Legal Beagle by Graeme Edgeler


Parliamentary Privilege, and the summonsing of Crown legal advice

The Epidemic Response has apparently resolved to call on the Director-General of Health, Commissioner of Police and Solicitor-General to provide it copies of relevant material, including legal advice prepared for the government. It is not clear whether summonses have been signed, and served on them, but this appears to be the case and a number of interesting legal questions arise.

This is going to be a long post, so I am going to start with a summary of my view:

  1. The House, and its Committees have the power to call for papers from any person.
  2. The House has the same powers and privileges that the House of Commons had in 1865.
  3. In 2018, the House of Commons exercised a power to require the government of the United Kingdom to provide it legal advice that had been prepared for the Government.
  4. The House of Commons treated the failure of Government to comply with that order as a contempt of the House.
  5. While care should be taken before directing the production of government legal documents to the House of a committee, there is a clear justification for the Epidemic Response Committee to seek the material it has sought, and there a no major concerns around comity with the Courts.
  6. The practice of the government providing the House or its committees with advice from the Law Officers is not unknown in the New Zealand Parliament, but there is an 1875 ruling of Speaker Bell that, if applied, would mean that neither the House, nor its committees could order official opinions of the Law Officer to be produced.
  7. It is not clear whether this does apply, or whether the assertion is correct, especially in light of that the UK direction, and other inroads into government legal professional privilege (including eg the Official Information Act).
  1. If summonses have been served on the Director-General of Health, Commissioner of Police and Solicitor-General, those persons should comply with them, unless or until their effect is stayed, or they are set aside.

This is one of those occasions where the distinction between the House of Representatives and the Parliament of New Zealand matters, so I will endeavour not to stuff it up, but it is useful to begin with the question of where the House gets its powers. In short, it gets them from Parliament. The House acts in a multitude of ways. Parliament acts in one way: enacting legislation.

And, both helpfully, and unhelpfully, the powers of the House are recognised in statute. At least two sections of the Parliamentary Privilege Act 2014 are relevant, although neither wholly answers the question. First is section 8(1):

8 Privileges exercisable and part of laws of New Zealand, journals as evidence

Privileges, etc, exercisable

(1) The privileges, immunities, and powers exercisable by the House, committees, and members are every privilege, immunity, or power that complies with both of the following:

(a) it was on 1 January 1865 (by parliamentary custom or practice and rules, statute, or common law) exercisable by the Commons House of Parliament of Great Britain and Ireland, its committees, or its members; and

(b) it is not inconsistent with, or repugnant to, the New Zealand Constitution Act 1852 of the Parliament of the United Kingdom as in force on (the date of the coming into operation of the Parliamentary Privileges Act 1865, namely) 26 September 1865.

The principal question is therefore: what powers did the House of Commons have on 1 January 1865 that still apply in New Zealand? To answer this, we turn to McGee’s Parliamentary Practice in New Zealand:

The power of the House to inquire into anything that it sees fit has long been held to imply a concomitant power of compulsion to obtain information necessary to carry out an inquiry. These powers—the power to inquire and the power to obtain evidence coercively—have often been regarded as synonymous. However, the House of Commons never claimed a privilege to administer an oath in aid of its inquiry power; and it is not always the case that a body with a power to inquire also possesses a power to order production of evidence. It is preferable, therefore, to regard these as separate powers. The power to secure evidence by requiring the attendance of witnesses and the production of documents is viewed as supporting the power to inquire, and is usually (but not always) co-extensive with it.

The question about whether the House can require the administering of an oath is resolved by the Parliamentary Privilege Act:

24 House or committee may administer oath or affirmation

The House or a committee respectively may, for the purpose of taking evidence on oath or affirmation from a witness examined before the House or that committee, administer an oath or affirmation to the witness.

Importantly McGee notes that:

The New Zealand Bill of Rights Act 1990 is binding on the House and controls any coercive exercise of the power to obtain evidence. Persons against whom the power is exercised have the right to be free from unreasonable search and seizure.

McGee also notes that the power to order witnesses to attend, or for documents to be produced may be delegated by the House to Committee. It has done this in standing orders 196 (which deals with committees with the power to send for person, papers, records) and 197 (which deals with committees without that power). Under the resolution adopted by the House creating the Epidemic Response Committee, the Epidemic Response Committee has the power to send for persons, papers, and records, so I quote only SO 196:

196 Exercise of power to send for persons, papers, and records

(1) A committee with the power to send for persons, papers, and records may order that a summons be issued to any person—

(a) to attend before that committee to be examined and give evidence:

(b) to produce papers and records in that person’s possession, custody or control to that committee.

(2) Every summons issued under this Standing Order—

(a) must state the time and place at which it is to be complied with by the person to whom it is addressed, and

(b) is signed by the Speaker and served upon the person concerned under the Speaker’s direction.

Importantly there is a disjunction between a summons that requires a person to attend to answer evidence, and one that requires production of documents.

This is because of section 25 of the Parliamentary Privilege Act:

25 Privileges and immunities of witness giving evidence on oath or affirmation

(1) A witness examined before the House or a committee and giving evidence on oath or affirmation has, in respect of that evidence, the same privileges and immunities as has a witness giving evidence on oath or affirmation in a court.

(2) This section is subject to any contrary enactment.

A look through the historical antecedents to this section suggests this is mostly directed at the privilege against self-incrimination, but I see no reason why this would not cover oral evidence touching on material that is protected by legal professional privilege. I wondered whether it may be an oversight in the process that adopted the Parliamentary Privilege Act that this extends only to evidence on oath or affirmation, but have concluded it was the intention.

