order to the law concerning collateral challenge in the United Kingdom) overruled the judgment of the Divisional Court in Bugg v Director of Public Prosecutions, which would have excluded collateral challenge in criminal cases, as “contrary to authority and principle”:
It does not apply in a civil case when an individual seeks to establish private law rights which cannot be determined without an examination of the validity of a public law decision. Nor does it apply where a defendant in a civil case simply seeks to defend himself by questioning the validity of a public law decision. These propositions are established in the context of civil cases by four decisions of the House of Lords: .... One would expect a defendant in a criminal case, where the liberty of a subject is at stake, to have no lesser rights. Provided that the invalidity of the byelaw is or [may be] a defence to the charge[,] a criminal case must be the paradigm of collateral or defensive challenge.
Although Boddington and the cases referred to in it concerned breach of orders or rules made by subordinate legislative bodies or administrative agencies, the reasoning in the speeches provides no basis to draw any distinction according to whether the defensive challenge follows breach of administrative order or delegated legislation on the one hand and orders made by courts on the other. Unless there is some abuse of process, the general rule that material error may be raised in penalty proceedings applies. Abuse of process may arise where the defendant was a party to the proceedings in which the order was made and had rights of appeal. Where however he was not a party, defensive challenge without more is no basis for suggesting abuse.
Indeed, as indicated, it is far from clear that any other proceedings in which such challenge could be made were available as of right in the present case. The High Court, by conventional legal doctrine, may not be judicially reviewed except by appeal in the proceedings in which its challenged order or determination is made. In Broadcasting Corporation of New Zealand v Attorney-General this inconvenience was overcome by making an informal application to the High Court to have the order varied or rescinded, which was then removed directly into the Court of Appeal. It is said that the courts have been entertaining informal requests for variation or review of suppression orders, especially by the news media. But the status of these requests is not clear and without clear entitlement they amount to little more than prompts to the judge to reconsider orders in his discretion. Their application to those who are not members of the press but who are recognised to have rights to impart information under s 14 of the New Zealand Bill of Rights Act is still more obscure. It is uncertain whether s 66 of the Judicature Act 1908 would permit an appeal from a decision to decline such reconsideration.
These difficulties are illustrated by the present case, in which Mr Siemer’s direct application for reconsideration of the suppression order was rejected for filing. His attempt to bring the matter before the Court on an originating application did not fit easily within the categories in which such procedure can be invoked as of right and he was obliged to seek leave to file it, a request which was declined, it seems because it was seen as abuse of process because a collateral challenge to the findings of contempt which had already been made. Ms Bright’s earlier attempt to seek reconsideration in the criminal proceedings themselves was rejected, apparently because she lacked standing in them.
It was difficulties such as these that led the Privy Council in Independent Publishing to reject as “an impossible argument” the suggestion that the appellant should have applied to the trial judge for variation or discharge of the orders. It also made the point that it was “no answer to a constitutional claim of this nature to say that some other remedy would or might have been open to the claimant”. In a case where rights under the New Zealand Bill of Rights Act are engaged, I consider that insistence on procedural particularity is similarly inappropriate. The concern should be the practical administration of the law. And while frivolous defences may be an abuse, there is no question but that freedom of expression is here engaged and is at the heart of the challenge the appellant makes.
The opportunity to challenge an order by judicial review in separate proceedings is expensive distinct litigation in which relief is discretionary. Its potential availability does not warrant treating a defence of illegality in contempt proceedings as an abuse of process and refusing to hear it. In the case of an informal application for reconsideration of an order made in the High Court (an application in which the standing of the defendant is uncertain, as the present case may illustrate), the procedure is not as of right and it is unclear whether there are rights of appeal. Such opportunities do not in my view properly justify excluding defensive challenge in the contempt proceedings that the order was not lawfully made.
Moreover, whatever the pre-publication opportunities for challenge, the proceedings for contempt are effectively a common law offence in which the nature of the offence and the appropriate defences to it are matters for judicial determination. The common law contempt proceedings permit much more serious penalties than the fines prescribed by the statutory offence for breach of the orders authorised by statute. The human rights of the publisher as well as his liberty are in jeopardy in the proceedings. Effective opportunity for challenge, not simply an opportunity available with effort, is appropriate. So too is a procedure which confers a right of appeal, consistently with the need for a full and effective remedy for breaches of the New Zealand Bill of Rights Act. I would frankly recognise such a defence in modern conditions.
The contempt proceedings to enforce the order are the appropriate proceedings in which a defendant not a party to the making of the order can challenge it by way of defence. They are proceedings brought to vindicate the order against someone who has not been heard on it. They are the convenient process for such challenge. Given New Zealand law as to standing and procedure, permitting challenge in contempt proceedings would not allow the defendant to evade the rules applicable in judicial review. And concerns sometimes expressed as to the competence of the court in which enforcement proceedings are brought to assess the legality of the rule infringed (and which are sometimes raised to resist challenge said to be “collateral”, although rejected in Boddington) do not arise in the common law contempt proceedings in the High Court.
Statements about the obligation to obey court orders in cases not concerned with defensive challenge in contempt proceedings to enforce orders made against the world do not answer the rule of law reasons why enforcement by contempt proceedings or prosecution should admit defensive challenge to the validity of the orders. In Isaacs v Robertson, the breach was by a party to the litigation. He was held not to be entitled to treat the order as a nullity, although Lord Diplock in that case pointed out he could have applied, irrespective of formal procedure in the rules, to have the order set aside “ex debito justitiae”. Whether Isaacs v Robertson would be followed in a similar case in New Zealand today may need further consideration if the point arises. It might be thought that there is something unsatisfactory and formalistic in acknowledging a right to apply to have an order set aside ex debito justitiae but not allowing such application to be run by way of defence to proceedings for contempt. If the case ever arose, it would be necessary to consider whether such outcome is consistent with the right to natural justice contained in the New Zealand Bill of Rights Act. For present purposes, however, it is sufficient to say that Isaacs v Robertson did not involve an order made against someone not a party to the proceedings in which it was made.
As already indicated, I am unable to agree that the common law of New Zealand should treat contempt as conclusively established upon breach of an order that has not been set aside in other proceedings at the time of the breach. The convenient forum in which to challenge the validity of an order made against the world and in the making of which there was no opportunity for the person against whom contempt proceedings have been brought to be heard is in the contempt proceedings themselves.
New Zealand authority does not compel a different result. The Full Court in the present case relied upon the judgment of Richmond J in Taylor to hold that “[i]t is not a defence to a charge of breaching the order to raise non-jurisdictional defects in the process by which it was made”. Richmond J was, however, the only member of the Court of Appeal in Taylor to suggest directly that the validity of the order could not be challenged in the contempt proceedings for error of law “within the jurisdiction”. Wild CJ seems to have treated its validity as able to be raised by way of defence, but concluded the order was valid. Woodhouse J would have allowed the appeal on the basis that the order should not have been made both because the court did not have a power of the sort relied on and because, even if it did, the use of the power was not justified. More importantly, the distinction drawn by Richmond J between “jurisdictional” errors and errors within jurisdiction is contrary to later New Zealand case law rejecting such classification, in application of Anisminic. It is the same reasoning adopted by Woolf LJ in Bugg in the context of a defensive challenge to delegated legislation but subsequently repudiated by the House of Lords in Boddington.
The limited authority in New Zealand supportive of a rule that defensive challenge to court orders against non-parties cannot be raised in contempt proceedings may not be surprising. It is to be expected that questions of collateral challenge will have arisen principally in cases concerning the enforcement of delegated legislation or administrative decisions, rather than in connection with court orders, because it is perhaps only in the case of court orders against the world that challenge within the proceedings (including by appeal) is not available. The absence of cases dealing with such orders (perhaps associated with doubts as to their validity following Leveller and Independent Publishing) may explain the comparative leanness of authority concerning defensive challenge to court orders in contempt proceedings against non-parties.
In the United Kingdom, there are other authorities of some relevance apart from those discussed above in relation to Boddington. So, in Attorney-General v Lundin the disobedience of a court order by a journalist required to reveal his sources was treated in committal proceedings by the Court of Appeal as not inevitably amounting to contempt. And challenges to judicial orders justifying detention may be raised in proceedings by way of habeas corpus. Hadkinson v Hadkinson, cited in the judgment of the majority at  to , is not a case of defensive challenge to contempt proceedings but a case where a party in breach of a court order was prevented from being heard further in the substantive proceedings while in breach. It is not authority for any rule that the validity of the order breached cannot be raised as a defence in proceedings for committal for contempt.
In Australia, the authorities do not support a rule against defensive challenges to court orders in contempt proceedings. Jackson v Sterling Industries Ltd (cited in the majority judgment in support of the proposition that there is a rule in common law jurisdictions preventing invalidity being raised as a defence in contempt cases) concerned not a defensive challenge in proceedings for contempt but whether the Federal Court had jurisdiction to grant a Mareva injunction. Indeed, even the question of whether an invalid order had to be complied with was expressly left open by the majority in the High Court. The other cases cited in the majority judgment in the present appeal are all examples of orders made against named parties.
Only in Canada is there high authority for a rule that the validity of a court order cannot be raised collaterally in contempt proceedings. The most forceful expression of such rule is in R v Wilson. It was not a contempt case, but one where a wiretap authorisation made in a superior court was not able to be ruled invalid (and had to be observed) in the lower, trial court. In the more recent decisions of the Canadian Supreme Court it has been concerned to ensure that what it acknowledges to be a formalistic approach to collateral challenge should not work injustice. For that reason it has admitted a number of exceptions to the rule. These are justified by the Supreme Court on rule of law grounds. They arise especially where permitting defensive challenge in contempt proceedings is necessary to enable effective challenge to an invalid order. In Dagenais v Canadian Broadcasting Corporation (a case concerning the setting aside of an injunction), McLachlin J was concerned to stress that effective challenge (absence of which admitted an exception to the general rule) should include a right of appeal. And in Attorney-General for Quebec v Laroche the acknowledged procedural difficulty in obtaining review of the order of a superior court was a consideration in favour of a further exception.
I do not think the Canadian authorities, which can be seen as examples of refusal to countenance abuse of process, and which in any event reflect that jurisdiction’s approach to jurisdictional error, prompt the adoption of a general rule against collateral challenge. In result, and because of the exceptions recognised, the outcomes reached under the Canadian approach and under the Boddington approach may not be too different. The reasoning in Boddington seems to me however to be more in tune with New Zealand authorities and compelling in its application to defensive challenge in contempt proceedings for breach of a court order. The general rule in my view should permit such challenges except where they amount to an abuse of process.
As has been indicated, proceedings for contempt at common law may be contrasted with the statutory offence of breaching orders of the court authorised by s 138 of the Criminal Justice Act. There, the penalty prescribed is limited to a fine of $1,000. The scope of available defences to the statutory offence of breaching court orders under s 138 is not in issue in the appeal. The nature of the penalty may well be relevant in determining whether the offence is complete on knowing breach of such orders without more. Whether it is, is a matter of statutory construction which is not presently before us.
Punishment under the common law contempt jurisdiction is not limited to a fine. The sentence available depends on the judge’s assessment of the seriousness of the contempt. The potentially severe punishment of which the defendant is at risk in the case of common law proceedings for contempt is a consideration which pulls against treating defensive challenge as an abuse.
I would hold that any error of law material to the validity of the order was a defence to the proceedings for contempt and should have been permitted to be raised by the appellant. In the present case, that would include a challenge that the order was an unreasonable restriction on Mr Siemer’s right to communicate the information suppressed. I would therefore have allowed the appeal on this ground on the basis that the Full Court should have heard the defensive challenges Mr Siemer sought to raise. If not for the indication by counsel for the Solicitor-General that rehearing was not sought, I would have remitted the contempt proceedings for rehearing in the High Court.
For the reasons given, I would allow the appeal. I consider that the inherent power to make orders of the kind made by Winkelmann J was excluded by s 138 of the Criminal Justice Act. I would have quashed the sentence on that basis. Since that is a minority view, it has been necessary to explain why I differ from the majority in their reasons, which are dispositive of the appeal. I am of the view that a non-party to proceedings in which an order is made against the world may raise error of law in the making of the order as a defence in contempt proceedings based on its breach. On this ground I would have allowed the appeal and quashed the sentence.
McGrath, William Young & Glazebrook JJ
(Given by McGrath and William Young JJ)
This is an appeal from a judgment of the Court of Appeal, upholding a judgment of a Full Court of the High Court finding Mr Siemer to be in contempt of court. The conduct in issue was Mr Siemer’s publication of a judgment, given in the course of criminal proceedings, which had been the subject of a suppression order. The Court of Appeal had also dismissed an appeal by Mr Siemer against the sentence of six weeks’ imprisonment imposed on him by the High Court.
On 9 December 2010 Winkelmann J delivered a judgment making pre-trial rulings in criminal proceedings on questions of severance and whether all defendants should be tried by judge alone without a jury. The proceedings involved 18 accused, variously charged with participation in an organised criminal group and offences involving unlawful possession of firearms and other restricted weapons. After evaluating the circumstances, including the number and nature of the charges faced by individual accused, the volume of evidence likely to be presented at the trial and the imposition on potential jurors of sitting on a trial having a likely duration of 12 weeks, Winkelmann J ruled that the trial of three accused should be severed from those of the remainder, who should be tried by judge alone.
The front page of the reserved judgment delivered by Winkelmann J carried this heading:
THIS JUDGMENT IS NOT TO BE PUBLISHED (INCLUDING ANY COMMENTARY, SUMMARY OR DESCRIPTION OF IT) IN NEWS MEDIA OR ON INTERNET OR OTHER PUBLICLY ACCESSIBLE DATABASE OR OTHERWISE DISSEMINATED TO THE PUBLIC UNTIL FINAL DISPOSITION OF TRIAL OR FURTHER ORDER OF THE COURT. PUBLICATION IN LAW REPORT OR LAW DIGEST IS PERMITTED.
It is plain that the heading was intended to be an order of the Court. Expressing suppression orders using a banner on the front page of judgments is regularly done in New Zealand as the most practical means of making clear to all receiving the judgment that restrictions on publication of its contents have been imposed and informing them of what those restrictions are. It was not necessary to record that the order had been made or made in this way in the reasons for judgment.
Shortly after the judgment was delivered, the appellant, who was not a party to the proceedings, published an article on each of two identical websites he operates, which are called “Kiwisfirst”. So far as relevant, the article, headed “JUDGE OR BE JUDGED”, read that:
Chief High Court Judge Helen Winkelmann .... ordered yesterday that the “Urewera terrorist” prosecution (R v Bailey) against 15 accused will be by judge alone trial. The landmark ruling was sought on application by the Crown and had been opposed by the accused.
The remaining three of the eighteen listed defendants were ordered separate trials.
Winkelmann J ordered the public not be told about her order. In the past Winkelmann has stated the reason for such secrecy was to ensure the jury pool is not prejudiced by pre-trial information. Her latest order prohibiting a jury states the potential length of the trial, or that jurors would use “improper reasoning processes”, each provide a sufficient ground for denial of jury trial in her mind.
