Otago Law Review
In the middle of last year, the New Zealand Parliament passed retroactive legislation to rescue MP (and Government minister) Harry Duynhoven from the effects of s. 55 (1) (c) of the Electoral Act following his re-acquisition of Dutch nationality. Duynhoven had applied earlier in 2003 to reacquire his citizenship of the Netherlands, taking advantage of an opportunity provided by the government of the Netherlands to those whose Dutch citizenship had been cancelled by a law passed in 1984. His application was successful. But he did not realise at the time what the legal effect in New Zealand would be. Like many countries, New Zealand tolerates dual citizenship and apparently there is no objection to citizens of another country sitting in the House of Representatives provided they are qualified as voters. But section 55 (1) (c) of the Electoral Act provides that:
The seat of any member of Parliament shall become vacant— [i]f he ... does or concurs in or adopts any act whereby he .. may become a subject or citizen of any foreign State or Power, or entitled to the rights, privileges, or immunities of a subject or citizen of any foreign State or Power.
It is not a question of Mr. Duynhoven having broken any law or violated any prohibition. There is nothing wrong with acquiring foreign citizenship. The problem is simply that the law attaches a particular consequence to it: if you are a Member of Parliament your seat becomes vacant, and it must be filled by a by-election. Read literally, the Act seems to say that the seat becomes vacant by operation of law; but apparently the literal reading is superseded by a practice whereby Parliament’s Privileges Committee considers the issue of disqualification and notifies their view of the matter to the Speaker of the House, who then gives notice of the vacancy, initiating the process of filling it. In Mr Duynhoven’s case, the Select Committee found (by a majority) that s. 55(1) (c) did apply. But before the Speaker could declare the seat vacant the government introduced the Electoral VacanciesAmendment Bill, suspending the operation of the sub-section that was causing the difficulty, and giving that suspension retroactive effect, so that Mr. Duynhoven’s seat was deemed not to be vacant after all. This bill was passed under urgency on the 6th, 7th and 8th of August, by a small majority in the face of quite furious opposition, and it was given the Royal Assent within a day or so of completing its Third Reading.
Needless to say, the furious opposition that confronted the Bill made much of its retroactive character. And it is this issue of retroactivity that I would like to address in this article. I want to say at once that when I ask in the subtitle of this article: “How dodgy was Duynhoven?”, I am not referring to the conduct of Mr Duynhoven himself. I am using his name as a sort of metonym for the Government’s solution to the problem posed by his application for Dutch citizenship and the impact of s. 55
(1)(c). The Select Committee says Mr. Duynhoven acted openly and honourably; and I propose to make no comment on that. Some say it would have been better had he resigned his seat, once it became clear that the Speaker was hesitating in carrying out his duty to declare the seat vacant. I make no comment on that suggestion either, or on the behaviour of the Speaker of the House. What I am interested in is not how Mr. Duynhoven behaved, but how dodgy the legislative manoeuvres were that were employed to rescue him, and I want to use the incident to reflect more generally on the topic of retroactive legislation. I will come back to some specific aspects of the Duynhoven case at the end of this article.
I say the issue in Duynhoven was retroactivity, though section 5 of the amending Act says that it has “[r]etrospective effect” — retrospective, not retroactive. Section 5 reads:
Retrospective effect of change to vacancy rules made by section 4
For all purposes, the question of whether the seat of a member of Parliament became vacant before the commencement of this Act must be determined as if section 4 of this Act had come into force on 14 August 2002.
Is this just a verbal issue? Not quite. I do not want to get too pedantic about this. But I think we will find a distinction between retroactivity and retrospectivity helpful in what follows.
Retrospective legislation is legislation that attaches some legal consequence now and for the future to an event or transaction that took place in the past. For example, we decide on a Tuesday that a certain tax will be imposed on a transaction that took place on Monday. Now suppose you still have till Friday to pay the tax, so strictly speaking the requirement to pay is prospective. Someone might still complain about not having had the chance to avoid this tax liability, because he did not know when he performed the transaction on Monday what the tax liability would be.
Retroactive legislation is more radical. A retroactive law is one that operates on past events as though it were in force when the past event took place. Had section 55(1)(c) been suspended in June at the time Harry Duynhoven applied to reactivate his Dutch citizenship, his seat would not have become vacant or it would not have been liable to be declared vacant. But section 55(1)(c) was not suspended at that time. Suspension actually took place in August. But the magic of retroactivity is that we are now to proceed as if the suspension were in effect in June.