Prior to 2014, the legislated basis for parliamentary privilege was contained in the Legislature Act 1908. It had a more extensive section on witness immunity. A person required under oath to answer questions of either House, or any committee or joint committee, could argue that requiring an answer might incriminate them. If the Committee still felt in needed answers to satisfactorily deal with the issue it was considering role, it could apply to the house for a resolution requiring answers. If such a resolution was issued, answers were then required, but the witness was provided with a certificate of immunity which could be used to stay any civil or criminal case against them for anything they revealed in their answers.

Why would this apply to oral evidence and not documents? Because there’s isn’t a privilege against incrimination in pre-existing documents. A search warrant can be used to obtain them, and a production order can be used to require a person to produce them. That simply isn't what the privilege against self-incrimination means in New Zealand.

The Legislature Act even included a near-identical immunities section to that contained in section 25(1):

(5) Every witness sworn and examined under this or the last preceding section shall have, in respect of the testimony given by him when so sworn, the like privileges, immunities, and indemnities in all respects as are possessed by or belong to any witness sworn and examined in the Supreme Court.

What does this mean? The same thing it means now. Mostly, it is about not being sued, but it too could arguably apply to allow a witness to refuse to answer questions directed at piercing legal professional privilege.

But this doesn’t get us much further toward the question of whether the House (and by extension. Its committees) has the power to require production of government documents protected by legal professional privilege. To advance this further we can look at the experience of both the New Zealand Parliament, and the United Kingdom Parliament.

In respect of the United Kingdom, the position is admirably simple. On 13 November 2018 the House of Commons adopted a resolution requiring “… that the following papers be laid before Parliament: any legal advice in full, including that provided by the Attorney General, on the proposed withdrawal agreement on the terms of the UK’s departure from the European Union including the Northern Ireland backstop and framework for a future relationship between the UK and the European Union.”

The Government refused and on 4 December 2018 the House adopted a motion: “this House finds Ministers in contempt for their failure to comply with the requirements of the motion for return passed on 13 November 2018, to publish the final and full legal advice provided by the Attorney General to the Cabinet concerning the EU Withdrawal Agreement and the framework for the future relationship, and orders its immediate publication.”

The Government published the material the following day.

My research hasn’t been exhaustive, but there appear to have been no substantive law changes between 1865 and now that would have created a power for the House of Commons that it could exercise in 2018 that it did not also possess in 1865. Certainly, there is a long-standing convention that advice from the Law Officers is not disclosed outside Government. Erskine May (the United Kingdom’s equivalent of McGee) records:

By long-standing convention, observed by successive Governments, the fact of, and substance of advice from, the law officers of the Crown is not disclosed outside government. This convention is referred to in paragraph 2.13 of the Ministerial Code. The purpose of this convention is to enable the Government to obtain frank and full legal advice in confidence. Therefore, the opinions of the law officers of the Crown, being confidential, are not usually laid before Parliament, cited in debate or provided in evidence before a select committee, and their production has frequently been refused; but if a Minister deems it expedient that such opinions should be made known for the information of the House, the Speaker has ruled that the rules of the House are in no way involved.

But in light of the 2018 experience, that “usually” only gets you so far, as Erskine May notes, footnoting the events listed above:

Although the opinions of the law officers of the Crown given to Ministers have generally been withheld from Parliament (see para 21.27 ), the failure of the Government to comply with a resolution calling for the production of the Attorney-General's legal advice to the Government has been judged to be a contempt.

The experience in New Zealand is not nearly so recent, but also not as non-existent as the Attorney-General suggested in his recent Facebook address. This debate has in fact come up before, and the arguments were remarkably similar. In 1875, the Public Petitions Committee of the House of Representatives was considering a petition in relation to land grant made in Wanganui for an industrial school. One member noted during the debate “that the Committee had over and over again been supplied with opinions from the Attorney-General”. Another member noted that this was the first such request he had known to be refused.

But importantly, the debate was over a motion in the following form:

Motion made, and question proposed, “That it is desirable that the Public Petitions Committee should be furnished with a copy of the Attorney General's opinion on the case of the Wanganui Industrial School Grant.”—(Mr. T. Kelly.)

This is not a motion like that in the United Kingdom House of Commons in 2018. It is more in the nature of a request, with Hansard recording the following exchange with the Speaker:

Mr. MACANDREW asked if the Committee had exercised due diligence, and used the power with which it was invested, to call for persons, papers, and reports. In this case it appeared to him that the Committee had ample power to call for papers.

Mr. SPEAKER said the Committee had not the power to order papers which the House could not order to be produced. The House could not order official opinions of the Law Officer to be produced, and therefore the Committee could not.

While the House voted in favour of Mr Kelly’s request, and the motion recording the view of the House that it was desirable that the opinion be made available, this ruling from Speaker Bell may resolve the issue: if the House of Representatives (for some reason unlike the House of Commons), lacks the power to direct the official opinion of a Law Officer to be produced, then neither can the Epidemic Response Committee.

Of course, in light of the United Kingdom experience, it is not clear that it accurately reflects the privileges of the House of Commons, at 1865 or since. And given that there are public indications that summons have already been issued, there appear to be some indications that the current Speaker, Rt. Hon. Trevor Mallard, may not be following this precedent, so it is useful to consider other matters.