The word “ruling” in the first paragraph was a hyperlink which gave readers of the article immediate electronic access to a copy of the 9 December 2010 judgment. No application for variation of the suppression order was made by Mr Siemer to permit this publication. In the copy of the judgment accessed from the websites, the heading recording the suppression order was altered so that the order purported to read: “PUBLICATION IN LAW REPORT OR KIWISFIRST IS PERMITTED”. It was obvious from the typeface that the suppression order had been altered, and the change made it clear that Mr Siemer was aware of the order.
On 17 December 2010 Crown Counsel in the Crown Law Office wrote to Mr Siemer, by email and facsimile, advising him that the Solicitor-General was of the view that publication of the article and the judgment on websites for which he was responsible “constitutes a clear and deliberate breach of the suppression order”. The letter required him to remove both the article and judgment from the websites. On 18 December Mr Siemer published a second website article headed “CROWN TO PERSECUTE WHERE LAW PREVENTS PROSECUTION”. This article stated that defendants in the criminal proceedings would appeal against the ruling for judge alone trials, adding:
Meanwhile, Crown Law has sent notice that it intends to prosecute kiwisfirst publisher Vince Siemer for publishing Winkelmann’s judgment, on the grounds Winkelmann ordered the public not be told about it. Crown Law is seeking Siemer be imprisoned.
The threat to prosecute comes despite [the prosecutor] advising the High Court and Crown Law that they intend to seek rescission of all suppression orders on behalf of the prosecution on the grounds publication of Winkelmann’s judgment “cannot possibly prejudice the fair trial rights of the accused, and (the issues in the judgment) are a matter of genuine public interest.”
On 21 December, Winkelmann J heard counsel on whether a more limited suppression order would be appropriate. The Crown had applied for a variation of the suppression order on the basis it went further than was needed. All defendants opposed the application. Following that hearing, the Judge issued a minute which said:
In a trial of this nature and given the extent of publication in connection with it occurring outside the mainstream media, I have concluded that the prudent course is to continue to suppress the content of the judgment except that part of the judgment which declares the result ....
In giving these reasons for varying the suppression order Winkelmann J made explicit that the reason for making the order and for maintaining it in modified form was to prevent prejudicial publicity that might impact on jury trials of some defendants.
High Court judgment
The Solicitor-General then brought proceedings in the High Court seeking to have Mr Siemer committed for contempt. Mr Siemer defended the proceeding. The proceeding was heard by a Full Court of the High Court, comprising Mackenzie and Simon France JJ, which held that because of the way the case had been pleaded by the Crown, the application for contempt of court should proceed solely on the basis that the appellant had breached a court order by which he was bound. That basis did not require proof by the Crown that the publication in breach of the order caused a risk to a fair trial.
It was common ground, and accepted by the Court, that if there were no jurisdiction to make the suppression order, the application for contempt for breaching it would fail.
No reasons for making the suppression order were expressed in the judgment by Winkelmann J. The High Court said of this:
The reason for making such orders is self-evident, namely that publication of some or all of the material contained in the pre-trial ruling may prejudice the right of any accused to a fair trial. Pretrial rulings such as bail, severance and admissibility of evidence can all contain material that will be inadmissible at trial and which should not be publicised to the potential pool of jurors. Although there are variations in the scope of such orders, the general effect of them is a temporary ban on publication until the trial is complete. The ability to make such orders is an important component in discharging the court’s responsibility to ensure the fair trial of an accused.
The Court also observed that, while there was no power under statute to make an order suppressing a judgment, the traditional view was that the source of judicial power to make them lay in the inherent jurisdiction of the High Court or inherent powers of the District Court to give effect to its extensive criminal jurisdiction. This had been established in New Zealand in Taylor v Attorney-General. The High Court held that, although the Court of Appeal’s judgment had recently been rejected in Independent Publishing Co Ltd v Attorney-General of Trinidad and Tobago, Taylor remained good law in New Zealand.
The High Court also considered whether s 138 of the Criminal Justice Act 1985, which confers power to forbid reports of proceedings, had abrogated the inherent power, and concluded it had not. It held that the statutory provisions were in substitution for the common law powers, but only addressed situations where a court wished to control publicity of events in open court that had been witnessed by those present, such as the media. The statute had no other impact on the inherent jurisdiction or powers of a trial court. It followed that Winkelmann J had power to make the suppression order in relation to the judgment making pre-trial rulings.
The Court also rejected a submission that the appellant’s criticisms concerning absence of reasons for the suppression order and the breadth and vagueness of its terms provided a defence to the contempt proceedings. Once it was accepted there was power to make the order, it was binding until set aside. It was open to persons, including Mr Siemer, to apply to the court for variation or removal of the order but, until that was done successfully, the suppression order was binding. It was not a defence to a charge of deliberately breaching the order to say the original Court was wrong to make the order. As well, as the purpose of the order was clearly to protect the fair trial rights of the accused, it was squarely within the court’s functions and no question of jurisdiction to make the order arose.
Mr Siemer acknowledged that he published the judgment on his websites and that he knew of both the order of 9 December and variation on 21 December. The Court held that the order covered the form of publication undertaken by Mr Siemer. The Court was satisfied beyond reasonable doubt that he had published the judgment in breach of the order. The deliberate nature of the conduct merited a finding of contempt. The Court subsequently sentenced Mr Siemer to six weeks’ imprisonment.
The High Court delivered its judgment finding Mr Siemer in contempt of court on 4 July 2011. On 25 August 2011, Mr Siemer attempted to file an application in the Criminal Registry of the High Court challenging the lawfulness of Winkelmann J’s suppression order. The application was rejected by the Registrar on the basis that the Registry had no jurisdiction to accept it.
Mr Siemer drew the Registrar’s attention to a minute of the High Court issued on 26 July 2011 in which Brewer J had directed the Registry not to receive for filing an application by Ms Bright, an acquaintance of Mr Siemer, for permission to commence an originating application seeking variation of Winkelmann J’s suppression order. Mr Siemer said that although the Full Court had said it was open to persons, including Mr Siemer, to apply for a variation or removal of the suppression order, it was proving impossible to find out how that was done. In his minute, Brewer J had held that, under the rule governing that particular application, the High Court had no jurisdiction to make the orders sought by the applicant.
Court of Appeal judgment
Mr Siemer appealed. The Court of Appeal observed in its judgment that there was no statutory power for the High Court either to order suppression, or to require postponement of publication, of its judgments. At issue was whether the Court had inherent power under the common law to make such an order. On that point the Court agreed with the High Court that Taylor remains good law in New Zealand and did not follow the differing approaches taken in English cases decided since Taylor.
The Court of Appeal identified several difficulties with the English approach, which does not recognise the existence of an inherent power to make suppression orders binding the world at large. In particular, the Court explained that such an approach makes the threat of sanction for contempt the only means of control over publications that put fair trials at risk. Sanctions imposed after the event cannot, however, repair damage which may already have been done to the fair administration of justice. In the New Zealand context, the Court of Appeal saw the exercise of the inherent power to make suppression orders as being the most flexible and effective means of securing a fair trial. It pointed to approval of Taylor’s recognition of the inherent power in a number of subsequent decisions of the Court of Appeal and of this Court. The Court of Appeal, in agreement with the High Court, also held that, on its terms, s 138 of the Criminal Justice Act did not oust the Court’s inherent power under the common law to suppress publication of a judgment.
Finally, the Court of Appeal rejected the submission that the suppression order should not have been made without reserving a power or prompt right for Mr Siemer to seek review of the order. The Court said:
Mr Siemer was not a party to the proceeding. However, he was entitled, if he wished, to apply to review the order. Instead, he elected immediately to breach the order and must bear the consequences. As Elias CJ and McGrath J observed in Siemer v Solicitor-General, our constitution requires that Court orders are obeyed until properly challenged or set aside.
On 19 July 2012 this Court gave Mr Siemer leave to appeal against the Court of Appeal’s judgment upholding his conviction. The approved ground of appeal was whether New Zealand courts had inherent power or jurisdiction to suppress judgments in criminal cases. The Court declined to give leave on whether a collateral challenge to the High Court’s suppression order could have been brought in the subsequent contempt proceedings alleging breach of the order.
The Court heard argument on the inherent power or jurisdiction issue on 15 November 2012. Following that hearing, we decided that we should also hear argument on whether a collateral challenge was permissible, which we had not originally approved as a ground of appeal. We accordingly gave leave to appeal on 5 December 2012 on the following additional grounds:
The appeal was then set down for resumed hearing on these grounds on 14 February 2013.
Did the Judge have power to suppress publication of her judgment?
Our general approach
The original approved ground raised two issues:
whether New Zealand courts have (or had) inherent power to make orders suppressing publication of information relating to proceedings which are binding on non-parties to proceedings (“non-party suppression orders”); and, if so:
the extent to which such power was excluded by s 138 of the Criminal Justice Act.
Although these issues may appear to be logically distinct, they are in fact closely intertwined. For this reason we propose to discuss them in the same section of our reasons.
The leading New Zealand cases are the Court of Appeal’s judgments in Taylor and Broadcasting Corporation of New Zealand v Attorney-General. They, together with the early English case R v Clement, support the existence of an inherent power to make non-party suppression orders.
Counsel for the appellant, however, maintain that such inherent power was displaced by s 138(5) of the Criminal Justice Act and that, in any event, we should not follow Taylor and Broadcasting Corporation given the subsequent enactment of the New Zealand Bill of Rights Act 1990 and developments in other jurisdictions. Their broad position is that an inherent power to make non-party suppression orders is unnecessary in light of the ability to prosecute in relation to any publication which can be shown to have prejudiced the administration of justice.
Against this rather complex background, we have decided to structure our discussion of the two issues set out above under the following headings:
The inherent powers of the New Zealand courts;
Clement, Taylor and Broadcasting Corporation;
Subsequent practice in New Zealand;
The evolution of the relevant statutory provisions;
Did s 138 of the Criminal Justice Act 1985 confer power to make an order suppressing publication of judgments?
Did s 138(5) exclude what might otherwise have been the inherent power to suppress publication of the judgment?
Summary of the position in relation to the interaction between s 138 and the inherent suppression power;
Does it matter that Winkelmann J sat in private when she varied the order?
The approach required by the New Zealand Bill of Rights Act 1990;
Overseas authorities; and
Conclusion on the inherent suppression power.
The inherent powers of the New Zealand courts
All courts in New Zealand have inherent powers. While these powers have in the past sometimes been described as part of the “inherent jurisdiction” of the courts, we think that the term “inherent powers” more aptly describes them. “Jurisdiction” and “power” are two distinct concepts. The jurisdiction of a court is its substantive authority to hear and determine a matter. Jurisdiction may be inherent in a particular court or it may be conferred by statute. But every court has inherent powers which are incidental to or ancillary to its jurisdiction, whether that jurisdiction is inherent or statutory.
In Mafart v Television New Zealand, which did not involve contempt of court, the majority judgment of this Court referred to the ancillary powers of the courts as:
.... the authority of the judiciary to uphold, to protect, and to fulfil the judicial function of administering justice according to law in a regular orderly and effective manner.
The courts’ inherent powers include all, but only, such powers as are necessary to enable a court to act effectively and uphold the administration of justice within its jurisdiction. Their scope extends to preventing abuse of the courts’ processes and protecting the fair trial rights of an accused. The inherent powers of a court do not, however, extend to furthering the general public interest beyond that concerned with the due administration of justice. Examples of the inherent powers which are necessary to enable a court to act effectively within its jurisdiction include powers to dismiss or stay proceedings, to control barristers and solicitors and to issue orders to preserve evidence.
Clement, Taylor and Broadcasting Corporation
Clement involved circumstances where publicity had the potential to threaten the administration of justice by jeopardising fair trial rights. It was a sequel to the prosecution at the Clerkenwell Sessions House of a number of defendants for treason arising out of the Cato Street conspiracy. The Judges (including Abbott CJ, Dallas CJ, Richards CB, and Richardson and Best JJ) directed that each defendant was to be tried separately but with the trials to follow each other in quick succession. At the commencement of the first trial, Abbott CJ (speaking for himself and the other judges) directed that no reports of any trial should be published until all trials were concluded. The direction was, he explained, necessary in the furtherance of justice, to protect the others remaining to be tried. In defiance of this order, Clement, who was the editor of a newspaper, published reports of the first two trials. He was prosecuted for contempt and fined. His challenge to his conviction and attempt to be discharged from the fine were later dismissed in the Courts of King’s Bench and Exchequer.
We see this as an important case. It is perfectly clear that the Judges at the Clerkenwell Sessions House had made an order and not merely given a warning. Indeed the substance of the direction was recorded as an “order of prohibition” and entered in the records of the Court. The power to make such an order was not challenged by any of the Judges in either the Court of Kings Bench or the Court of Exchequer. Including the trial Judges who were present when Abbott CJ pronounced the order, no less than ten of the then twelve common law Judges must have been of the view that there was an inherent power to make non-party suppression orders. It is therefore a case of substantial authority. We also note that the current edition of Arlidge, Eady and Smith on Contempt records that, prior to the Privy Council’s decision in Independent Publishing, which we discuss below, Clement was treated as authority in England that orders might be made in certain circumstances preventing information from being released to the public on either permanent or temporary bases.
Taylor arose out of a prosecution under the Official Secrets Act 1951. At the start of the trial, the trial Judge made an order prohibiting publication of anything that might lead to the identification of members of the New Zealand Security Service. Mr Taylor, who knew of the order, deliberately disclosed the real name of one of the security service officers who had given evidence. He was prosecuted for, and found guilty of, contempt of court and his appeal to the Court of Appeal was dismissed, with Woodhouse J dissenting. For reasons which we will explain later, the Attorney-General did not rely on the then current statutory provisions which permitted the making of certain non-party suppression orders.
The majority (Wild CJ and Richmond J) held that the High Court has an inherent power to order permanent suppression of the names of witnesses in a trial open to the public and that the effect of such an order is not confined to the parties to the case. Richmond J referred to the courts’ inherent powers as existing only “because they are necessary to enable the Courts to act effectively within their jurisdiction in the primary sense”. Richmond J cited in support of this approach to inherent ancillary powers a dictum of Lord Morris:
There can be no doubt that a Court which is endowed with a particular jurisdiction has powers which are necessary to enable it to act effectively within such jurisdiction.
Wild CJ also recognised the Court as having “the inherent power .... to act effectively within its jurisdiction”. Those powers could be exercised “in respect of matters which are regulated by statute or by rule of court, so long as [the Court] can do so without contravening any statutory provision”. Both Wild CJ and Richmond J held that the ancillary power to make orders for the purposes of the administration of justice was sufficiently broad to authorise suppression of the name of a witness in order to preserve the effectiveness of the security service.
Woodhouse J agreed with the majority’s approach to the inherent powers of the Court. He added that such powers arise “in relation to and for the purpose of giving proper support for the functioning of the Court as a Court of justice”. They could be exercised “for the purpose of controlling not only the actions of persons associated with the proceedings but the world at large”. But he differed from the other Judges on the application of these principles because he considered that the purposes of the administration of justice did not encompass the protection of the operations of the security service.