The particular terminology does not matter. An etymologist might point out that “retrospective” looks to the past — from the Latin spectare = to look at — whereas “retroactive” acts on the past. Unfortunately, however, we have to use “prospective” (with the “spectare” root) as the antonym of both, since
“pro-active” has a different meaning altogether. Some American commentators use the terms “weak” or “secondary” retroactivity and “strong” or “primary” retroactivity to refer to what I am calling retrospectivity and retroactivity.
strongly retroactive weakly retroactive primary retroactivity secondary retroactivity retroactive retrospective
Now, it would be a mistake to suppose that what I am calling retroactivity is necessarily worse than retrospectivity, or that what the commentators call strong or primary retroactivity is worse than weak or secondary retroactivity. They are somewhat different phenomena, and they are objectionable or worrisome to varying degrees, depending on the area of law and the implications of particular enactments. An obvious difference is that some pieces of backward looking legislation confer benefits or remove burdens, i.e. they operate beneficially or remedially; while others operate penally, adding punishments or imposing burdens or disqualifications. But this is true of both retroactive and retrospective laws — so you may have a retrospective law imposing a burden (like the tax law I imagined a moment ago) or a retroactive law conferring a benefit, as in Mr. Duynhoven’s case — and obviously that dimension will affect one’s assessment of the situation as well. Let me illustrate the difference — and what I think is the importance of the difference — between retroactivity and retrospectivity with a couple of examples
— one from New Zealand, the other from Australia. The New Zealand example I have in mind is a case, from two or three years ago, well known to most of you. It is a case that came before the Court of Appeal: The Queen v Pora  NZCA 403;  2 NZLR 37. This involved a statutory provision enacted in 1999which purported to increase the minimum non-parole period of imprisonment that a judge would have to impose (in connection with a mandatory life sentence) when he sentenced someone for a murder — a murder involving home invasion — even when that murder was committed in 1992. And the question was what to do about that provision in the light of other aspects of New Zealand’s criminal justice legislation and human rights legislation which specifically prohibit retrospective increases in punishment.
Most everyone agreed that this was objectionable, but it is objectionable as retrospective not as retroactive. To increase the punishment associated with an offence is to change what we do now in respect of something that happened in the past, which makes it retrospective, but it need not require the fiction that the law then was as we are saying it should be now. The provision complained of in Pora did not purport to affect the legal status of what happened in 1992. It was murder then and it is still regarded as murder now. But it was retrospective in the sense that it required the judge to attach a legal consequence in the year 2000 in respect of an action performed in 1992 which could not have been attached, given the state of the law at the time the act was performed. Aretroactive statute in this area would be one that not only changed the penalty, but penalised behaviour that was not previously criminal — legislating in 2002 to punish actions performed by citizens in 2001, which were lawful, legally unobjectionable at the time they were performed. Fortunately we have no example of this in New Zealand — or none that I know of. But here’s one from Australia. In 2002, the Federal Government proposed legislation — Criminal Code Amendment (Anti-hoax and Other Measures) Bill 2002 — making it an offence to send hoax material through the mail, in order to create an anthrax scare or some other sort of terrorist scare.The Bill proposed to make this retroactive, to October 16, 2001,to take account of a large number of hoaxes that were perpetrated in October and November (following anthrax incidents in the United States). So imagine yourself – a smart-ass student playing a merry prank on your fellow citizens, doing something you know is wrong, but the best legal advice you have is that it is not unlawful. And you certainly would not have done it had you known it was a crime. And then suddenly, retroactively, it turns out to be criminal after all and you are open to a ten year penalty.
This would seem to be an unfair trick to play, to change the status of this action today, from permissible to criminal, in advance of the legal change we are considering. We might not have much sympathy with the prankster — for he was hardly engaged in a legitimate activity; still as the Senate Standing Committee for the Scrutiny of Bills reminded the Australian Attorney General, “[n]ot every
‘illegitimate’ activity is ‘criminal’ activity.”He might have known that he was taking a risk given the rather hysterical atmosphere around the world in the wake of September 11, but still there was no law to warn him off this conduct. This, I think, is exactly the sort of situation William Blackstone had in mind when he talked about ex post facto law —
...after an action is committed, the legislator then for the first time declares it to have been a crime, and inflicts a punishment upon the person who has committed it; here it is impossible that the party could foresee that an action, innocent when it was done, should be afterwards converted to guilt by a subsequent law: he had therefore no cause to abstain from it; and all punishment for not abstaining must of consequence be cruel and unjust.
And I think the Senate Committee in Australia was right to find it disturbing.
(I believe the retroactive provision remained in the version of the Bill that was finally passed on March 21, 2002.)