First, there is the question of waiver. The Attorney-General’s Facebook address was remarkable for a number of reasons, but most pertinently that he may have gone so far in his explanation of the government’s legal advice, that he waived privilege in it. He went well beyond what might normally occur, which might include a statement that the Government had obtained legal advice, and was confident of its position. Rather the address included statements such as “on the advice we have received from Crown Law there is no gap in the legality of the powers of enforcement under levels 3 and 4.” The government’s legal position is expanded upon at some length:

I consider that subsection is apt to cover a direction to all New Zealanders for isolation and quarantine, with exceptions as mentioned and is not limited to use only in relation to specified individuals. The provision is that persons can be isolated or quarantined (with no statutory pre-condition that they may have or carry the disease), so there is no stretch of language; and this is entirely consistent with the purpose of these powers – which is to prevent the outbreak or spread of infectious disease.  I observe that the powers set out in s 70 may be contrasted with the powers in Part 3A of the Health Act, which apply to the conduct of an “individual” considered by a medical officer of health to constitute a public health risk.  In cases under Part 3A, the directions in question are directed to the “individual” concerned.

If your position is that the advice is confidential, it is odd that you would decide to publicly air so much of it. I cannot go so far as to say that this amounts to a waiver, or even partial waiver, but if the question is raised at the Epidemic Response Committee, it is something the Chair would properly consider when ruling on an objection being raised over a summons.

The Attorney-General also argued that there was no need for the Committee to see the advice, and that in investigating the legality of the Government’s response to the epidemic, it was potentially trampling on the role of the Courts. I cannot accept this.

It is the role of parliamentary committees to report to the House of action it might wish to take. If a committee considers that there is need for legislative action, it need not wait for a Court to rule that the Government behaved unlawfully.

Parliament passes more validating legislation than we would hope necessary, and while some of it arises because of Court decisions, much of it does not. Neither of the times Parliament validated the swearing-in of Police officers was its decision to do so preceded by a Court decision, and a submission I made to the Government Administration Committee on the Members of Parliament (Remuneration and Services) Bill saw the committee recommend changes to the legislation to validate several years of payments to mid-term replacement list MPs. Waiting for the Courts to rule on any challenge to the lockdown is one possibility, but it does not impinge upon the role of the Courts for a committee, or Parliament itself to form its own view.

Equally, the legal advice available to the Government when it made the decision not to request Parliament to provide new laws for the level four lockdown is relevant to the question of whether the House should maintain its confidence in the Government. Depending on the advice, the decision may have been reckless, or may show ministers being insufficiently sceptical of public servants’ assertions of power. I realise this is anathema to our political system with its very strong whip to actually imagine the House withdrawing confidence, but one option available to a Committee in the right circumstance might be to recommend that the House withdraw confidence, if for example, the committee determined the government had substantially over-reached.

I’d call this highly speculative, but of course it isn’t. We *know* that even if this was recommended, it wouldn’t happen. But the thought experiment establishes the principle. The Government’s legal advice may be relevant to the House’s decision of a matter, or relevant to a committee’s business. And this is something the House has been clear about since at least 1875.

Finally, the concern that complying with a summons might affect the Government’s position in relation to court cases that are now on foot, and that it should be careful . Of course a committee should be careful, but that decision is not one that Parliament has left wholly to the Attorney-General. The Attorney-General noted that the Official Information Act recognised the importance of legal professional privilege, providing that it is a ground upon which release of information may be refused. He did not note, but we should, that it is a discretionary ground of refusal, where release of information must be weighed against the public interest in release. Parliament has entrusted the Ombudsman – an officer of Parliament – with over-ruling government decisions on release of legally privileged material. It is not ridiculous to think that in affording the House all of the privileges and powers of the House of Commons, it might have done the same.

Of course, the House should be careful, but there will times, as the House of Commons determined in 2018, that it’s interest in conducting its business outweighs the Crown’s interest in keeping relevant material from it. I would go so far as to say that if the law does not permit to House to require the production of advice from the Crown's lawyers, it should be extended to allow it.

The government’s expressed concern is that release of its advice may jeopardise legal proceedings that are now before the Courts. I would first note that the Committee’s original requests pre-date the Court proceedings now on foot, but more importantly, in light of the new Parliamentary Privileges Act, I do not see that it would necessarily follow that acceding to a summons would require a waiver in Court. A litigant may be able to see the advice, but that is not the same as being having a right to adduce it in Court. Much Parliamentary material is protected from being used in Court, even though Parliament broadcasts it on live television. I think there is every chance that even if the Solicitor-General’s advice was provided to the Epidemic Response Committee under a summons, a Court will still find the material entitled to the protection of privilege for the purpose of court proceedings.

What next?

Well, the simple point is that three people have (apparently) been served summonses. Those summons require the production of documents. The summons may be beyond the power of the House, but that is first a question for the Epidemic Response Committee (whose chair might rule, for example, that any claimed privilege was waived), and then the House to decide.

I have seen it suggested that the Government should take the “safe” course, and decline to release the material. That course strikes me as anything but safe. Realistically, with the Privileges Committee split evenly between government members and opposition members, the risk isn’t a contempt finding, but two of those who have been summonsed currently occupy roles as the Solicitor-General and the Police Commissioner. People who hold such positions should not consider themselves free to ignore a summons issued by a body with a clear power of summons.

If you have been served a summons, and especially if you are the Solicitor-General or Police Commissioner, there are basically two proper courses of action available to you:

(1) make your best efforts to comply with the summons;

(2) challenge the validity of the summons, and while that challenge is being determined, apply for, and obtain, a stay of the summons pending the determination of a challenge to it.