Broadcasting Corporation arose out of a sentencing hearing in respect of which the Judge had made orders suppressing publication of information and had excluded members of the public (including news reporters) from the Court. The orders were intended to provide protection for the defendant who had cooperated with the police and whose sentence was accordingly discounted. The orders were challenged by the Broadcasting Corporation in proceedings which were removed into the Court of Appeal. This touches on a procedural point of some interest to which we will revert later.
In Broadcasting Corporation, the Court of Appeal accepted the existence and nature of the inherent ancillary powers identified in Taylor. The Court held, unanimously, that the orders made were too broad and substituted more appropriate suppression orders. The majority, however, accepted that the Judge had jurisdiction to make orders which went beyond those contemplated by statute. All the Judges proceeded on the basis that Taylor had been correctly decided. This is despite Taylor having been the subject of adverse comments in the House of Lords in Attorney-General v Leveller Magazine Ltd (Leveller), a case which we discuss later in these reasons.
The non-party suppression orders in issue in Taylor and Broadcasting Corporation were permanent and had been made for reasons which at most were only tangentially, if at all, related to fair trial considerations. In contradistinction the order made in the present case was temporary, in that it was expressed to endure only until trial. As well, although no reasons were given, it is perfectly clear that the order in the present case was made to protect the fair trial rights of the defendants. Although we later address whether Taylor and Broadcasting Corporation should be overruled, it must be kept steadily in mind that (a) the issue before us is whether Winkelmann J had power to make the order which the Solicitor-General wishes to enforce and (b) the order she made was both less extensive in scope (because it was only temporary) and more orthodox (in that it was unquestionably based on the protection of fair trial rights) than the orders in Taylor and Broadcasting Corporation.
Subsequent practice in New Zealand
Section 344A was inserted into the Crimes Act 1961 by s 3 of the Crimes Amendment Act 1980, which came into force on 1 January 1981. This provision introduced a formal procedure for pre-trial evidence admissibility (and other) rulings. As well, s 379A provided for pre-trial appeals. From an early stage, the courts acted on the basis that there was power to suppress publication of pre-trial judgments, and the practice of suppressing publication of such judgments, pending trial, soon became virtually universal. The ways in which pre-trial hearings have since been conducted, and subsequent judgments expressed, are based on the premise that such a power is available. As well, the Court of Appeal judges who determined Broadcasting Corporation and those involved in the subsequent evolution of the relevant statutory provisions (which we are about to discuss) must have been well aware of the practice, pending trial, of suppressing publication of pre-trial judgments.
Non-party suppression orders are not confined to criminal cases. They are frequently made in civil proceedings, particularly in cases which involve issues of the same kind as those involved in cases which are the subject of statutory non-publication provisions.
As will become apparent, New Zealand courts have consistently confirmed that they have the inherent power to make non-party suppression orders where the jurisdiction to do so has been challenged.
The evolution of the relevant statutory provisions
The inherent powers of the courts are subject to statutory regulation, limitation or abolition. The present appeal requires us to determine whether or not the inherent power of the courts to make non-party suppression orders has been excluded by statute. This will require some analysis of the legislative history. For reasons which will become apparent, however, it is sufficient for present purposes to start with the legislation as it was in 1975.
At that time, the relevant provisions were s 46 of the Criminal Justice Act 1954 (which was to the same effect as s 140 of the 1985 Act and in very similar language) and, more importantly, s 375 of the Crimes Act 1961. Section 375 of that Act was in these terms:
The first case in which s 375 was considered in any detail was Taylor. At the Official Secrets Act trial which preceded Taylor, the Judge had power under s 46 of the Criminal Justice Act 1954 to suppress publication “in any report” relating to the proceedings, of the names of witnesses (as persons “connected with the proceedings”). As well, it would have been open to the Judge under s 375(2) of the Crimes Act to have suppressed publication of any report or account of the evidence of the security service officer. But Mr Taylor had merely identified the security service officer in question as having given evidence at trial and did not otherwise attempt to provide a “report” of what happened at the trial. Neither s 46 of the Criminal Justice Act nor s 375 of the Crimes Act authorised orders which prohibited publication otherwise than in a “report” or “account” of proceedings. Concerned that the courts might conclude that Mr Taylor’s identification of the officer as a witness was not in the nature of a “report” or “account” of the proceedings, counsel for the Attorney-General sought to justify the orders as having been made under the inherent power of the court. There was also another problem in that the security considerations which motivated the Judge in making the orders were not within the purposes provided for in s 375(1).
In upholding the conviction for contempt, the majority in Taylor also held that the inherent power to make non-party suppression orders had not been abrogated by s 46 of the Criminal Justice Act 1954 or s 375 of the Crimes Act. Woodhouse J was not sure on this point and expressed the view that:
.... the statutory provisions may be intended (so far as they extend) to be in substitution for the pre-existing inherent jurisdiction for the Court.
The only legislative amendment that directly followed Taylor was in 1976, when the phrase “newspaper reporter” in s 375 of the Crimes Act was replaced with “news media reporter”. Thus the statutory scheme under consideration in Broadcasting Corporation was substantially the same as had been under consideration in Taylor.
For present purposes, the important features of Broadcasting Corporation are:
After Broadcasting Corporation, s 375 was recast with effect from 1 July 1983 so as to provide:
We see several of the changes to s 375 as a response to the judgments in Broadcasting Corporation:
The addition of “the security or defence of New Zealand” in the grounds listed in s 375(1) was presumably a result of the lacuna revealed by the Taylor case. In the same vein, the proviso was altered so as not to apply where “the interests of security or defence” so required. But, because the recast s 375 did not expressly say so, we do not see that section as having excluded all common law powers to make non-party suppression orders.
Directly relevant to the present appeal is s 138(1) to (6) of the Criminal Justice Act 1985 which was in force in December 2010 and provided:
Also contextually relevant are other provisions of that Act which provided for suppression orders. The most significant of these was s 140, which relevantly stated:
As will be apparent, s 138 of the Criminal Justice Act was largely taken from the 1983 version of s 375 of the Crimes Act but with some variations.
For the sake of completeness, we note that s 138 of the Criminal Justice Act has now been replaced by s 194 and following of the Criminal Procedure Act 2011. The new sections are far more detailed than their precursors. Of interest:
Section 196(2) of the Criminal Procedure Act provides that every “hearing” (other than one on the papers) is open to the public, subject to ss 97 and 199 (which relate to complainants in sexual cases) and s 197, which allows the clearing of the court, but only for the reasons specified and only where a statutory suppression order is not sufficient to mitigate the risk.
Did s 138 of the Criminal Justice Act 1985 confer power to make an order suppressing publication of judgments?
Section 138 of the Criminal Justice Act and its precursor, s 375 of the Crimes Act, were very much focused on the desirability of trials being open to the public. They sought to regulate the circumstances in which judges might exclude the public and, at the same time, made provision for limiting reports of what transpired in open court, including publication of evidence.
Section 138(1) made it clear that the primary focus of the section was to ensure that courts are open to the public when “sitting” in criminal proceedings. And pursuant to s 138(6), even where the public has been excluded from a hearing, the announcement of the result must take place in public. There is no express statutory power to suppress publication of such a result or the reasons for it. However, s 138(2)(a) and (b) did provide for the making of orders suppressing publication of other material relating to criminal proceedings.
We see s 138(2)(a) as applicable to the reporting of what happens at the sitting of the court in question, that is, the evidence adduced and the submissions made at that sitting. This conclusion follows from the structure of the section, with s 138(2) being a carve out from what would otherwise be the consequences of s 138(1). It is also consistent with the legislative history just discussed. Although s 138(2)(b) permitted prohibitions which are not confined to what might appear in a “report or account” of proceedings, we see it as applicable to the identification of any person who gives evidence at “the sitting” referred to in s 138(1). This reading of the two subsections is consistent with s 138(2)(c), which supports the view that s 138(2) deals with the same general subject matter as s 138(1), namely what happens at a “sitting of [a] court”.
Because Winkelmann J’s judgment contained references to evidence which was intended to be adduced at trial (and may have been adduced at the preliminary hearing), submissions made to her and, to a limited extent, witnesses who would be giving evidence, it might be thought arguable that in those respects the order was within s 138(2)(a) and (b). But even to the extent that her order incidentally precluded publication of what she said about that evidence and, more significantly, those submissions, we do not see it as coming under s 138(2)(a). This is because we see the powers conferred by s 138(2)(a) as extending to prohibition only on direct reporting of the evidence adduced, or submissions made at a hearing and thus not to publication of a judgment in which incidental reference is made to evidence or submissions. To amplify this point, we do not see a judgment which refers to evidence or submissions as being in the nature of a “report” or “account” of proceedings. If it were otherwise, the release of a judgment which referred to evidence or submissions which were subject to suppression orders would, itself, be a breach of the section.
Accordingly we are of the view that Winkelmann J only had the power to make the 9 December 2010 order if there is an extra-statutory, and thus common law, power to suppress publication of judgments.
Did s 138(5) exclude what might otherwise have been the inherent power to suppress publication of the judgment?
Messrs Ellis and Edgeler for the appellant understandably placed heavy reliance on s 138(5), contending that the statutory powers conferred by s 138 to make the orders provided for in s 138(2) displaced what might otherwise have been the inherent power of the courts to make non-party suppression orders. Ms Laracy for the respondent, on the other hand, contended that the exclusion of the inherent power is only in respect of orders of the kind specified in s 138(2) rather than suppression orders generally.
As is apparent, we see s 138(5) as a response to the opinion expressed by Cooke J in Broadcasting Corporation to the effect that the courts retained inherent power to make an order excluding the public from courts which was not subject to the constraint provided by the proviso to s 375(1). Section 138(5) expressed more clearly the intended effect of the 1983 amendment: to repudiate the view that there could be a power under the inherent jurisdiction of the court to make orders of the kind provided for by statute which could operate free of the constraints imposed on the statutory power.
Against this background, the meaning to be attributed to the words “of any kind described in subsection (2) of this section” and “any order of any such kind” is reasonably clear. The “kind” of orders “described” in s 138(2) are orders (a) suppressing publication of evidence adduced or submissions made at the hearing in question, (b) prohibiting the publication of the names of witnesses who give evidence at such hearings, and (c) excluding the public. The words cannot sensibly be read as a generic reference to suppression orders. Significantly, the legislature has never set out to provide for all types of non-party suppression orders which might be necessary to protect fair trial rights. Most obviously, the statutory powers to make non-party suppression orders do not extend to the making of orders suppressing publication of:
evidence which one party wishes to adduce but which is held to be inadmissible; or
a defendant’s prior criminal record.
There are two further points we should mention:
The powers conferred by s 138(6) are not expressly made subject to s 138(5). So even where the announcement of a verdict or decision must be given in public, there is no express statutory provision precluding interim suppression (perhaps pending the determination of other proceedings) of that verdict or decision. Indeed, it might be thought that as s 138(6) permitted a judge, in exceptional circumstances, to decline to give reasons in public, it would be odd if it was not open to such a judge instead to give reasons publicly, but to suppress, at least temporarily, publication of them.
In both Taylor and Broadcasting Corporation, the Court of Appeal held that the express statutory provision to make suppression orders of certain kinds did not exclude the inherent power of the courts to make orders of different kinds. The limited nature of the legislative responses is telling. They show that s 138 was not intended to occupy all the ground as to non-party suppression orders.
Against that background, we consider that much clearer words than those which appear in s 138(5) would be required to exclude the inherent power to make orders of a kind not “described” in s 138(2). We thus conclude that s 138(5) did not exclude the inherent power of the courts to make non-party orders suppressing publication of judgments and rulings.
Summary of the position in relation to the interaction between s 138 and the inherent suppression power
It may be of assistance if we summarise at this point the conclusions we have reached:
Since the 1970s, New Zealand courts, in particular in Taylor and Broadcasting Corporation, have asserted an inherent power to make non-party suppression orders.
The legislature responded to Taylor and Broadcasting Corporation but only in limited respects and has sought neither to enact a general and complete regime for the making of non-party suppression orders, nor to exclude generally the inherent power to make such orders.
Section 138(2) of the Criminal Justice Act did not provide for the making of an order suppressing publication of a judgment.
Since the s 138(5) exclusion of inherent jurisdiction is confined to the making of orders “of the kind” provided for in s 138(2) it did not apply to the order made by Winkelmann J.
Does it matter that Winkelmann J sat in private when she varied the order?
Mr Ellis was critical of the process followed by Winkelmann J when she varied the suppression order having heard from the parties by teleconference. He maintained that the hearing which preceded the variation of the order was required by s 138(1) to have been in public.
We do not accept that judges dealing with ancillary issues were necessarily always required to do so at formal court sittings. Although “every sitting” of a court dealing with criminal proceedings was required to be in public, s 138(1) did not specify that a judge may only deal with an issue ancillary to the determination of criminal proceedings at a “sitting of [the] court”. Plainly the section applied to trials, sentencing and appeals and to final decisions (for example, verdict and sentence). As well, given s 138(6), it also applied to committal proceedings. Remand appearances also must have been within the scope of the section.
But some steps in criminal proceedings are sometimes dealt with informally, either on the papers or by teleconference, such as bail variations which are consented to, adjournment applications and middle-banding and the like. And judgments on ancillary issues associated with criminal proceedings are often reserved and later delivered through the Registrar. The same is true of judgments on appeal. Throughout our professional lives, there have always been some steps associated with criminal proceedings which have occurred otherwise than at a public sitting and we are of the view that s 138(1) cannot have been intended to change that.
Nor do we accept that s 138(6) requires all incidental decisions made in the course of criminal proceedings to be given in public. Like s 138(1), we see it as limited to final decisions and other important stages of a criminal trial.
We see no reason in the present case to reach a definitive conclusion on whether it was open to Winkelmann J to deal with the variation of the order by way of teleconference. In part this is because s 138 has now been replaced by a suite of sections in the Criminal Procedure Act, including s 196, which is not expressed in the same way as s 138(1), meaning that resolution of the issue would have no precedential value. As well, it could not seriously be argued that a breach of s 138(1) in relation to the variation could affect the validity of that order, let alone the initial version of the order which the appellant breached.
The approach required by the New Zealand Bill of Rights Act 1990
We now turn to consider the impact of the New Zealand Bill of Rights Act on the inherent power to make non-party suppression orders. One of the reasons why Mr Ellis invited us not to follow Taylor and Broadcasting Corporation was that they both preceded the enactment of the New Zealand Bill of Rights Act. Mr Ellis submitted for the appellant that freedom of expression should be the “starting point” in assessing whether an inherent power to make suppression orders exists. He argued that the continued existence of a power to make non-party suppression orders was an unjustified limit on freedom of expression. Alternatively, he submitted that if an inherent suppression power exists, it is limited in accordance with the New Zealand Bill of Rights Act. Mr Ellis was also critical of the absence of any analysis of the New Zealand Bill of Rights Act in the judgments given by the Court of Appeal and High Court. We now evaluate those arguments in the context of the particular order made by Winkelmann J, that is, an order of temporary duration which was made to protect the fair trial rights of the defendants.