So that is my distinction. Increasing the penalty is retrospective, because what it does is attach a new consequence for now to an action that society already had the right to punish. Imposing a penalty on an act previously regarded as innocent is retroactive, because it changes the status of the action ex post facto from innocent to criminal. The distinction is a delicate one and one can imagine a number of intermediate cases where it would not be so clear. It probably would not hold up if the effect of the backward-looking enactment were not just to increase the punishment but to change the status of an existing offence, to convert a misdemeanour (say) into a felony, for example, in jurisdictions that recognise and give consequence to that distinction. (Or, in the United States, to change a category of homicide from non-capital to capital.) But it does hold up, I think, when the increase in punishment is modest, or when (as in Pora) there is simply an increase in the mandatory non-parole period rather than the removal of an offence from one category to another.
Again, the distinction probably would not hold up if you thought that penalising an activity was just like taxing it, i.e. if you thought that making an action criminal did not really change its legal status; it just attached a consequence. And no doubt a skeptic — a Legal Realist, for example — could make it dissolve altogether under certain sorts of pressure. The Legal Realist would say:
What’s all this about the status of an action in the eyes of the law? Who cares about that? All that matters is the legal consequence. There is the action and there is the legal consequence: one is the occasion for the other. And we either attach legal consequences to past actions or we don’t. But there are not two ways of doing it — one, changing the status of the action and two, just changing the consequence.
That is what the Legal Realist might say, and he might continue: You can phrase it in terms of status if you like but it adds nothing, and it can be applied as easily to Pora as to the more egregious case we imagined. What the statutory provision in Pora does is change the status of an action from one that has non-mandatory-non-parole status to one that has 13-years-mandatory-non-parole status. Different label, different status, if you like. But all it tells you is what the consequence is. And changing an act to crime from non-crime is essentially no different.
Well I do not buy that. It is not a matter of the momentousness of the change. The difference between a 13 year mandatory non-parole period and the normal tariff for murder may be much larger and more significant for the offender than the punishment for a retroactively established misdemeanour. It is not the momentousness of the change.
Instead, I think that in some cases, what I am calling the status of the act matters, because it affects the way the action figures in a whole array of laws and legal procedures. The status of an action is important in relation to what I am going to call “the systematicity of the law”.
When a type of act is made criminal, the action changes its legal character comprehensively, and it becomes significant in all sorts of interrelated ways. For example, if I know that I have — for whatever reason — performed something that may be regarded as a criminal act, then I need immediately to seek legal advice, to secure a good recollection of what happened, names of witnesses and so on. If others are aware of the situation they may want to lay a complaint; victims or bystanders may want to approach the police. When the Crown gets wind of this, then police and prosecutors also have to go on alert, to try and find out what happened, and go about securing evidence, securing me, and so on. And with all this, various formal procedures and protections come into play, for me and for the others involved. Statutes of limitations begin running, indictments must be laid, and so on. What I am saying is that it is not merely at the moment of sentencing — where a penalty is finally attached to my action as a formal consequence — that the status of the act becomes important. In an ordinary case of the possible or purported violation of an existing legal provision, the status of my act is significant immediately in a whole variety of ways. That is why a retroactive law, imposed later, is so problematic. Much of this complex array of subsidiary legal provisions needs to go into effect more or less immediately upon the performance of the action complained of. But it cannot if, at that moment, there is no provision designating the action complained of as an offence. No one knows that they should seek legal advice at the time or secure evidence, or be careful what they admit to — for the action was not at that time criminal and there was nothing to put them on alert. We sometimes take too simple-minded a view of this. We recite the Latin tag
“Nulla poena sine lege” or Dicey’s principle —
[N]o man is punishable or can be lawfully made to suffer in body or goods except for a distinct breach of law established in the ordinary legal manner before the ordinary Courts of the land.
— as though it were simply a principle about fair warning. Now, certainly the absence of fair warning is certainly one of the main headings of objection to retroactive penal legislation. But it is much more than that. The principle is rather that there is to be no question of punishing anyone except under the auspices of the whole systematic apparatus of criminal law in a mature legal system, with its complex and systematic array of principles, procedures and protections which operate from the first reporting of the offence, perhaps even from the moment of its commission, all the way through to the discharge of sentence. And what the retroactive creation of an offence does is simply short-circuit that systematic element in our criminal law. So that is what I mean when I say that a retroactive statute changes the status of the action, and that this is different from a mere retrospective variation in the penalty.