I see no reason for this to be different because the summons emanates from a committee of the House of Representatives, instead of a Court, except that the process for applying for that stay is a little more opaque. McGee notes that: “The House has established procedures to be followed if individuals object to the use of the power to send for persons, papers and records.” Although it doesn’t specify what these are, and they are not obvious what processes there may be other than consideration of a possible contempt.

The Attorney-General has requested the Speaker refer the question of whether a summons directing the provision of Crown legal advice is in order the Privileges Committee, but such a reference would not automatically stay the effect of any already issued summons. Standing Order 2 provides that, where standing orders do not provide for a rule around something, the Speaker decides the issue, guided by previous Speakers’ rulings and the established practices of the House. It may be that the Speaker can stay the effect of the summons, pending consideration, or it may be that the Privileges Committee forms the view that any question of that nature is a matter for it, or even for the Committee that issued the summons, or the House.

But, assuming they have been served facially valid summonses (ie signed by the Speaker, etc.) I do not think it proper for the Solicitor-General and Police Commissioner to even technically risk being in contempt. It is not enough for them to do nothing while the Attorney-General seeks a solution, if the summons exists, has been served and has not been stayed, they should comply.


A draft submission on the Electoral (Registration of Sentenced Prisoners) Amendment Bill

There are a few days left to put in a submission on the Electoral (Registration of Sentences Prisoners) Amendment Bill.

The bill would allow prisoners serving sentence of imprisonment under three years to vote, essentially restoring the status quo ante that existed before the members bill advanced by then National MP Paul Quinn was passed by a slim majority

For anyone interested in my views, they're published below. I've been sufficiently organised this time to publish them here a few days before submissions close, so if there are any errors, please let me know. 

The Justice Committee

Electoral (Registration of Sentenced Prisoners) Amendment Bill

Submission of Graeme Edgeler


My name is Graeme Edgeler. I am a Wellington barrister with a strong interest in electoral law. I thank the Committee for the opportunity to make a submission on the Electoral (Registration of Sentenced Prisoners) Amendment Bill (“the bill”). I look forward to the opportunity to speak to my submission.

I support the overall intention of this bill. Indeed, I believe it should go further. I consider that being sentenced to imprisonment should not, by itself, remove the right to vote. I would welcome any further opening up of the right to vote, including, for example, raising the threshold, and encourage the Committee to recommend this to the House.

I recognise that a policy decision has been made, so while I touch on the arguments for this, I also address some other matters of policy and drafting raised by the bill.

My submission addresses the following matters:

  • Whether denial of electoral rights is ever justified as part of criminal punishment.
  • The need to avoid arbitrariness in the application of law, including around the timing of sentences, and the applicability of bail laws.
  • The inconsistency between the treatment of civil prisoners and service prisoners.
  • The proposed changes around the unpublished roll.

Is Denial of Electoral Rights ever justified as part of a criminal punishment?

I accept there are situations where it may be justifiable to temporarily remove an otherwise qualified voter’s right to vote. But I would limit this to circumstances of offending where that offending has a connection to the electoral process.

One example would be that contained in section 100 of the Electoral Act, which provides for names to be included on the corrupt practices list, where a person is convicted of a corrupt practice, or found to have committed a corrupt practice on the trial of a (civil) election petition.

An alternative to a system whereby imprisonment is the determining factor could include one where upon conviction, a judge has the power to order that someone be entered onto something like the corrupt practices list for a time. With appropriate criteria, this could present a justifiable limitation on the right to vote. This would also avoid some of the arbitrariness I will discuss later: under the current law, placement of the corrupt practices list is for three years, ensuring that everyone on it suffers the same penalty (missing one general election and one local election).

A process of this nature could even apply to some people convicted but not sentenced to prison, where the nature of their conviction was such as to undermine the integrity of the electoral system (for example, offending related to an election, but charged as forgery, as we saw with the conviction of Daljit Singh, who was sentenced to community detention).

Avoiding Arbitrariness

People should not lose their rights on an arbitrary basis, and the law should not provide that two identical sets of offending have different consequences based on matters beyond any person’s control.

This can occur under the law as it stands. And while the bill reduced the possibility of arbitrariness, unfortunately it will not eliminate it.

Home detention

One anomaly the bill will fix is the distinction between short term prison sentences and home detention. A sentence of home detention is only available if a convicted person’s offending was so serious that they would have been sentenced to imprisonment: if the sentence is short enough, and the aims of sentencing can still be met by home detention, it can be imposed.

However, for some convicted people, lacking in financial means, or living in precarious or overcrowded accommodation can mean they are effectively ineligible for home detention, and may be sentenced to imprisonment where a more wealthy person would receive home detention.

The three-year rule removes this anomaly, because home detention is only available for sentences of two years or less.

Arbitrariness in Sentencing Dates

Under the Electoral Act, the determinant of whether a convicted person is eligible to vote is whether they are in prison under a sentence of imprisonment. This can mean that a person sentenced to six years imprisonment, and paroled after serving 40% of their sentence may never actually lose the right to vote in a general election, if their sentence is timed right, but a person sentenced to six months in prison, serving only three months, may lose the right to vote if the sentence is imposed shortly before an election.

The determinant of this timing may not even by the timing of the offending. Two near identical offences committed at the same time might have different outcomes depending on where they occur in the country, which could mean the local policing priorities or resourcing result in quicker charging, or waiting times at different courts may result in different trial and sentencing dates.

Changing the rule to require a three-year sentence does reduce the likelihood for inequity, but it does not eliminate it.