Developments in the common law must be consistent with the rights and freedoms contained in the New Zealand Bill of Rights Act. Fair trial rights are protected by s 25:
The right to a fair trial is not only a fundamental right of an individual facing criminal proceedings; it is also essential to the administration of criminal justice and the integrity of the courts. Fair trial rights are not the only protected rights at stake, however, when a suppression order is made. The New Zealand Bill of Rights Act also protects freedom of expression:
The right to express ideas, including critical or unpopular opinions, is basic to our democratic system. When a person is prohibited, by a court order, from publishing information, the freedom to express that information is circumscribed. Likewise, the correlative right of the anticipated audience of that publication to receive that information is limited. The right to freedom of expression is not, however, absolute. Also relevant for present purposes is s 5 of the New Zealand Bill of Rights Act, which states:
The New Zealand Bill of Rights Act does not exclude the courts’ inherent power to make suppression orders. Whether or not a suppression order is a limitation on freedom of expression that complies with s 5 will depend on the circumstances of the particular case.
A suppression order can be made consistently with the New Zealand Bill of Rights Act where that represents the appropriate resolution of the tension between freedom of expression and fair trial rights. New Zealand courts have recognised that the right of freedom of expression supports contemporaneous discussion of events in the criminal justice process and must be taken into account along with the right of an accused person to a fair and public hearing by an independent court. Both values must be given serious consideration and, so far as possible, fair trial rights and freedom of expression should each be accommodated. But, where publication of certain information would give rise to a real risk of prejudice to a fair trial right, freedom of expression may be temporarily limited by a suppression order in order to avoid that risk. In our view, this approach properly recognises the special importance of fair trial rights.
An interim ban, pending trial, on the publication of material which gives rise to a real risk of prejudice to a fair trial, is a reasonable limit on the s 14 right of freedom of expression. As a limit imposed by an order of court made under the common law, it is prescribed by law in terms of s 5. As well, the protection of fair trial rights is a sufficiently important objective to warrant a temporary limitation on freedom of expression. The requirement, before a suppression order can properly be made, that publication of the material would create a real risk of prejudice to a fair trial, ensures that suppression orders are only made where that is rationally connected to the objective of protecting fair trial rights. Fair trial rights are important and, where there is a real risk that they will be negated, a pre-emptive but temporary publication ban is a reasonable and proportionate limit on freedom of expression, to avoid that risk. The scope of such a suppression order (for example, the material suppressed or the duration of the order) should be defined in such a way that ensures freedom of expression is limited only to the extent reasonably necessary to preserve fair trial rights.
It is not, however, necessary or appropriate for us to consider whether or not the order made by Winkelmann J in this case was consistent with these principles. For reasons we give later, the rule against collateral challenge precludes such an analysis in the context of contempt proceedings.
The remaining question is whether prosecution and punishment of breaches of court orders under the law of contempt of court is a reasonable limit of freedom of speech under s 5. We deal later with that issue.
Next we consider the approach that has been taken to the courts’ inherent powers overseas. Supporting the existence of an inherent power to make non-party suppression orders, and thus consistent with the New Zealand jurisprudence, is the judgment of the Supreme Court of Canada in Dagenais v Canadian Broadcasting Corporation. This case started with an application to prevent the broadcasting of a particular programme which was thought likely to prejudice trials that were currently underway and pending. But in dealing with the appropriateness of the orders made, the Supreme Court confirmed the long-standing common law position that a judge hearing (or scheduled to hear) criminal proceedings has a power to issue non-party suppression orders. Significantly, Dagenais was decided, and the inherent suppression power confirmed, in the context of the Canadian Charter of Rights and Freedoms which contains provisions equivalent to those in our New Zealand Bill of Rights Act which provide for freedom of expression but also the right to a fair trial.
Turning to the position in England and Wales, Clement, which we have already discussed in detail, is no longer seen as authoritative. This is a consequence of the decisions of the House of Lords in Leveller and the Privy Council in Independent Publishing.
Leveller involved a situation which was similar to that in Taylor, in that it concerned the identification of a prosecution witness and there was a security overlay. The magistrates had allowed the witness to give evidence without revealing his name but information, which he provided when giving evidence, made it possible for his identity to be ascertained, as it was. It was later published. A prosecution for contempt resulted. In allowing the publisher’s appeal, some of the Judges discussed whether a non-party suppression order could be made and in this context Taylor was mentioned by Lord Diplock and Viscount Dilhorne. Lord Diplock did not express a definitive opinion but Viscount Dilhorne was unequivocally of the view that Taylor did not represent the law of England. Clement was not discussed.
One result of Leveller was the enactment of ss 4(2) and 11 of the Contempt of Court Act 1981 (UK). Section 4(2) provides for orders postponing publication of prejudicial material and s 11 addresses the particular situation which arose in Leveller. Together they provide a statutory framework for the making of non-party suppression orders. Later, when the trial judges in England and Wales were given statutory powers to make certain types of preliminary rulings in criminal cases and rights of appeal were provided, statutory restrictions on publication were enacted at the same time. This was in marked contrast to what happened in New Zealand when s 344A of the Crimes Act was enacted. When considering the applicability of decisions from England and Wales it is important to recognise that the statutory regime in that jurisdiction has been premised on an absence of an inherent power to make non-party suppression orders whereas in New Zealand, the statutory scheme has evolved against a background in which the courts have asserted a general inherent power to make such orders.
The common law position in England and Wales was extensively reviewed by the Privy Council (on appeal from Trinidad and Tobago) in Independent Publishing. The Privy Council judgment contains a survey of the cases, including Clement, Taylor and the Canadian and Australian authorities. Clement was in part explained on the basis that the prosecution was based not just on the breach of the order but also on the tendency of the publication to prevent or obstruct the course of justice. Some doubt was expressed as to whether what Abbott CJ had said was in the nature of an order (as opposed to a warning). To the extent that Clement was authority for the proposition that there is an inherent power to make non-party suppression orders, it was overruled. Dagenais was discounted on the basis that what was actually in issue in the case was an interlocutory injunction to prevent a contempt of court. Taylor was seen as wrongly decided.
The Australian authorities generally support the existence of an inherent power to make temporary non-party suppression orders where such orders are necessary to preserve fair trial rights. This is the position which has been adopted by the Court of Appeal of Victoria. The Court of Appeal of New South Wales has also upheld the existence of such an inherent power but has indicated that judges should be cautious about its exercise. In Hogan v Hinch, French CJ expressed the view that there is inherent jurisdiction or implied power in limited circumstances to restrict the publication of proceedings conducted in open court. The other Judges left the issue open.
Conclusion on the inherent suppression power
For the avoidance of doubt, we reiterate that we are addressing in this case the power to make an order of the kind made by Winkelmann J, that is, an order of temporary duration which was made to protect the fair trial rights of the defendants. On the appellant’s case, there is no inherent power to make non-party suppression orders but any publication that prejudices the administration of justice may be addressed, after the fact, by prosecution for contempt of court. On the respondent’s argument, the courts have a power to make non-party suppression orders, breach of which is punishable by contempt irrespective of whether actual prejudice to the administration of justice has been established.
Our discussion of the New Zealand cases indicates that, since the 1970s, New Zealand courts have exercised the power to make non-party suppression orders which go beyond anything provided for by statute. We have demonstrated that this power has not been extinguished by either the Criminal Justice Act or by any earlier enactment. Neither s 138 of the Criminal Justice Act nor the provisions in the Criminal Procedure Act purport to provide anything like a code in relation to non-party suppression orders. There is thus a pattern of legislative action and inaction founded on the assumption that the courts have the power to make non-party suppression orders. And the way in which criminal courts deal with pre-trial applications and appeals in part reflects an assumption that non-party suppression orders promote fair trial rights.
As explained, we are of the view that the New Zealand Bill of Rights Act (which post-dated both Taylor and Broadcasting Corporation) does not exclude the continued existence of the inherent power to make suppression orders.
We are not persuaded by Leveller or Independent Publishing that we should depart now from the established New Zealand approach. In Broadcasting Corporation, the Court of Appeal followed Taylor’s conception of the inherent powers, despite what was said in Leveller. In Muir v Commissioner of Inland Revenue, the Court of Appeal continued to follow Taylor, despite Independent Publishing. In Mafart, the majority judgment of this Court approved Taylor’s explanation of the inherent ancillary powers. Furthermore, the New Zealand approach is consistent with Clement and Dagenais. Very importantly, since 1975, the New Zealand statutory regime has been developed against an assumption that the courts have inherent power to make non-party suppression orders.
We accept that risks to fair trial rights can be mitigated otherwise than by using non-party suppression orders. The default position is that publication of material which is likely to interfere with the administration of justice is punishable as contempt of court. This approach can be practically reinforced by judicial warnings about publication and perhaps a practice that publication of what happens when the jury has been excluded from the courtroom is presumptively prohibited, so that publication will be treated automatically as a contempt of court. As well, there are other mechanisms which might be deployed. For instance, a case might be heard in camera (although where this happened s 138(6) applied). Witnesses might be permitted to give evidence without disclosing their names or otherwise in circumstances which should protect their anonymity. As well, there is a jurisdiction to grant an injunction to restrain a threatened contempt of court.
These mechanisms, however, are by no means fail-safe and to date the New Zealand view has been that they do not adequately protect fair trial rights and the administration of justice. We consider that this view is well-founded:
A system which leaves publication decisions (particularly the assessment whether a publication will prejudice fair trial rights) entirely to third parties (who may be neither dispassionate nor fully informed) creates a risk that those third parties will get it wrong, resulting in prejudice to fair trial rights which cannot be remedied, after the fact, by prosecution for contempt of court.
There is not much point permitting witnesses to give evidence anonymously if those involved with, or who take an interest in, the case and either know, or can find out, their identities, can publish those names with impunity.
Assumptions as to the extent of the power to make non-party suppression orders are material to the way in which litigation is conducted and the extent to which the legislature sees it necessary to provide a statutory basis for non-party suppression orders. For instance, a decision by a judge not to sit in camera may be influenced by the availability of a power to suppress publication of the sensitive aspects of what happens in court. The willingness of judges to deal with evidential and other issues pre-trial may be diminished if there is no power to make non-party suppression orders. Pre-trial judgments which are susceptible to publication may be expressed in more guarded terms than those which can be suppressed. In these ways, the absence of any inherent power to make non-party suppression orders may have undesirable downstream consequences for the way in which the justice system operates.
We note in passing that there is no general statutory power to suppress information associated with civil proceedings. Yet such orders are not infrequently made. If it should transpire that there is no inherent power to suppress such information, there must be a very substantial number of orders which will necessarily be rendered invalid and thus a large number of people who will have conducted litigation on inaccurate assumptions as to suppression.
Given the temporary nature of the order made in the present case, and the desirability of protecting fair trial rights, we consider that these factors mean that the rule and practice so far adopted by the New Zealand courts is appropriate. A power to make such orders is necessary for the administration of justice and the protection of fair trial rights and, as we have explained, has not been excluded by statute. Against that background, we do not propose to overrule Taylor and Broadcasting Corporation.
It follows that we are satisfied that Winkelmann J had an inherent power to suppress publication of her judgment of 9 December 2010.
Our general approach to the additional grounds of appeal
In the second half of our judgment we address the two additional grounds of appeal in turn. The first is whether a person who wishes to act in a manner contrary to a suppression order may seek to have it rescinded or varied. The second is whether, in contempt proceedings based on breach of an order of court, the defendant may raise as a defence that the order should not have been made or made in the terms it was. After addressing the first additional ground of appeal, we structure our approach to the second additional ground under the following headings:
Contempt of court based on breach of a court order;
New Zealand authority on the rule against collateral challenge;
The United Kingdom position;
Does Boddington apply?
The position in Canada;
Exceptions to the rule against collateral challenge;
The continuing operation of the rule against collateral challenge;
Contempt of court and the rule against collateral challenge: justified limitations; and
Application of these principles in the present case.
Who has standing to apply to the court for variation or rescission of a suppression order?
The first additional ground of appeal raises the issue of whether a person who wishes to act in a manner contrary to a suppression order may seek to have it rescinded or varied.
The rule of law requires that court orders are not immunised from challenge or review. There should be some means by which, where an order is made in error, it can be corrected.
Further, Canadian cases have recognised that when court orders affect the rights of individuals the law should provide a means to vindicate those rights. In R v Domm, the Court of Appeal of Ontario described this as a “remedial” dimension to the rule of law. Doherty JA, delivering the Court’s judgment, said:
The rule of law, however, does more than demand compliance with the law. To validate this demand, the law must provide individuals with meaningful access to independent courts with the power to enforce the law by granting appropriate and effective remedies to those individuals whose rights have been violated.
One means of accommodating those who, not being party to the proceedings, maintain that an order of the court has been made that inappropriately affects their rights is to allow them to seek variation or rescission of the order. New Zealand has taken that approach in relation to review of suppression orders. In Taylor, Richmond J said that he had no doubt that Mr Taylor, had he been so advised, could have applied to the Court for a variation or termination of the order suppressing the name of the witness at the earlier trial. Mr Taylor, of course, was not a party to the proceedings in which the suppression order was made. Following Taylor, it has become well-established in New Zealand that media interests will be heard by the courts on the issue of suppression and may apply for suppression orders to be discharged, rescinded or varied. The decision of the Court of Appeal in the Broadcasting Corporation case was an early decision on just such an application. The same approach has been followed without comment in this Court.
Both the Court of Appeal and the High Court were of the view that it was open to persons in the position of the appellant to apply to the Court for a review of the suppression order that was binding on him. This proposition is supported by Richmond J’s judgment in Taylor, which was endorsed by Cooke J in Broadcasting Corporation. We agree that in principle, and subject to any legislative provision, New Zealand law should continue to permit any member of the public, who wishes to publish material which may not be published under a suppression order of general application, to approach the registrar of the court which made the order, seeking its variation or rescission. The approach must be made in writing and set out the reasons why review of the order, or its application to that person, is sought.
Once such an application is made, the requirements of natural justice, which will vary with the circumstances, must be complied with. The application must be considered by a judge, who first will decide whether the parties to the proceedings in which the order was made should have the opportunity to respond and generally how the application should be addressed. These matters, including whether the applicant should be heard orally before the application is decided, will always depend on the circumstances of the case, such as the nature of the interests at stake and any potential disruptive effect on continuing proceedings.
This opportunity for individuals to seek informally amendment to or discharge of a suppression order operates in tandem with the right of the media to be heard on the matter, ensuring that court orders are susceptible to appropriate review, and that freedom of expression is given proper consideration and effect. We are satisfied that, in the New Zealand context, this meets the requirements of the rule of law and natural justice.
Statutory rights of appeal vary depending on whether the subject matter is criminal or civil. Currently the rights of appeal from the High Court to the Court of Appeal in civil cases are governed by s 66 of the Judicature Act 1908 whereas rights of appeal in criminal cases from both the High Court and District Court in relation to cases dealt with on indictment are provided for in the Crimes Act 1961. Appeals from the District Court in civil cases are provided for in the District Courts Act 1957 whereas appeals from the District Court in respect of the summary trial of offences are addressed in the Summary Proceedings Act 1957.