Arbitrariness in Application of Bail Laws

One consequence of the law, diminished but not removed by the bill, is the possibility that it may treat more hardened offenders less severely.

Appropriately, both the current law and this amendment permit remand prisoners to vote. This means that a person who serves a substantial period of their sentence in pre-sentence remand spends less time ineligible to vote than a person who does not.

This could even apply with two co-offenders, charged with the same crime, one denied bail throughout the court processes (perhaps because of a more extensive criminal history), the other bailed early in the process. After 18 months awaiting trial, both are convicted and receive the same sentence: 4 years’ imprisonment. Two people convicted of the offending, and receiving the same sentence, but one serves the entire non-parole period of their sentence eligible to vote, the other serves it while ineligible to vote.

Again, while the bill will reduce the incidence of this arbitrariness, it does not eliminate it.

Service Prisoners

These types of arbitrariness are a consequence of the policy decision taken under both the current law, and as evidenced in the bill, however there are other apparently arbitrary distinctions that the law and bill make, which could be addressed without altering the fundamental policy choices involved.

The Armed Forces Discipline Act provides for different sentences than the civilian courts, including a sentence of detention, and a more serious sentence of imprisonment. The effect of the current law, and the bill as drafted is that a service detainee (that is a person convicted at court martial and sentenced to a sentence of detention) retains the right to vote while serving their sentence. This is appropriate.

However, the effect of both the current law and of the bill is that a service prisoner (that is a person convicted at court martial and sentenced to a sentence of imprisonment) is entitled to vote, even after conviction. However if a service prisoner is transferred from a service prison to a civilian prison, they would lose the right to vote (a service detainee transferred to a civilian prison would retain the right to vote).

There are several aspects to this that are arbitrary. While there are service specific offences, courts martial can also be held in respect of ordinary civil offences. A person’s entitlement to vote after being convicted and sentenced to imprisonment should not turn on whether the conviction was in a military court or a civil one. It especially should not turn on the decision of a military authority to move a prisoner from a service prison to a civil prison or back. Under both the current law, and the amendment, it does.

Unpublished Names

Clause 8 of the bill provides for an automatic right for prisoners eligible to vote, while in prison, and all prisoners upon release, to be placed on the unpublished roll if they wish.

This not a right accorded to non-prisoners. A person must ordinarily make an application to the Electoral Commission, and must satisfy the Commission that their being placed on the unpublished roll is necessary.

While I am of the opinion that being sentenced to imprisonment should not result in a person having fewer electoral rights than other citizens, I cannot accept that it should result in the prisoner having more electoral rights.

Prisoners who have reason to think that the inclusion of their names on the published electoral roll would be prejudicial to the personal safety of themselves or others should have to meet the same requirements as any other person.

If there is a particular need, then, at most, the amended section 115(4) could provide for an additional type of document that might be used to satisfy the test (for example, allowing a statutory declaration from a prison employee/case officer/social worker, etc, in the same manner as paragraph (c) currently provides for police officers.

It may be that the Committee thinks that the current requirements are too stringent. If that is so, they should be relaxed for everyone.


I would prefer that this bill went further than it does. While it removes some of the inequities present under the current law, it does not remove all of them:

  • The Committee should consider whether there are other, less restrictive means of achieving the bill’s aim or ensuring that certain categories of prisoner should not be able to vote, while avoiding the anomalies the bill maintains.
  • The Committee should consider whether the bill should treat service prisoners in the same way as civil prisoners, in particular, it should reconsider the distinction the Electoral Act and Corrections Act make based on where a service prisoner is directed to serve a sentence of imprisonment.
  • If there are changes to the requirements for going on the unpublished roll, these should apply equally to prisoners and non-prisoners.

I look forward to appearing before the Committee in support of my submission.

Graeme Edgeler


Waiver, the singular Crown and the conduct of Crown legal business

Much has been written about the importance of discretion in an emergency situation, and the concerns raised by the potential for it to be exercised arbitrarily. Given the quality of the discussion, there seemed little point in adding to it at any length. In particular, I point to the evidence given this morning by Professor John Hopkins of the University of Canterbury to the House of Representatives' Epidemic Response Committee. The concerns raised by him and by others have raised are real and remain, and are a matter for ongoing vigilance by the Epidemic Response Committee, and by the press, and if necessary, by the Courts.

But one highly technical aspect of today’s discussion hasn’t been remarked upon, so I saw some use in writing this post.

This morning, members of the Epidemic Response Committee were questioning Police Commissioner Mike Bush. I did not watch the discussion in full, but media reports indicate the Commissioner advised the Committee Police had obtained legal advice from Crown Law and were relying on that in the exercise of their extraordinary powers under the Civil Defence Emergency Management Act. ACT MP David Seymour at least has request (and perhaps others have as well?) that Police release this advice: “The rule of law requires that rules are clear and publicly accessible. That is not the case at present. Police can't demand respect from the public but must earn it. Police have an opportunity to encourage public respect for the law by being open and transparent and releasing Crown Law advice.

At the committee meeting, Bush didn’t agree to release the information, but indicated Police would seek advice from Crown Law on whether to release it.

While these has been some disappointment at this, it isn’t surprising: as a matter of Government practice, Bush can’t release it.

Paragraphs 4.69 – 4.72 of the Cabinet Manual record the position of the Government on the release of Government legal advice:

Release of legal advice

4.69 As part of the Attorney-General’s constitutional role, the Attorney-General represents the Crown in the courts and provides legal advice to the government. Day-to-day instructions to legal advisers are usually provided by departments, agencies, or other Ministers under the authority of the Attorney-General. Nevertheless, the constitutional responsibility of the Attorney-General remains. The Attorney-General has the right to:

(a) obtain copies of all legal advice provided to the Crown (from whatever source);

(b) determine whether to release that advice; and

(c) instruct all lawyers acting for the Crown.

4.70 When determining whether to release legal advice that has been provided to the government, or to refer to the content of such advice, and waive (or potentially waive) legal privilege, there is a need to:

(a) ensure a coordinated government approach to release;

(b) avoid any adverse impact from a release on current or potential legal proceedings; and

(c) ensure that no single release will create an undesirable precedent.

4.71 Where a Minister or a government department considers that it is necessary to release legal advice or refer to the content of such advice, the matter must first be referred to the Crown Law Office. The Crown Law Office will in turn refer the matter to the Attorney-General’s office for approval.

4.72 Where a request is made under the Official Information Act 1982 or the Privacy Act 1993, the decision on release must be made by the Minister or chief executive who received it. The Attorney-General (through the Crown Law Office) should be consulted about the request.

This reinforces the long-standing position of the Government is that there are not multiple “Crowns”. It is one legal personality, and though it can exercise its powers in different ways, and through different state actors, it is still “the Crown” which is acting. Because of this, its position is that the Crown’s legal advice belongs to the Crown as a whole, and in the case, does not belong to Police.

This conception of the Crown means that neither the Police Commissioner, nor Police generally are the client (“the Crown” is) and thus Police are not in a position to waive legal privilege in it.

This is, of course, only government policy, and there are partial exceptions. As noted in 4.72 above, the Official Information Act and Privacy Act can require the release of legal information (it’s rare, while the exception isn’t absolute, it’s close, but theoretically there will be occasions where the release of legally privileged material is in the public interest, which is a decision that has to be made by the person who holds the information).

And there’s another one. The House of Representatives has a power to summon people and documents, a power which it granted the Epidemic Response Committee. If the Committee wants to see this advice, it can insist.

Of course, it may not need to go quite that far. It strikes me as unlikely that any Crown Law opinion on the exercise of legal discretion under state of emergency has been directly disseminated to frontline Police. The direction given by Police command to frontline officers might be informal, or may have been issued as general instructions issued by the Commissioner under section 28 of the Policing Act, or a Commissioner’s circular. These are effectively “orders” by the Commissioner to police, and (if lawful) must be followed.

The House of Representatives has a duty to ensure the public accessibility of the law (I’ll leave the debate over whether this is moral or legal to another day), and it has given the Epidemic Response Committee the power to fulfil this responsibility at least. It has the power to insist upon being told the guidance underpinning the police exercise of statutory discretion, irrespective of the ordinal crown approach to the Conduct of crown legal business. If it needs to, it should exercise it. 


Pandemic Preparedness and the New Zealand general election

New Zealand is planning to hold a general election on Saturday 19 September 2020. With the COVID-19 pandemic, I’ve had several queries about how New Zealand’s electoral laws would cope with this event. There’s obviously a long way to go until the election, but because I was asked, and because it’s useful to have these things written down accessibly to point to later, I've prepared a brief Q&A.

Although not related to COVID-19, the Electoral Act was updated last week, adding things like election day enrolment, but also updating the laws around interruptions of election day. Spurred on, I think, by the Christchurch earthquakes, the law still mostly covers short-term or localised disruptions. We’re not in lock down yet, and may never be, but what does the law say about disrupted elections?

Can the election be delayed?

Yes. Although the Prime Minister has announced the date of the election, the legal formalities have not been completed, and will not be for some time. It is open to the government to announce a different date for the election. It could do this right up until the 52nd Parliament expires. It is possible (and probably advisable) that this would occur after seeking input from the opposition, but this is not required.

When does Parliament expire? And what’s the last date an election could be held?

Unless it is sooner dissolved, Parliament will expire on 12 October 2020. This is because the term of the New Zealand Parliament is three years. This three years starts from the date set for the “return of the writ” of the preceding election (ie the date when the writ – the formal document signed by the Governor-General ordering the Electoral Commission to hold an election – directs that itself be returned to the Clerk of the House of Representatives declaring who was elected). For the last election, the date set for the return of the writ was 12 October 2017, so three years later is October 2020. If we waited until this last day, the usual process would start then – the writ ordering the election must be signed within seven days (so 19 October 2020) and (under that law change from last week) it must be returned within 60 days (so 18 December at the latest). Working backward from this date, this gives a realistic last day for the election of 5 December 2020, allowing time for the count to take place after overseas votes are allowed in), but this could technically be pushed to 12 December 2020 if the regulations were changed (that’s probably unlikely).

That’s under normal circumstances, are there powers to delay the vote beyond that time if there’s an emergency?

Yes, there are rules (also updated last week) about what to do if there is an “unforeseen or unavoidable disruption” to an election. They’re mostly designed to deal with short-term, or localised disruptions (think earthquake or particularly bad weather), and only once an election has been called, but they could be used during an epidemic. The power enables the Chief Electoral Officer to adjourn voting for up to three days, and then subsequently for recurring 7 day periods. During an adjournment, the Electoral Commission could make arrangement for alternative voting processes, including, for example, extending the electronic voting process overseas voters can use to New Zealanders.

With the uncertain end point, and the fact that while this is going on, there isn’t a Parliament, it is not an ideal process to use when the disruption is known, and seems very much to be a last resort, for a very short period.

Can the Government delay the election beyond this year?