The legislature has generally specified particular criminal appeal processes in respect of statutory suppression orders which, at present, do not extend to provide the media with rights of appeal. As will be apparent, this has not prevented the media being heard in respect of such orders.
There is no explicit provision in any enactment for rights of appeal in respect of non-statutory suppression orders. The position is similar in Canada and Canadian courts have held that this means that such orders are not subject to appeal to the Ontario Court of Appeal. On the other hand, in Victoria, the Court of Appeal has held that such appeals are competent, essentially as civil appeals. The relevant statutory provisions in issue in the Victorian case have some differences from those in New Zealand but the approach taken by the Court of Appeal is broadly similar to the approach of this Court in Mafart. Against this background, it is certainly arguable that there is a right of appeal under s 66 of the Judicature Act in respect of non-party suppression orders made in criminal cases. Whether that is so does not arise for determination in the present case as no-one attempted to appeal against the order made by Winkelmann J.
For completeness, we observe that, under the provisions of the Criminal Procedure Act concerning the standing of those who are “members of the media”, such persons will enjoy special rights of standing in relation to suppression orders. That group, which is defined in a way that includes any persons reporting on the proceedings with the permission of the Court, will enjoy the right to seek variation or revocation of a suppression order. The effect of this statute on the common law position we have described remains to be worked out in future cases. It is not relevant to the present case.
The rule against collateral challenge
Contempt of court based on breach of a court order
Breach of a court order will generally constitute contempt of court. Where contempt proceedings are based upon publication of information subject to a suppression order, it is not necessary to show, in addition to breach of the order, that the publication in fact interfered with fair trial rights or another aspect of the administration of justice. To the extent that Leveller or Independent Publishing suggest that an actual interference with the administration of justice is necessary to establish liability for contempt, the approach taken in those cases was influenced by the view that there was no power for courts to make binding non-publication orders, breach of which could give rise to contempt proceedings. In contra-distinction, the predominant contemporary view in England and Wales is that breach of a non-publication order made under the Contempt of Court Act constitutes contempt without there being any necessity to show an interference with the administration of justice. As the authors of Arlidge, Eady and Smith on Contempt explain:
.... since a s 4(2) order can only be made for the purpose of protecting the administration of justice, it is difficult to see how a publication of prohibited material with knowledge of such an order could be other than a contempt.
The same reasoning applies to orders made under the courts’ inherent powers in New Zealand.
New Zealand authority on the rule against collateral challenge
In that context, we next address the issue raised by the second additional ground of appeal, as to whether a defendant can defend a contempt proceeding on the basis that the order that has been breached should not have been made at all or in the terms it was. Such a defence would essentially involve a collateral attack to a court order, in the form of a challenge brought in proceedings which are not directed to rescinding, varying or setting aside the order. A rule against such collateral challenge in contempt proceedings based on breach of a court order was recognised by Richmond J in Taylor, where he said:
I have no doubt that the appellant, had he been so advised, could have applied to the [High] Court for a variation or termination of the order made by Beattie J. But once it be found that the order was within the jurisdiction of the Judge to make then the appellant could not challenge it by disobedience.
Although dissenting in the outcome, Woodhouse J also indicated his support for this principle, and its application to orders made not just against parties but against all the world. Taylor involved such an order. In explaining the importance of court orders being expressed in clear and unambiguous terms, he described the effect of the order made in Taylor in the following terms:
It is not concerned merely with the interests of parties or individuals. It is addressed to the world at large. If made with jurisdiction a failure to obey it can be made the subject of the great coercive powers of proceedings for contempt.
In the present appeal, the parties accepted that the appellant had the right in the contempt proceedings to challenge whether the Court had power to make an order of the kind made by Winkelmann J. Because we have decided that Winkelmann J did have that power, it is not necessary for us to express a view on whether in New Zealand a defence will be available to a charge of contempt where it can be established that the court did not have the power to make an order of the relevant kind.
Provided the court had power to make an order of its kind, a court order is binding and conclusive unless and until it is set aside on appeal or for other reason lawfully quashed. Collateral attacks on such orders are not permitted. Neither the parties, nor other persons subject to an order, are permitted to arrange their affairs in accordance with their perceptions of its flaws, including any individual views they may have concerning the validity of the order. The position is the same whether the order has been made in the High Court or in the District Court.
The rules that breach of a court order will constitute contempt of court and that collateral challenge in contempt proceedings is not permitted maintain stability in the law and protects the ability of the courts to exercise their constitutional role of upholding the rule of law. As previously explained in this Court:
Effective administration of justice under our constitution requires that the orders of the courts are obeyed unless properly challenged or set aside. Public confidence in the administration of the law, also necessary for its effective administration, requires that there is a strong expectation that those who ignore court orders are quickly brought to account. Achieving these aims is part of the objective of the law of contempt.
The rule against collateral challenge and the importance of the reasons which underpin it are also recognised in other common law jurisdictions.
The United Kingdom position
The origins of the rule against collateral challenge lie in such early authorities as Woodward v Lincoln (Earl) and Russell v East Anglian Railway Co. In England, the judgment of Romer LJ in Hadkinson v Hadkinson is also regularly cited in support of the rule against allowing collateral attacks on court orders. Mrs Hadkinson had caused her son to be removed from England to Australia in breach of a custody order directing that the child should not be taken out of the jurisdiction without the Court’s consent. She then appealed against an order of court obtained by Mr Hadkinson requiring her to return the child. Mr Hadkinson made the preliminary objection that the Court of Appeal should not hear the appeal because the mother was in contempt of court.
A majority of the Court of Appeal was of the view that Mrs Hadkinson had disobeyed the order and was in continuing contempt. Her appeal would not be heard until she had taken the first step of purging her contempt by returning the child to England.
In his judgment, Romer LJ said:
It is the plain and unqualified obligation of every person against, or in respect of whom, an order is made by a court of competent jurisdiction, to obey it unless and until that order is discharged. The uncompromising nature of this obligation is shown by the fact that it extends even to cases where the person affected by an order believes it to be irregular or even void. “A party who knows of an order, whether null and void, regular or irregular, cannot be permitted to disobey it .... It would be most dangerous to hold that the suitors, or their solicitors, could themselves judge whether an order was null and void – whether it was regular or irregular. That they should come to the court and not take upon themselves to determine such a question: that the course of a party knowing of an order which was null and irregular and who might be affected by it was plain. He should apply to the court that it might be discharged. As long as it existed it must not be disobeyed.” (Per Lord Cottenham L.C. in Chuck v Cremer.)
The majority in Hadkinson decided that Mrs Hadkinson’s breach of the court order was a contempt, putting her at risk of punishment. The only consequence determined by the Court of Appeal, however, was that she could not bring her appeal until she had purged herself of the contempt. The relevance of the case to the rule against collateral challenge lies in the reasoning of Romer LJ that persons knowing of orders that they believed were defective could not disobey them. They should apply to have the orders discharged but, as Lord Cottenham had said in Chuck v Cremer, while the order existed it had to be obeyed.
The passage from the judgment of Romer LJ was cited and applied in Isaacs v Robertson, a judgment of the Privy Council, which is direct authority for the rule against collateral challenges. The defendant failed to comply with an interim injunction restraining him from trespassing on land. The plaintiff applied for committal of the defendant for his contempt in failing to obey the Court’s order. The Court of Appeal of St Vincent held that the order of injunction ought not have been made and the defendant would have been entitled to succeed had he applied to have the order set aside but he had not. Accordingly, he was in contempt in disobeying the order while it was in force.
On appeal, the Privy Council agreed that defective orders had to be obeyed even though they had been made improperly. Lord Diplock adopted the passage from Hadkinson set out above in giving the Privy Council’s judgment on the point, observing that it “says all that needs to be said upon this topic” and was “in itself sufficient reason for dismissing this appeal”.
In 1994, the House of Lords applied Isaac v Robertson in Re M. That case concerned contempt proceedings brought by a person who had been refused political asylum, against the Home Office and Home Secretary. The plaintiff was deported before his applications to bring judicial review proceedings were finally determined. A High Court Judge ordered that he be returned to the United Kingdom. The order was set aside a few days after it was made on the ground that the Judge did not have power to make the order. But actions of the Home Secretary, while it was in force were held, in the contempt proceedings, to be in breach of the Court’s order. In the Court of Appeal, Lord Donaldson MR said that, although the order should not have been made, it did not follow that it was made without jurisdiction. That was because the High Court was a Court of unlimited jurisdiction. He cited Isaacs v Robertson as authority for the proposition that:
An order which is made by a court with unlimited jurisdiction is binding unless and until it is set aside. Common sense suggests that this must be so. Were it otherwise court orders would be consistently ignored in the belief, sometimes justified, that at some time in the future they would be set aside.
The Master of the Rolls added that the binding effect of orders of courts of unlimited jurisdiction should cause no problem for those subject to them in deciding what to do. There were, however, situations that could produce a dilemma for persons bound by orders where compliance with the order would render nugatory the right to apply to have the order set aside. In such situations, the burden of decision was:
.... firmly on the person to whom the order is addressed. If he does not immediately comply with its terms, he will technically be in contempt. If, however, he preserves the situation pending the result of a prompt application to the court, no penalty would be imposed if the order were subsequently set aside and in all likelihood no penalty would be imposed if, in the event of it being affirmed, there was full and immediate compliance. On the other hand he would be liable to severe penalties if, pending such an application, he acted to frustrate the order, whether or not it was affirmed.
This limited indication of flexibility in how the courts should deal with breaches of their orders was, however, confined to the question of penalty rather than of whether they were in contempt.
Both sides appealed. The judgment of the House of Lords delivered by Lord Woolf was unanimous in dismissing the appeal. On the point in issue, the House of Lords held that, there being jurisdiction for the order made by the Judge, it was not a defence to suggest that it was inappropriate for him to make it.
Lord Woolf nevertheless did conceive of a means of providing flexibility in application of the strict rule requiring obedience. He referred to the reference by Lord Donaldson to “technical contempt” and said:
.... it is undesirable to talk in the terms of technical contempt. The courts only make a finding of contempt if there is conduct by the person or body concerned which can, with justification, be categorised as contempt. If, therefore, there is a situation in which the view is properly taken (and usually this will only be possible when the action is taken in accordance with legal advice) that it is reasonable to defer complying with an order of the court until application is made to the court for further guidance then it will not be contempt to defer complying with the order until an application has been made to the court to discharge the order. However, this course can only be justified if the application is made at the first practicable opportunity and in the meantime all appropriate steps have been taken to ensure that the person in whose favour the order was made will not be disadvantaged pending the hearing of the application.
The Home Secretary was unable to justify his actions under this formulation, in part because he had not taken adequate steps to protect the position of M pending application to the Court.
Does Boddington apply?
In his submissions on the second additional ground, Mr Ellis also relied on Boddington v British Transport Police, in which the House of Lords upheld the right to challenge the validity of by-laws, and administrative decisions made pursuant to by-laws, as a defence to a prosecution for breach. As the judgments in Isaacs v Robertson and in Re M preceded that in Boddington, we accept that it is necessary to decide whether its approach should be applied by analogy to court orders.
The defendant had been convicted for breaching a by-law by smoking a cigarette in a railway carriage where smoking was prohibited by British Railways Board by-laws. Underlying the reasons of the Law Lords is their concern that both the rule of law and the requirement of fairness to accused persons facing criminal charges, required that defendants have a fair opportunity to challenge legal measures promulgated by executive bodies and to vindicate their rights in court proceedings. Judicial review of the lawfulness of the subordinate legislation and administrative decisions involved was inadequate for these purposes as it was not a realistic or satisfactory option for most defendants. In those circumstances, it was held that the public interest in orderly administration had to be circumscribed to accommodate the constitutional importance of defendants to criminal charges being able to defend themselves by contending there had been unlawful action by the public body.
Boddington has not altered the English approach to the rule against collateral challenges to court orders in contempt cases. In Bell v Tuohy, Neuberger J cited Isaacs v Robertson in affirming that an order made by a judge of unlimited jurisdiction must be obeyed. Failure to do so could amount to contempt of court, however irregular the order might be, unless and until it was set aside. In that case, however, collateral challenge was permitted because issue of the warrant for possession in question was an administrative act, not a judicial one, and the order had been made by a court of limited jurisdiction. Forresters Ketley v Brent similarly affirmed Lord Diplock’s statement in Isaacs v Robertson that the defendant had an obligation to obey a court order unless it was set aside. The Court refused to entertain submissions directed at challenging orders that had been set aside. These Court of Appeal decisions indicate that the application of the rule against collateral challenge to court orders continues unchanged by the different approach taken, since Boddington, to administrative actions. The authors of Arlidge, Eady and Smith on Contempt also confirm that the general rule against collateral challenge to court orders endures in the United Kingdom, despite Boddington.
As well, we accept Ms Laracy’s submission for the respondent that it would be inapt to draw an analogy between Boddington and the circumstances of the present case. The constitutional position with court orders is different from that of a collateral challenge to an executive or administrative action, and a different approach is required. The common law rule against collateral challenge to court orders is itself premised on the centrality of the rule of law. The special need in our society for compliance with judicial orders is the constitutional reason for treating disobedience of court orders differently from conduct in breach of subordinate legislation or administrative directive. No such rule of law concerns are raised when a defendant, charged in a criminal court with breach of the directions of an executive government agency, raises their unlawfulness, and resulting invalidity, as a defence. The defendant, of course, in such a case takes the risk of incurring a criminal conviction and punishment if his or her expectation of illegality proves to be wrong. Disobeying the government’s administrative directions in this way does not impede the exercise of its functions. But if disobedience of court orders in that way were to be tolerated, the Court’s authority and ability to discharge its functions would become seriously impaired. For that reason, the common law of contempt by disobedience of a court order rests on the existence of an order that was made with legal authority, and was thus lawful, as we have held in the present case. It does not matter whether or not the order should have been made at all or made in those terms.
For these reasons, Boddington is distinguished. It does not support the type of collateral challenge that the appellant seeks to make in the present case. These United Kingdom authorities affirm the rule against collateral challenge to court orders. We respond below to the suggestion in Re M that there should be some flexibility in the rule.
The position in Canada
A rule against collateral challenge is also recognised in Canada. The judgment of the Supreme Court of Canada in Wilson v Rdid not concern a case of contempt. A trial judge had ruled that the decision of a Supreme Court to authorise a wiretap was wrong and had excluded evidence thereby obtained. On a Crown appeal against the subsequent acquittal of the accused, the Supreme Court held that the superior court’s decision had to be treated as “absolute verity” so long as it stands unreversed. McIntyre J, in his judgment, said:
It has long been a fundamental rule that a court order, made by a court having jurisdiction to make it, stands and is binding and conclusive unless it is set aside on appeal or lawfully quashed. It is also well settled in the authorities that such an order may not be attacked collaterally – and a collateral attack may be described as an attack made in proceedings other than those whose specific object is the reversal, variation, or nullification of the order or judgment.
McIntyre J went on to refer to authorities which had discussed whether, despite the fundamental rule, court orders could subsequently be attacked on the ground of nullity. The answer that had been given to that submission, with which McIntyre J expressed agreement, was that the order of a superior court was never a nullity. He said that the authorities:
.... confirm the well-established and fundamentally important rule .... that an order of a court which has not been set aside or varied on appeal may not be collaterally attacked and must receive full effect according to its terms.
He added that, because the authorisation of a wiretap was granted on ex parte application, it could be reviewed on application to the court, and he outlined the process by which this could occur.
Dickson and Chouinard JJ agreed, while pointing out that the general rule against collateral challenge was subject to modification by statute.
Subsequent decisions of the Supreme Court of Canada in cases involving contempt have affirmed the approach taken in Wilson. The Court has given particular emphasis to the duty of a person bound by an order of court to obey the order:
The duty of a person bound by an order of court is to obey the order while it remains in force regardless of how flawed he may consider it or how flawed in fact it may be. Public order demands it be negated by due process of law, not by disobedience.
The reasons for the Canadian position are similar to those informing the New Zealand rule. Referring to the rule against collateral challenge, Iacobucci J said in R v Litchfield:
The rationale behind the rule is powerful: the rule seeks to maintain the rule of law and to preserve the repute of the administration of justice. To allow parties to govern their affairs according to their perception of matters such as the jurisdiction of the court issuing the order would result in uncertainty. Further, “the orderly and functional administration of justice” requires that court orders be considered final and binding unless they are reversed on appeal.
Exceptions to the rule against collateral challenge
The Supreme Court of Canada has concluded that maintaining the rule of law may in some circumstances also require limited exceptions to the rule against collateral attack. As LeBel J said delivering the majority judgment of the Supreme Court of Canada in Attorney-General of Quebec v Laroche:
Limited exceptions, controlled by the courts, to the rule against collateral attacks do not offend the principle of the stability of judicial decisions, which retains its place as a central element in a sound administration of justice and an orderly judicial system. Recognition of such exceptions, on the other hand, also allows for the integrity of the fundamental rule of law to be maintained, by ensuring compliance with that rule in situations where constitutional rights would otherwise be severely impaired, absent such a remedy.
The Supreme Court of Canada cited Domm, in which the Ontario Court opined that such an exception to the collateral challenge bar may be necessary where a person has no effective means to challenge a court order or where compliance with the court order, would result in an “irretrievable loss of a right which could not be salvaged by a subsequent setting aside of the order”.
To illustrate this approach, the Ontario Court in Domm referred to its earlier decision in R v Fields. Mr Fields had been a witness at a trial who was convicted of contempt in the face of the Court after refusing to answer questions about his political affiliations. On his appeal against his conviction he was held to be entitled to raise the issue of the relevancy of the questions, despite that being an attack on the correctness of the trial Court’s direction. The Court of Appeal in Domm said that, in those circumstances, Mr Fields’ right to refuse to answer irrelevant questions would have been “irretrievably lost” if he had complied with the direction to answer the question. As well, Mr Fields had no available means to challenge the validity of the judge’s order made in the course of the trial. We note that in New Zealand the refusal of a witness to answer questions in criminal proceedings and the ground on which this is permissible was governed by s 352 of the Crimes Act, which has now been replaced by s 165 of the Criminal Procedure Act.
In Domm, however, it could not be said that the defendant lacked alternative means to challenge the suppression order in question. The defendant had a right of appeal against the suppression order but had chosen not to exercise it. As well, requiring the defendant to comply with the order and resort to the appeal procedure would not have resulted in an “irretrievable loss” of his freedom of expression. The Court accepted that in some cases delay in publication may effectively amount to an absolute ban, for example, if by the time the ban is lifted, the speaker has lost his audience and his message has lost its purpose. But Domm was not such a case: his purpose of informing the public of what goes on in criminal courts would still have been served by publication after a challenge by appeal. In the absence of either of these exceptions, the Court upheld the rule against collateral challenge. Significantly, Domm, like the present case, involved an order made against all the world.
The 1929 judgment of the High Court of Australia in James v Cowan concerns contempt proceedings. A public servant had been subpoenaed to produce minutes of books in his possession in an action in the High Court of Australia. He refused to do so and was committed to prison for contempt of court by Starke J. An appeal to the Full Court of the High Court was dismissed. The Court dealt with the collateral challenge point succinctly:
Much of the argument in support of the appeal was addressed to the questions as to what use could be made of the books when brought into Court and whether the order is one we ourselves would make, but it is not necessary for us to express an opinion on these questions because it is enough for us to say that the order to bring the books into Court as directed by the writ of subpoena was made by a competent Court, and that a refusal to obey that order was a defiance of the authority of the Court and therefore a contempt of Court.
More recently, in Jackson v Sterling Industries Ltd, the High Court of Australia had to consider whether an order by a Judge of the Federal Court who had ordered that the respondent provide security in the sum of $3,000,000 had jurisdiction to make that order. A majority of the High Court decided that the order went beyond the Federal Court’s powers and allowed the appeal. As there were contempt proceedings pending against the respondent, the majority judges went on to indicate the effect on those proceedings of the invalidity of the Federal Court order. Each Judge said that it should be no excuse that the order had been made in the improper exercise of jurisdiction. Wilson and Dawson JJ cited Russell v East Anglian Railway Co when saying:
The principle remains, however, that the order of a competent court must be obeyed whilst it remains in force.
The High Court has since confirmed that orders by a superior court of record are valid until set aside. The ban on collateral challenges of superior court orders also remains. Kirby J has on occasion suggested there should be flexibility in that rule but the circumstances in which he would allow such exceptions would not undermine it.
The continuing operation of the rule against collateral challenge
The approach in the overseas common law jurisdictions that we have considered is consistent with the expression of the rule by Richmond J in Taylor. No subsequent cases in New Zealand take a different position. We accordingly confirm the continuing application in New Zealand of the rule against collateral challenge in order that the authority of the courts and the rule of law are maintained. Provided the court had power to make an order of the relevant kind, it is not open to a person facing contempt proceedings based on breach of a court order to establish a defence, by collateral attack, on the basis that the order should not have been made, or made in the terms it was. The rule applies even where the court order in question was an order made ex parte or against the whole world, binding persons who did not have an opportunity to be heard before the order was made.
The application of the rule should, however, be consistent with its underlying objectives. We have recognised that, so far as practicable, there should be a process by which a person who has become subject to an order of court, without being heard, can approach the court seeking variation or rescission of the order concerned. In addition, we now indicate that certain aspects of the law of contempt by disobeying court orders, which are recognised in the United Kingdom and Canada, should, in the interests of justice, apply in New Zealand.
As the judgment of the House of Lords in Re M pointed out, before holding a breach of a judicial order to be a contempt, the court must be satisfied that the conduct in issue is of a contemptuous nature. If a person who is bound by a court order but who was not heard at the time the order was made makes an immediate application to the court seeking variation or rescission of the order, or having a right of appeal brings an appeal promptly, the court hearing contempt proceedings may decide that non-complying conduct pending determination of the challenge to the order, is not contemptuous. This will be dependent on the person bound not acting to the disadvantage of persons whose interests the order seeks to protect or in a way that frustrates the effect of the order if it is affirmed. This approach helps to ensure that there is a meaningful and practically available opportunity for those subject to court orders to challenge them in proceedings directed to that issue. It is not strictly a collateral challenge.
We would also recognise that, while the rule against collateral challenges continues to apply in New Zealand, a person who has no other available and effective means to seek reconsideration or review of a court order by which he or she is bound should be able to challenge that order in contempt proceedings brought because of failure to comply with that order. The interests of justice require that, before a person who is not a party to proceedings in which the order is made is punished for contempt, he or she has an opportunity to challenge the lawfulness of the order. This limited exception to the rule against collateral challenges to court orders will apply where challenging an order in the course of contempt proceedings is the only meaningful opportunity for a person to do so. Such situations will normally only arise where the defendant is not party to the proceedings in which the order is made, there is no time to seek variation or rescission or otherwise challenge such an order, and where to defer compliance and follow an appeal or review procedure, whether formal or informal, would result in the “irretrievable loss” of an important right. The defence will fail if the defendant’s non-complying conduct goes beyond what is needed to avoid such a loss or, of course, if the defence is otherwise unsuccessful on the merits. Neither the rule of law nor natural justice requires that a person have a second opportunity to challenge a court order in contempt proceedings based upon its breach.
In that restricted class of cases, the rule of law is not being undermined by permitting collateral challenge to the order in the enforcement proceedings. In all other circumstances, the rule of law and the authority of the courts require that individuals subject to court orders obey such orders unless and until they are set aside. Those objectives require that persons bound by court orders challenge them by appeal or application to the court, and not by conduct which is destructive or frustrates the purpose of a court order. Breaching a court order and then raising a collateral attack as a defence in consequent contempt proceedings cannot provide a further opportunity to challenge a court order where a challenge by application to the court or appeal was available.
Contempt of court and the rule against collateral challenge: justified limitations
In his submissions at the resumed hearing, in further response to the new permitted grounds of appeal, counsel for the appellant argued that the appellant should have been permitted to raise freedom of expression, as well as legal errors by the Judge in making the suppression order, as a defence to the charge of contempt by breach of the order. Much of counsel’s argument was based on the importance of the appellant’s right to freedom of expression and the associated principle of open justice. The thrust of this submission is that the law should permit a defence by way of collateral challenge based on freedom of expression or the New Zealand Bill of Rights Act. The rule against collateral challenge, however, subject to the exceptions we have explained, precludes arguments, in the course of contempt proceedings, that an order should not have been made at all, or in the terms that it was. This is so whether the challenge is based on freedom of expression or some other perceived error in the judge’s decision. We see no sound basis for making a general exception to the rule against collateral attack to allow such challenges based on freedom of expression.
Counsel for the appellant also submitted that there must be some consideration of whether the appellant’s conviction for contempt of court for breach of the court order is a reasonable limitation upon the right to freedom of expression. He argued that the law of contempt represents an unreasonable limitation on that freedom. We next consider whether, under the New Zealand Bill of Rights Act, prosecution for contempt of court for breach of a suppression order is a reasonable and justified limit upon freedom of expression, as required by s 5.
We are satisfied that the power of the High Court to hold those who disobey court orders guilty of contempt of court, when exercised in accordance with the principles we have explained, is a justified limitation on the right to freedom of expression. Without such a power there would be no sanction for breach of the order. Contempt of court, its principles being found in the common law, is a limit “prescribed by law”. The reasons why its objective is of sufficient importance to override the rights and freedoms in the New Zealand Bill of Rights Act are those discussed by Elias CJ and McGrath J in their reasons for judgment in Mr Siemer’s previous contempt case set out above at . That case concerned the right to a jury trial in contempt of court proceedings. We consider those reasons to be equally applicable to contempt by breach of a suppression order made under the courts’ inherent powers. In both contexts, contempt of court operates to uphold the authority of the court, which is of fundamental importance to maintaining the rule of law in our society.
Suppression orders in this area are made to give effect to what the court perceives to be a justified limitation of freedom of expression rights in accordance with s 5 of the New Zealand Bill of Rights Act in order to protect the right to a fair trial in the circumstances of the case. Enforcing such orders by bringing contempt of court proceedings against those who disobey them is likewise a reasonable limit on freedom of expression.
As well, subject to the principles we have discussed, the rule against collateral challenge is a justified limitation on freedom of expression. Permitting collateral challenges, whether based on freedom of expression or other alleged errors, would have the effect of allowing third parties to make their own publishing decisions in breach of court orders, according to their own judgment, which would undermine the effective operation of those orders. That would be highly detrimental to fair trial interests. It would also damage the authority of the courts in the exercise of their function for the reasons already identified in particular at  above.
We recognise that suppression orders may at times be made inappropriately, or cease to be appropriate with changing circumstances. For that reason we have explained that individuals may approach the court, seeking variation or rescission of a suppression order that affects their rights or freedoms. We have also indicated that to defer compliance with a court order while such an application is made will not invariably be contemptuous provided that the person bound has not acted in a way that frustrates the purpose of the order. As well, there is a limited exception to the rule against collateral challenge to ensure a meaningful opportunity to challenge suppression orders is available to those not parties to the relevant proceedings. But the fundamental principle is that, unless and until an order of court is modified or discharged by due process, it must be complied with. Fair trial rights and the administration of justice justify that temporary restriction on freedom of expression and those who publish suppressed information in defiance of an order risk prosecution for contempt of court.
Overall, we are satisfied that these principles of the law of contempt of court are a proportionate means of achieving its objective, which is to protect the effective administration of justice. This approach to the law of contempt also involves the minimum degree of intrusion necessary for the courts to uphold the judicial function.
Application of these principles in the present case
We turn to the present case. It was open to the appellant as a member of the public to apply informally to the Court to seek variation or rescission of the suppression order made by Winkelmann J. The appellant could have made the point that he was seeking to inform the public about the decision and facilitate public scrutiny of it. He could have argued that his position was analogous to that of accredited media who have been held to have standing to apply for discharge, rescission or variation of suppression orders. Indeed, it was the appellant’s argument in this case that the publications on his websites were on or akin to others on media outlets. Alternatively, the appellant could have waited to see if others took the initiative, as actually happened, with the result that the Judge reviewed the suppression order and modified it to allow publication of the result of the judgment on 21 December. Even if he thought his prospects of being permitted to publish were slim, the appellant could and should have approached the Court or encouraged or waited for others to do so rather than disobey the order.
Instead, the appellant chose to publish immediately what had been suppressed despite having ample opportunity to reflect on the situation and address what he wanted to do in other ways. Requiring the appellant to seek rescission or variation of the suppression order through proper means would not have resulted in an “irretrievable loss” of his freedom of expression; the present case is analogous to Domm in this regard. There is no basis for applying an exception to the rule against collateral challenge in the present case. It was accordingly not open to the appellant to raise a defence to the charge of contempt based on breach of a court order on the basis that the order should not have been made at all or in the terms that it was.
Nor is there any basis for holding that the appellant’s breach of the court order was not contemptuous. The appellant’s actions frustrated the very purpose of the order made by Winkelmann J which had been made to protect the public interest in a fair trial and the defendants’ right to a fair trial. Permitting a person in his situation to act on his individual view of the inappropriateness of the Judge’s order would be damaging to the protection of fair trial interests and the stability of the law in that respect.
After he had been found guilty of contempt in July 2011, the appellant made an application challenging the order made by Winkelmann J, possibly with regard to his position at the time of sentencing or to an appeal. We do not regard the problems he encountered with an application at that time as of any relevance. The circumstances bear no relation to what they would have been had he brought an application immediately after the order for suppression order was made in December 2010. If the appellant had applied to the Court before breaching the order, and had met with the same response, he may have had the benefit of the exception to the rule against collateral challenge. But the fact of the matter is that the appellant did not make such an application and proceeded to disobey the order by publishing what the Judge had suppressed in order to protect the other person’s fair trial rights. That was in breach of the fundamental rule that, until he had successfully had the prohibition on publication removed, he was bound to obey the order. His defiance of the Court orders by his actions undermined the administration of justice and for that reason was contemptuous.