The Government couldn’t. Not by itself. But if it was thought necessary, Parliament could delay the election. It has done this before: despite the standard three-year term, the 19th Parliament (elected in 1914) lasted almost 5 years, after Parliament passed the Parliamentary Elections Postponement Act 1916 because of the First World War. Parliament passed a law extending the term of Parliament to four years in 1934 (extending the term of the then current Parliament as well), possibly as part of the response to the Depression. This was later reversed, but the election was also delayed twice during the Second World War, with a Prolongation of Parliament Act passed in each of 1941 and 1942.

Under our current system, a law delaying the 2020 election beyond early December would require legislation, which would have to be supported by at least 75% of the House. In the current Parliament, this means both National and Labour would have to agree.

We’re not at that stage yet, but what if the need to delay the election didn’t arise until after Parliament has stopped sitting?

If Parliament had adjourned, but not been formally dissolved, it could be recalled to act. If the house had been dissolved, or expired, then one way or another there would be an election. It might be briefly delayed using the Chief Electoral Officer’s adjournment powers, but it would have to happen at some point, in relatively short order.

If the election were so deficient, for example, in turnout, with large groups of voters unable to vote because of an unforeseen or unavoidable disruption, it would still result in the election of MPs, but it is possible that the new Parliament might see itself as a sort of caretaker Parliament. The caretaker convention is about executive government in times where the Government doesn’t have the confidence of Parliament (for example, in the period after the vote, but before the result is counted), but something analogous might apply to a Parliament that couldn’t be sure it really had the support of voters – it could ensure continuity, make necessary changes and approve spending (preferably by consensus), and hold the fort while arrangements were made for a new election. This is a long way off however.

Are there other options beyond delaying the election?

Depending on the cirsumstances, Parliament could recognise the issues inherent in holding an election during, or soon after an epidemic, and change the laws around voting - perhaps allowing people within New Zealand to cast special votes electronically in the same way that voter overseas can. Telephone voting, made available to members of the Deaf community would be another option, as would going to a postal vote. Most of these changes would require amendments to the Electoral Act, some might require a supermajority, but some of the smaller ones might be able to be done through the Electoral Regulations. Thankfully, we have time to consider our options.


On the possibility of laws further regulating hate speech

I am among the most pro-free speech people I know. I don’t doubt there are others more strident than me, but I’m certainly up there. And my support for freedom of expression isn’t limited to opposition to government-imposed restrictions. It extends to a belief that employers should have limited powers over their employees non-employment-related speech (I was firmly of the view that, had Israel Folau’s controversial facebook comments been subject to New Zealand employment and non-discrimination law, they would have been protected, and if it turned out they weren’t protected, the law should be changed so they were), and to a preference for companies that form part of the infrastructure of the Internet to avoid content-based restrictions (I oppose campaigns to get companies to refuse to provide services to unpopular people and groups – should be able to decide what comments it publishes, but the company who leases them server space, and the companies that carry their content to you over the Internet, shouldn’t consider it their function to play a part – if certain types of content should be banned, decisions should be taken by governments with approval or oversight from independent courts, not corporations).

My support for freedom of expression seems to come from a slightly different place than it does for most of people who are prominent in New Zealand in their advocate of free speech: I’m not sure I actually agree that free speech enables the marketplace of ideas, that good ideas will beat out bad ones by force of argument, but I generally oppose laws limiting speech anyway, because of a belief that the imposition of criminal consequences is often more harmful than the harms we might seek to prevent by passing criminal laws.

Not only is convicting someone harmful, but prosecuting someone is harmful. Arresting someone is harmful. Arguing that we need to pass a criminalising some conduct to “send a message” is good way to ensure I will reflexively oppose it. We should pass criminal laws because we consider that the conduct engaged in can be so bad, that when people engage in it we are willing to say to their children: you can’t have your parent around for a while, and your life should be made meaningfully worse in a way which increases your risk to society. Some conduct is that bad, but it’s a damn high test.

Even just prosecuting someone, without a sentence of imprisonment, or even without a conviction can have some of these consequences (arguments around why we shouldn’t impose the costs of prosecution on the victims of it are for another day).

But even starting from this point, I am not yet reflexively opposed to all regulation of hate speech. There’s a pretty good chance that I may oppose a particular proposal – and I don’t know what the government’s review is likely to recommend – but the idea that there could be some law change to recognise the harm caused by hate speech does not meet immediate opposition from me.

Of course, future hate speech laws need not be criminal. Much of the general public understanding of the regulation of hate speech comes from high-ish profile examples from the United Kingdom, where criminal law has been the predominant tool. This is not the only option. Defamation law limits speech, but does not impose criminal penalties. It is easy to imagine a law being enacted where hate speech was subject to civil damages, not fines or imprisonment. In fact, that is already the law in New Zealand. There are at least two “hate speech” provisions in the Human Rights Act: (1) a criminal offence around publishing threatening, abusive, or insulting material, with the intention of exciting hostility or ill-will against, or bringing into contempt or ridicule, any group of persons in New Zealand on the ground of the colour, race, or ethnic or national origins; and also (2) a civil prohibition on publishing such words where they are likely to have that effect (this is an objective test, so the effect need not be intended, which is required for the criminal offence). The criminal offence is punishable by conviction and a fine of up to $7000, or up to 3 months’ imprisonment. The civil prohibition is enforced by someone suing in the Human Rights Review Tribunal, for civil damages. Given that the victim of the publication would be a group, and the person making the claim likely an individual, it’s somewhat odd, but any damages would appear to be paid to the claimant.

This civil process was most recently used by Labour MP Louisa Wall, who sued Fairfax, the Marlborough Express and the Christchurch Press over the publication of two editorial cartoons by Al Nisbet found to be insulting to Māori and objectively offensive. The case, and an appeal from it, ultimately failed on the basis that the publication of the cartoons was not likely to excite hostility toward Māori. But they are useful to showing what laws we currently have, and that a criminal offence is not the only option.

In a sense, these sorts of laws are (or at least can be) similar to defamation, and privacy: different civil proceedings dealing with rather abstract concepts. Defamation laws protect against attacks on character or reputation, privacy laws against invasion of privacy, with these sorts of hate speech laws (there are others) protecting against attacks on dignity. It is easy to imagine new laws which might provide for further protection against embarrassment or harm to feelings.

Of course, the protection the law provides against attacks to character, and breaches of privacy is much greater than it currently provides to hate speech said to undermine dignity, but It is not clear to me why a person’s reputation is so much more worthy of legal protection that their dignity. Of course, one solution I might favour would be to drastically reduce the scope of defamation law.

But I have already said that I am not wholly opposed to all new regulation of hate speech, so the question I ought address is what sort of law might find my favour. It is helpful to consider what sorts of restrictions on freedom of expression there are, and which are justifiable.

When the state proposes to limit a right like freedom of expression, it ought to be able to point to a compelling government interest. Often, this is easy: the interest in prohibiting criminal activity is a compelling reason to make conspiracy to commit a crime illegal, especially where the crime is serious: getting together with others, and forming a common intention (through words alone) to murder someone, or to import methamphetamine, is illegal, even though the offending hasn’t even reached a stage of eg attempted murder.

Other times it is harder: our current defamation laws protect reputation even in the absence of other loss, and even in circumstances where publication has not in fact been shown to have diminished a person’s reputation. I don’t think they get the balance right, but know that others do.

Back on the clearly justified side: people shouldn’t feel threatened, so laws against threats, and against stalking are justified.

This last example is helpful in a discussion around hate speech. A part of the problem of stalking is the sense of unease it creates, and the fear and worry it can induce in people. It can have real limit on a person’s ability to live their life. People who have been stalked, like people who have been in situations of domestic violence, can be affected in all facets of their life. Sometimes it could manifest in a fear of being out in public. Other times, it may require a person to limit their profile, for example, not engaging in political advocacy in a way in which they may like to, for fear of being recognised, and re-victimised.

Certain types of hate speech may have a similar effect. Much of this hate speech is already illegal, including death threats, and other threats of violence or of sexual assault. For some other harmful speech, it is less obviously so. Sometimes it can be prosecuted by general laws, but this can be inconsistent. It does not seem necessarily unreasonable that the law would act to ensure that people should are able to go about their lives.

Examples are easy to imagine. A Muslim mother wishes to take her children to the beach on hot summer day. Her beliefs dictate that she should be modestly dressed in public, but she still wants to swim with her kids, so wears a burkini. At the beach, she’s verbally accosted by someone yelling “Go Back to Islam”, and other derogatory comments indicating she doesn’t belong in New Zealand. Now, maybe this is the type of speech we have to live with in a pluralistic society. But we shouldn’t pretend there is no harm. Her kids have heard it. Maybe they were worried for her safety, in the same way that some who hears a threat may fear for someone’s safety. Maybe they’re now scared to go to the beach, in case that bad man (or someone like him) is there.

This might well already be illegal. In 2013, a man was convicted for offensive language for crossing the street to tell two men “you've got Aids” and “you're a poofter”. But the same law has also been used to convict someone for saying a war commemoration should commemorate the dead on both sides of the conflict and not just our dead, and a related law was used to arrest Tiki Taane for performing N.W.A.s protest song “Fuck the Police”. As much as there may be victims of hateful speech, there is also a risk there will be victims of hate speech laws.

But even if we reject expanding hate speech laws, we should not ignore the fact that speech which causes someone to change their public life – not going to the beach because their kids are scared of being accosted; or driving to the supermarket, instead of walking because someone on the direct route yells out the n-word or the (other) f-word every time they walk past – is harmful.

Of course, we can address a lot of this harm without specific hate speech laws. And that may be the preferable approach: for example, instead of a specific hate speech law, you could instead expand the offence of intimidation to more clearly cover off speech which has the effect of diminishing public utility. Or we could expand civil laws – although that too comes with risk (is the threat of bankruptcy, and the lower standard of proof justified?). And, of course, the risks of criminalising hate speech may still be too great, but some of the factors that might mean I am less likely to oppose a proposal around a law designed to address hate speech where that law targets:

  • individualised speech, and not generalised speech
  • directed speech, and not non-directed speech
  • aggressive speech
  • speech which provably inhibits a person’s ability to be in public spaces, or participate in public life

When we think of hate speech laws, we tend to think of laws which would deal with general insults, like the law that was used by Louisa Wall to challenge Fairfax, but I think there is an important distinction for example, between a hateful post on your own facebook page, and one that someone has posted on a victim’s own facebook page (or been sent as a message). And wherever you draw the line, a street preacher giving a sermon against homosexuality generally, is not causing as much harm, as a person approaching someone and yelling at them that they have caused AIDS.

What would cause me to oppose a new hate speech law? The fear that such a law would be disproportionately used against poor and brown people, like most public order offences are. And that still might be enough. It is a very real concern that anyone proposing a law in this area need to address. I'm not sure I've really seen anyone attempt it yet.