The appellant knew of the order when he breached it and must have realised he was at risk of being held in contempt. He did not pursue the proper available course of action of approaching, or encouraging others to approach, the Court seeking variation or rescission of the suppression order. Instead the appellant acted impetuously in criticising what had happened, adopting the impermissible course of publishing what was suppressed in defiance of the Court’s order. That course of action was damaging to the effective administration of justice by the courts. It is not the fact that he was critical of the Judge’s decision that makes the appellant’s conduct contemptuous. It is his defiance of the court order.
We conclude that the appellant was correctly held to be in contempt by the courts below. Accordingly we dismiss the appeal.
Mr Siemer must surrender at the Registry of the High Court at Auckland at 9.00 am on Monday 15 July 2013.
 As they were described by Woodhouse J in Taylor v Attorney-General  2 NZLR 675 (CA) at 690.
 See Attorney-General v Leveller Magazine Ltd  AC 440 (HL) at 467–468 per Lord Russell.
 In force at the relevant time but now replaced by ss 196–198, 202, 205 and 207 of the Criminal Procedure Act 2011.
 Siemer v Solicitor-General  NZCA 188,  3 NZLR 43 [Siemer (CA)].
 Solicitor-General v Siemer  3 NZLR 101 [Siemer (HC)] (contempt) and Solicitor-General v Siemer HC Wellington CIV-2010-404-8559, 2 September 2011 (sentencing).
 R v Bailey HC Auckland CRI-2007-085-7842, 9 December 2010 [R v Bailey (9 December 2010)].
 Siemer (HC), above n 5, at .
 Siemer (CA), above n 4, at –.
 A “collateral challenge” is generally regarded as a challenge in proceedings not brought for the purpose of challenge.
 Independent Publishing Co Ltd v Attorney-General of Trinidad and Tobago  UKPC 26,  1 AC 190.
 Taylor, above n 1.
 Siemer (HC), above n 5, at –.
 At –.
 At .
 At .
 At –.
 The powers of a court in cases of abuse of process are discussed in Hunter v Chief Constable of the West Midlands Police  UKHL 13;  AC 529 (HL); and Lai v Chamberlains  NZSC 70,  2 NZLR 7.
 Although avoiding the terms “jurisdictional” and “non-jurisdictional”, the distinction drawn by the majority between power to make an order of the sort made (on which it reserves its position as one it is unnecessary to decide because of the concessions made in the present case) and errors in the exercise of the power (in respect of which they decide that no defensive challenge can be made) is a distinction between jurisdictional error in the narrow sense and other error.
 R v Bailey HC Auckland CRI-2007-085-7842, 7 October 2009; R v Bailey HC Auckland CRI-2007-085-7842, 15 December 2009; and R v Bailey HC Auckland CRI-2007-085-7842, 23 April 2010 [R v Bailey (23 April 2010)].
 R v Bailey (9 December 2010), above n 6.
 R v Bailey (23 April 2010), above n 19.
 Siemer (HC), above n 5, at .
 Siemer (HC), above n 5, at .
 Bright v Attorney-General HC Auckland CIV-2011-404-4294, 26 July 2011 at .
 At .
 Scott v Scott  AC 417 (HL) at 434 per Viscount Haldane, 440–441 per Earl of Halsbury, 445 per Earl Loreburn, and 476–477 per Lord Shaw.
 Broadcasting Corporation of New Zealand v Attorney-General  1 NZLR 120 (CA) at 129 per Cooke J.
 R v Forbes  1 AC 473 (HL) at  and Brown v Scott  1 AC 681 (PC) at 693. See also R v Hansen  NZSC 7,  3 NZLR 1 at ; R v Stewart (Eric)  NZSC 53,  3 NZLR 425 at ; Randall v R  UKPC 19,  1 WLR 2237 at ; and R v Horsham Justices ex parte Farquharson  1 QB 762 (CA).
 See discussion in Condon v R  NZSC 62,  1 NZLR 300 at ; and in R v Hansen, above n 28, at –.
 Article 19(3).
 New Zealand Bill of Rights Act 1990, s 3(a).
 Human Rights Committee General Comment No 34, Article 19: Freedoms of Opinion and Expression CCPR/C/GC/34 (2011) at .
 Taylor, above n 1, at 689.
 Siemer v Solicitor-General  NZSC 54,  3 NZLR 767.
 At .
 Boddington v British Transport Police  2 AC 143 (HL).
 Leveller, above n 2, at 471.
 At 471–472.
 At 465.
 At 468.
 Independent Publishing, above n 10, at .
 Leveller, above n 2, at 473.
 At 453 per Lord Diplock, 456 per Viscount Dilhorne, 465 per Lord Edmund-Davies, and 469 per Lord Russell.
 Shorter Oxford Dictionary (6th ed, Oxford University Press, Oxford, 2007) definition of “adduce”.
 Disclosure of a criminal record otherwise may well amount to contempt at common law in any event.
 Counsel for the Solicitor-General acknowledged that “if someone with standing appealed this suppression order” the Court of Appeal might well have said that the order was more than was required and that it should not have been made: Siemer v Solicitor-General  NZSC Trans 17 at 116.
 Contrast judgment of McGrath, William Young and Glazebrook JJ at –.
 Contrast judgment of McGrath, William Young and Glazebrook JJ at .
 Criminal Justice Act 1985, s 138(6).
 Contrast judgment of McGrath, William Young and Glazebrook JJ at .
 Such as the Bail Act 2000, s 18. The remote participation facilitated by the Courts (Remote Participation) Act 2010 does not affect whether proceedings are held in public or private.
 Contrast judgment of McGrath, William Young and Glazebrook JJ at .
 See, for example, Bulk Gas Users Group v Attorney-General  NZLR 129 (CA); and Peters v Davison  2 NZLR 164 (CA).
 Anisminic Ltd v Foreign Compensation Commission  2 AC 147 (HL).
 Boddington, above n 36, at 154.
 Attorney-General v Times Newspapers Ltd  AC 273 (HL).
 The Sunday Times v United Kingdom  ECHR 1; (1979) 2 EHRR 245 (ECHR) at .
 Boddington, above n 36.
 R v Bedwellty Justices, ex parte Williams  1 AC 225 at 232–233 per Lord Cooke.
 P F Sugrue Ltd v Attorney-General  1 NZLR 207 (CA) at 230; aff’d  UKPC 44,  3 NZLR 464.
 Leveller, above n 2, at 451 per Lord Diplock and 455–456 per Viscount Dilhorne.
 Independent Publishing, above n 10, at  and .
 At . In Leveller, above n 2, the House of Lords held that the actions did not amount to contempt.
 Leveller, above n 2, at 452 per Lord Diplock, 456–457 per Viscount Dilhorne, 465 per Lord Edmund-Davies, 467–468 per Lord Russell, and 471–472 per Lord Scarman; and Independent Publishing, above n 10, at –.
 R v Clement (1821) 4 B & Ald 218, 106 ER 918 (KB); and In Re Clement (1822) 11 Price 68, 147 ER 404 (Exch).
 Independent Publishing, above n 10, at , referring to the criticism of R v Clement, above n 65, in Arthur L Goodhart “Newspapers and Contempt of Court in English Law” (1935) 48 Harv L Rev 885 at 904–905.
 Crimes Act 1961, s 344A.
 A point made in relation to statutory powers by Lord Denning in R v Horsham Justices, ex parte Farquharson, above n 28, at 793.
 It is established that “prescribed by law” includes limits provided for by, or resulting by necessary implication from, statutes or regulations, as well as limits resulting from application of common law rules: Ministry of Transport v Noort  3 NZLR 260 (CA) at 272 per Cooke P, 283 per Richardson J, and 295 per Gault J.
 But see Hogan v Hinch  HCA 4, (2011) 243 CLR 506 at  per French CJ.
 Taylor, above n 1, at 689.
 Slater v R  NZCA 568, referred to in the judgment of McGrath, William Young and Glazebrook JJ at n 236, is concerned not with contempt but with fines imposed in the District Court for breach of ss 139 and 140 of the Criminal Justice Act 1985.
 Siemer v Solicitor-General, above n 34, a case where the injunction breached was in civil proceedings to which Mr Siemer was a party and in respect of which he had rights of appeal. His attempt to challenge the validity of the injunction in contempt proceedings was treated as an abuse of process.
 Boddington, above n 36, at 164 per Lord Slynn.
 At 153–154 per Lord Irvine, and 164 per Lord Browne-Wilkinson.
 Director of Public Prosecutions v Head  1 AC 83 (HL) at 104.
 Boddington, above n 36, at 173–174 per Lord Steyn (emphasis in original).
 See, for example, Mark Elliott “Boddington: Rediscovering the Constitutional Logic of Administrative Law”  JR 144; Ivan Hare “Collateral Challenge Defended or No Smoke Without Fine”  57 Cambridge LJ 429; Paul Craig Administrative Law (7th ed, Sweet & Maxwell, Great Britain, 2012) at [24–003]; and HWR Wade and CF Forsyth Administrative Law (10th ed, Oxford University Press, Oxford, 2009) at 237–239.
 Bugg v Director of Public Prosecutions  QB 473 (DC).
 Boddington, above n 36, at 172–173 per Lord Steyn.
 Broadcasting Corporation, above n 27, at 122.
 Section 210 of the Criminal Procedure Act 2011 gives standing to members of the media to make an application to renew, vary, or revoke a suppression order. This is consistent with a recommendation of the Law Commission in Suppressing Names and Evidence (NZLC R 109, 2009) at [6.55]. The Law Commission advised against extending standing to “other persons with a proper interest” such as witnesses and victims at [6.57].
 Independent Publishing, above n 10, at .
 At , referring to the judgment of the Privy Council delivered by Lord Cooke in Observer Publications Ltd v Matthew  UKPC 11 at .
 See discussion above at .
 Boddington, above n 36, at 161–162 per Lord Irvine, and 173 per Lord Steyn.
 McLachlin J expressed the view that recourse to an appellate tribunal is critical in providing a full and effective remedy for a breach of the Canadian Charter of Rights and Freedoms: Dagenais v Canadian Broadcasting Corporation  3 SCR 835 at 945.
 Certainly, the tendency to procedural exclusivity for judicial review indicated by O’Reilly v Mackman  2 AC 237 (HL) in the United Kingdom has been resisted in New Zealand: P F Sugrue Ltd v Attorney-General (CA), above n 60, at ; and Peters v Davison, above n 53.
 Such as Isaacs v Robertson  1 AC 97 (PC); and Hadkinson v Hadkinson  P 285 (CA).
 Isaacs v Robertson, above n 89, at 103.
 Lord Diplock in Isaacs v Robertson applied the approach he had earlier taken in Re Racal Communications Ltd  1 AC 374 (HL) in holding that a court of unlimited jurisdiction, not conventionally amenable to judicial review, could not be corrected by a superior court for material error of law if there was no right of appeal.
 Siemer (HC), above n 5, at .
 Taylor, above n 1, at 685–686 per Richmond J.
 Attorney-General v Lundin (1982) 75 Cr App R 90 (DC) at 95.
 Armah v Government of Ghana  1 AC 192 (HL).
 Hadkinson v Hadkinson, above n 89.
 Jackson v Sterling Industries Ltd  HCA 23; (1987) 162 CLR 612, referred to by McGrath, William Young and Glazebrook JJ at . The discussion about errors within and without jurisdiction (and the different consequence for nullity) follows from the distinctive Australian approach. See, for example, Craig v State of South Australia  HCA 58; (1995) 184 CLR 163, which was distinguished from the New Zealand position in Peters v Davison, above n 53, at 181.
 At 620–621.
 Wilson v R  SCR 594.
 R v Litchfield  4 SCR 333; Dagenais, above n 87; and Attorney-General of Quebec v Laroche  3 SCR 708.
 Dagenais, above n 87, at 945.
 Attorney-General of Quebec v Laroche, above n 100, at .
 Canadian Union of Public Employees Local 963 v New Brunswick Liquor Corp  2 SCR 227.
 R v Bailey HC Auckland CRI-2007-085-7482, 9 December 2010.
 As the High Court held at : Solicitor-General v Siemer  3 NZLR 101 [Siemer (HC)].
 At .
 At –.
 At  citing Taylor v Attorney-General  2 NZLR 675 (CA) at 687 per Richmond J and at 689 per Woodhouse J.
 At .
 At .
 Taylor, above n 108.
 Independent Publishing Co Ltd v Attorney-General of Trinidad and Tobago  UKPC 26,  1 AC 190.
 Siemer (HC), above n 105, at –.
 At .
 At .
 At .
 At – citing Siemer v Solicitor-General  NZSC 54,  3 NZLR 767 at  per Elias CJ and McGrath J.
 At .
 At .
 At  and .
 At .
 Siemer v Attorney-General  NZCA 188,  3 NZLR 43 at  [Siemer (CA)]. The Court of Appeal distinguished inherent power from inherent jurisdiction at .
 Independent Publishing, above n 112; and Attorney-General v Leveller Magazine Ltd  AC 440 (HL).
 Siemer (CA), above n 122, at .
 At –.
 The Court also noted, at , that s 205 of the Criminal Procedure Act 2011 had retained s 138 of the 1985 Act but without the abrogating provision in s 138(5).
 At [100(a)] (footnote omitted).
 Siemer v Solicitor-General  NZSC 57.
 Siemer v Solicitor-General SC 37/2012, 5 December 2012.
 Taylor, above n 108.
 Broadcasting Corporation of New Zealand v Attorney-General  1 NZLR 120 (CA).
 R v Clement (1821) 4 B & Ald 218, 106 ER 918 (KB). Also relevant is the sequel, In Re Clement (1822) 11 Price 68, 147 ER 404 (Exch). We will refer to both cases as Clement.
 See Siemer (CA), above n 122, at . The term “inherent jurisdiction” was used by Master Jacob in IH Jacob “The Inherent Jurisdiction of the Court” (1970) 23 Current Legal Problems 23 and was widely adopted in New Zealand. Despite our preference for “inherent powers”, in our discussion of the authorities and legislation we have, for accuracy, adopted the language used therein.
 As Rosara Joseph points out in “Inherent Jurisdiction and Inherent Powers in New Zealand” (2005) Canta LR 220 at 220–221.
 Taylor, above n 108, at 681–682 per Richmond J.
 See Department of Social Welfare v Stewart  1 NZLR 697 (HC) at 701.
 Mafart v Television New Zealand  NZSC 33,  3 NZLR 18 at , citing Jacob, above n 133, at 27–28.
 Television New Zealand Ltd v Rogers  NZSC 91,  2 NZLR 277 at  per McGrath J.
 See The Metropolitan Bank Ltd v Pooley (1885) 10 AC 210 (HL); Reichel v Magrath (1889) 14 AC 665 (HL); Reid v New Zealand Trotting Conference  1 NZLR 8 (CA); and, more recently, for example, Siemer v Stiassny  NZCA 1. Rule 15.1(4) of the High Court Rules, which deals with dismissal or stay of proceedings, expressly preserves the inherent jurisdiction of the court.
 See Harley v McDonald  1 NZLR 1 (PC) at ; and Black v Taylor  3 NZLR 403 (CA) at 408–409.
 See Busby v Thorn EMI Video Programmes Ltd  1 NZLR 461 (CA); and Donselaar v Mosen  2 NZLR 191 (CA).
 Clement, above n 132.
 R v Clement, above n 132.
 In Re Clement, above n 132.
 It has sometimes been suggested that the case can be explained on the basis that only a warning was given and that Clement was found guilty of contempt not on the basis of breach of the order but rather because his conduct prejudiced the administration of justice.
 It is set out in In Re Clement, above n 132, at 404–405.
 David Eady and ATH Smith (eds) Arlidge, Eady and Smith on Contempt (4th ed, Sweet and Maxwell, London, 2011) at [7-99].
 Taylor, above n 108.
 At 682.
 At 682, citing Connelly v Director of Public Prosecutions  1 AC 1254 (HL) at 1301.
 At 680.
 Taylor, above n 108, at 680 citing Jacob, above n 133, at 24.
 At 680 per Wild CJ and 684 per Richmond J.
 Albeit expressing his view in terms of “inherent jurisdiction”.
 At 689.
 At 689.
 Broadcasting Corporation, above n 131.
 See  below.
 A possible exception to this is the comment by Woodhouse P at 125 to the effect that s 46 of the Criminal Justice Act 1954 and s 375 of the Crimes Act 1961, when taken together, may have been intended to replace “the necessarily uncertain inherent powers of the Court”. He had expressed a similar view in Taylor, above n 108, see  below.
 Leveller, above n 123.
 See, for instance, R v Julian  1 NZLR 743 (CA); and R v Wilson  1 NZLR 316 (CA).
 See the cases referred to below at .
 To be noted is that in the Criminal Justice Act 1954 provision, the “report” contemplated was in relation to the proceedings, whereas in s 375(2), the “report or account” which could be the subject of an order was of “the evidence adduced”.
 At 693.
 By s 19(3) of the Summary Proceedings Amendment Act 1976.
 The facts of the Broadcasting Corporation case are set out at  above.
 At 126.
 At 128.
 At 134–135.
 Section 375 was recast by s 4(1) of the Crimes Amendment Act (No 2) 1982. This Act originated in the Crimes Amendment Bill (No 2) 1982, which in turn derived from the Official Information Bill 1981. That was a sequel to a report of the Danks Committee on Official Information: Alan Danks Towards Open Government: Supplementary Report of the Committee on Official Information (Department for Courts, 20 July 1981) at 97. This report preceded the judgment of the Court of Appeal in Broadcasting Corporation, above n 131, although several changes made to the Bill post-dated the Broadcasting Corporation decision. The discussion in the report is thus referable to the problems highlighted by Taylor, above n 108. There is no suggestion in the report of any intention to exclude the inherent power of the courts to make non-party suppression orders.
 See Taylor, above n 108.
 The statute permits a court not to state in public “all or any of the facts, reasons, or other considerations” it has taken into account, where either an order has been made under s 197 clearing the court or in relation to a statutory suppression order (see ss 197(3) and 207(2)). As well, s 205(1) replicates s 138(2)(a), as to which see the discussion below at –.
 We think that the phrase “evidence adduced” most readily denotes evidence which is actually given before the court, rather than material that may or may not be later given in evidence at another hearing. The same phrase appears in s 68 of the Summary Proceedings Act 1957 where it undoubtedly has the former meaning. To put it another way, we see the phrase “evidence adduced” as not encompassing “evidence not adduced”, for instance because it is held to be inadmissible.
 We do not accept that a reference in submissions to information confers a general power to suppress publication of that information; contrast the comment of the Chief Justice at . Section 138(2)(c) only provides for suppression of “any report or account of the whole or any part of .... the submissions made”. Assume a defendant has a prior conviction and this is referred to in submissions. Section 138(2)(c) permits suppression of “a report or account” of what was said by way of submission. But it does not provide for suppression of publication of the fact of the prior conviction, which is not in “a report or account” of the submissions.
 As well, we have already explained that the power in s 138(2) applies only to evidence actually adduced at the sitting of the court in question, see n 173 above.
 The phrase “and in no other case” was removed from the Criminal Justice Bill (No 2) 1984 by the Statutes Revision Committee at the same time as subs (5) was inserted. This makes it clear that the latter was intended to express more clearly the intended effect of the former.
 We do not accept the suggestion of the Chief Justice at  that the statutory suppression powers extend to such material. This is because we consider it to be clear that “evidence adduced” means evidence that is in fact admitted and does not include information that is not permitted to be adduced as evidence, see n 173 above.
 Bail hearings can now be heard in private, see Bail Act 2000, s 18. But before the Bail Act there was no explicit legislative provision as to this.
 This list should not be taken as being exhaustive of the circumstances in which a judge might not sit or announce a ruling in public. It may be that the hearing of a matter or giving of a decision in public will sometimes defeat the purpose of making the ruling or decision.
 As well, we have already explained that s 138(6) does not preclude interim suppression, after delivery, of a judgment that has been given in public, see  above.
 As the Chief Justice has noted (at ), middle-banding decisions were made following hearings “on the papers”. But the relevant provisions of the Summary Proceedings Act did not address the mechanisms by which such decisions were given. If the words “shall in every case” in s 138(6) were to be applied literally, such decisions should have been given in public, despite the hearing having been on the papers. What this illustrates is that, at the very least, s 138(6) applied only where s 138(1) was applicable, that is, at a “sitting of any court”.
 It should be noted, as we have already pointed out above at  that not every case where a suppression order is made will necessarily engage the right to a fair trial.
 See Hosking v Runting  1 NZLR 1 (CA), particularly at  per Gault and Blanchard JJ.
 Brooker v Police  NZSC 30,  3 NZLR 91 at – per McGrath J, and  and  per Thomas J.
 Gisborne Herald Co Ltd v Solicitor-General  3 NZLR 563 (CA). This decision applied the New Zealand Bill of Rights Act 1990 in a case involving contempt which did not involve a suppression order. It was an appeal, by a media organisation, against a conviction for sub judice contempt by publication of material prejudicial to a fair trial.
 The position is the same in respect of the temporary intrusion on the right under s 25(a) of the New Zealand Bill of Rights Act to open justice.
 Dagenais v Canadian Broadcasting Corporation  3 SCR 835.
 The Supreme Court of Canada described what we term an “inherent power” as a common law discretion to make publication bans. For earlier authority to the same effect, see R v McArthur (1984) 13 CCC (3d) 152 (ONHC); R v Unnamed Person (1985) 22 CCC (3d) 284 (ONCA) at 286–287; R v Church of Scientology of Toronto (1986) 27 CCC (3d) 193 (ONSC) at 207–208 where Clement, above n 132, was applied; and R v Barrow (1989) 48 CCC (3d) 308 (NSSC) at 315.
 Section 14.
 Sections 25 and 27.
 See above at –.
 The limited occasions on which it was cited and how it was interpreted are reviewed in Independent Publishing, above n 112, at –.
 Leveller, above n 123, at 453 per Lord Diplock, 455 per Viscount Dilhorne and 468 per Lord Russell.
 Independent Publishing, above n 112, at –.
 Leveller, above n 123, at 451–452 per Lord Diplock.
 At 455–456 per Viscount Dilhorne.
 See ss 37 and 41 of the Criminal Procedure and Investigations Act 1996 (UK).
 See  above.
 Independent Publishing, above n 112.
 At – and . As is apparent, we think it clear that there was an order, see  above.
 At .
 See News Digital Media Pty Ltd v Mokbel  VSCA 51, (2010) 30 VR 248.
 John Fairfax Publications Pty Ltd v District Court of NSW  NSWCA 324, (2004) 61 NSWLR 344.
 Hogan v Hinch  HCA 4, (2011) 243 CLR 506 at .
 At  per Gummow, Hayne, Heydon, Crennan, Kiefel and Bell JJ.
 We give reasons below at  for our agreement with the respondent’s submission.
 As well, they do not apply to civil cases.
 Muir v Commissioner of Inland Revenue  NZCA 277; (2004) 17 PRNZ 365 (CA) at .
 Mafart, above n 137, at  per Elias CJ, Blanchard and McGrath JJ.
 See Independent Publishing, above n 112, at . The suggestion was that where a court conducts a “trial within a trial” in the absence of the jury, the purpose of the order excluding the jury would be thwarted by publication, before the end of the trial, of what happened in the absence of the jury. The at least implicit suggestion was that publication in those circumstances would accordingly be a contempt of court.
 This is what happened in Leveller, above n 123.
 By contrast, the Civil Procedure Rules in England and Wales make specific provision for a power to make certain non-disclosure orders in civil proceedings: see r 39.2(4) in relation to the suppression of the identity of any party or witness. An equivalent rule has hitherto been seen as unnecessary in New Zealand.
 We disagree with the Chief Justice on this point, see  above.
 R v Domm (1996) 31 OR (3d) 540 (ONCA) at 546, 552 and 553.
 At 546.
 Taylor, above n 108, at 686.
 Fairfax New Zealand Ltd v C  2 NZLR 368 (CA) at .
 Broadcasting Corporation, above n 131, at 121, 122 per Woodhouse P and at 127 per Cooke J.
 Bain v R  NZSC 59, (2009) 19 PRNZ 524.
 Siemer (HC), above n 105, at ; and Siemer (CA), above n 122, at .
 Taylor, above n 108, at 686. See  below.
 Broadcasting Corporation, above n 131, at 127.
 Under s 27 of the New Zealand Bill of Rights Act 1990.
 Thomson Newspapers Ltd v R (1994) 121 DLR (4th) 42 (ONCA) at 47.
 News Digital Media Pty Ltd v Mokbel, above n 202, at 255.
 Mafart, above n 137, at .
 In particular, ss 198, 210 and 283.
 Sections 198(2)(b) and 210.
 On this issue we accept the submission for the respondent and reject the submission for the appellant, both of which are set out at  above.
 See Leveller, above n 123, at 473–474 per Lord Scarman; and Independent Publishing, above n 112, at .
 Arlidge, Eady and Smith on Contempt, above n 147, at [7.169]–[7.170].
 Taylor, above n 108, at 686. Wild CJ’s general approach in that case is also supportive of the proposition.
 At 690.
 See  above.
 In the United Kingdom and Australia, this issue turns on whether the court making the order is one of unlimited jurisdiction.
 Slater v R  NZCA 568 at . Mr Slater was not a party to the proceedings in which the order he breached was made.
 Siemer v Solicitor-General, above n 117, at  per Elias CJ and McGrath J.
 Woodward v Lincoln (Earl)  EngR 1865; (1674) 3 Swans 627, 36 ER 1000 (Ch) where Lord Nottingham said: “for whatever the mistakes be, the subject must obey below, and dispute here, for a contempt is not to be justified”.
 Russell v East Anglian Railway Co  EngR 870; (1850) 3 Mac & G 104, 42 ER 201 (Ch) per Lord Truro LC.
 Hadkinson v Hadkinson  P 285 (CA).
 With whom Somervell LJ agreed.
 At 288 (footnote omitted).
 Chuck v Cremer  EngR 1155; (1846) 1 Coop T Cott 338 at 342–343,  EngR 1155; 47 ER 884 at 885.
 Isaacs v Robertson  1 AC 97 (PC).
 Discussed at 101 of the Privy Council decision.
 At 101–102.
 Re M  1 AC 377 (HL).
 M v Home Office  1 QB 270 (CA) at 298–299.
 At 299.
 At 300.
 At 423.
 At 423–424.
 Boddington v British Transport Police  2 AC 143 (HL).
 At 161 per Lord Irvine.
 At 152 per Lord Irvine and 164 per Lord Steyn. We observe that the greater accessibility and particular importance of judicial review in New Zealand leaves scope for argument that the Boddington approach to collateral challenges to administrative decisions would not have general application in New Zealand. The approach, to date, of the New Zealand courts to collateral challenge to administrative action is discussed in Dean Knight “Ameliorating the Collateral Damage Caused by Collateral Attack in Administrative Law” (2006) 4 NZJPIL 117. We are not required to decide the question in this appeal.
 Bell v Tuohy  EWCA Civ 423,  1 WLR 2703.
 At .
 Forresters Ketley v Brent  EWCA Civ 324 at .
 Arlidge, Eady and Smith on Contempt, above n 147, at [9–212] and following.
 For the reason given in the passage in the judgment of Elias CJ and McGrath J in Siemer v Solicitor-General, above n 117, set out at  above.
 Compare in a different constitutional context Domm, above n 214, at 550 and 556.
 Wilson v R  2 SCR 594.
 Laskin CJ and Estey J joined McIntyre J’s judgment.
 At 599.
 At 604.
 At 607–609.
 At 614 per Dickson J. Chouinard J joined Dickson J’s judgment.
 Canada (Human Rights Commission) v Taylor  3 SCR 892 at 942 per Dickson CJ, quoting from Canada Metal Co v Canadian Broadcasting Corp (No 2) (1974) 4 OR (2d) 584 (ONHC). See also McLachlin J at 972–975; and R v Litchfield  4 SCR 333 at
347–349 per Iacobucci J.
 R v Litchfield, above n 268, at 349. See Domm, above n 214, at 547.
 Attorney-General of Quebec v Laroche  3 SCR 708 at .
 Domm, above n 214, at 555.
 R v Fields (1986) 56 OR (2d) 204 (ONCA).
 See Domm, above n 214, at 554.
 At 555.
 James v Cowan  HCA 46; (1929) 42 CLR 305.
 At 310–311.
 Jackson v Sterling Industries Ltd  HCA 23; (1987) 162 CLR 612.
 Mason CJ, Wilson, Brennan, Deane and Dawson JJ.
 Russell v East Anglian Railway Co, above n 239.
 Jackson v Sterling Industries Ltd, above n 277, at 620.
 Re Macks, ex parte Saint  HCA 62, (2000) 204 CLR 158.
 See, for example, Merribee Pastoral Industries Pty Ltd v ANZ Banking Group Ltd  HCA 41, (1998) 193 CLR 502 at 516; and DJL v Central Authority  HCA 17, (2000) 201 CLR 226 at 263–264. The rule against collateral challenge is also affirmed in Des Butler and Sharon Rodrick Australian Media Law (4th ed, Lawbook Co, New South Wales, 2012) at [5.420].
 We are not required to decide whether a defence would be available if the court had no power to make an order of the relevant kind, see  above.
 The orders in Taylor, above n 108; R v Slater, above n 236; and Domm, above n 214, were orders made against the whole world. The M cases (above n 247 and 248) involved an order made ex parte.
 The penalties for breach of a statutory suppression order made under s 138 of the Criminal Justice Act were set out in ss 138(7) and (8). The penalties for breach of statutory suppression orders made under the Criminal Procedure Act are set out in s 211 of that Act.
 Siemer v Solicitor-General, above n 117, at –.
 At –.
 At – above.
 Fairfax New Zealand Ltd v C, above n 217, at  approving R v L  3 NZLR 568 (HC) at 569 per Smellie J.
 See  above.
A J Ellis and G K Edgeler (instructed by N Dunning, Wellington) for Appellant.
M F Laracy and A R van Echten (instructed by Crown Law Office, Wellington) for Respondent.
all rights reserved