Now the retroactivity in Duynhoven’s case does not involve criminal law. There is no question of any offence being created or abolished by the Electoral Vacancies Amendment Act. But I will argue later that there are issues of systematicity involved in Duynhoven’s case nevertheless, namely the systematic character of our electoral laws. Very briefly, what I will want to say is that these bodies of law are rather like the rules of a game — like cricket or rugby. They form a system — in Joseph Raz’s sense, they form a system of joint validity,so that there would be little sense practicing just one or a few of these rules except in relation to the whole game, i.e. the whole set of activities that they constitute. The legal status of actions and situations — a person being citizen of a given country, a seat being vacant — make sense in relation to the whole game, to this whole system of law; and when we evaluate retroactive legislation with regard to its effect on such legal status, we need to evaluate it with regard to the integrity, the systematic integrity, of this whole body of law.
So that is my elucidation of the distinction between retroactivity and retrospectivity. Pora involved retrospectivity, whereas the Duynhoven rescue involved retroactivity — and key to my understanding of this difference is this business of changing the operation that one of the rules of the game has on the status of an action in relation to the game as a whole. I am not assuming, by the way, that this is a reason for saying that Duynhoven’s case is necessarily worse than the problem complained of in Pora, just because Duynhoven involves retroactivity rather than retrospectivity (or, in the other terminology I mentioned, because it involves primary or strong retroactivity). The distinction is not the be-all and end-all. There is also the separate distinction I mentioned a moment ago, of penal retroactivity as opposed to curative or beneficial retroactivity. But I will come back to this point about the link between retroactivity and systematicity, for I think there may be something objectionable to retroactivity — even in relatively innocuous cases like Duynhoven — which does not apply to the case of Pora, and which could survive an argument (and I am going to try and make this argument) that in fact Pora would have suffered no particular injustice if the offending section, increasing the mandatary non- parole period, had in fact been applied to his case. Let us go back now to the key question. What exactly is objectionable — what, if anything, is objectionable — about the sort of retroactivity we see in the Electoral Vacancies Amendment Act, and how does it compare with what is objectionable about other more egregious cases of retroactivity and retrospectivity, such as those we have just considered?
I want to consider this as part of a doctrine or ethic of legislation. The question is why should a Government hesitate before solving its political or administrative problems with enactments of this kind? What is wrong with this sort of legislation? Why is projecting our laws or the effect of our laws back in time something that our legislators should be reluctant to do, something they should be scrupulous about not doing? These are questions for legislators (and, of course, also for the citizens to whom they are accountable). They are slightly different from questions about the attitude that judges should take when they are confronted with what appears to be backward-looking legislation. There is a principle of interpretation which says that a statute should be read as prospective if at all possible, and obviously that reflects something about the perceived undesirability of retroactive and retrospective legislation. But some writers posit exceptions to this principle — exceptions for remedial, beneficial, or curative legislation, and a point I want to make is this: I do not think we can necessarily assume that an exception to the principle of interpretation should be read as though it implied an exception to the general principle advising against retroactive legislation. The reason for an exception in favour of curative retroactivity, for example, may have to do with how easy it is to ascertain or convey the intent of the legislature for such a statute; for in the end the principle of interpretation must yield to Parliament’s clearly expressed intention. But this need not imply anything about the relative propriety of curative retroactivity. The judge interpreting a piece of legislation is in a somewhat different position from a legislator deciding whether to vote for a bill, and the interpretive principles he uses operate differently and their exceptions operate differently from the general principle prohibiting legislation of this kind. So from the fact that Duynhoven’s law fell into the remedial category, we are not at all entitled to infer that it was a legitimate exercise of legislative power. In another sense, of course, judges have to deal with retroactive law-making all the time. Judge-made law is almost always retroactive, at least so far as the parties before the court are concerned. Here is a schematic example. In the case of P v Q, the court states the law in terms of a certain rule, call it Rule1, which in the circumstances gives judgment to the defendant. But in a subsequent dispute, R v S, in similar circumstances, the plaintiff is dissatisfied with Rule1 and argues very strongly that it should be overturned and replaced with Rule2 (giving judgment to the plaintiff) and his submission is persuasive, and the common law changes. This happens often. It is the life of the common law; that is how it develops.
But it is surely retroactive so far as the unfortunate defendant S is concerned, for S was relying and might have thought himself entitled to rely on Rule1. But now the court has changed the rule, and in the very nature of adjudication, it applies the new rule, Rule2, retroactively to the dispute before it. This is well known and it has been the basis of many attacks on judge-made law, most prominently the attack by Jeremy Bentham on the common law of England.
It is not just in private law — in contracts or torts — that courts change the law. One prominent example is the decision of the House of Lords in R v. R  2
All E.R. 481 to abandon the old principle that a husband could not be guilty of raping his wife if he forced her to have intercourse against her will, applying its new doctrine of course to Mr. R standing in front of them.
To resist the conclusion that judge-made law is retroactive, you have to resort to the old Blackstonian doctrine that judges never make the law, they only find it — and that even when the law appears to change at the hands of the judges, what is changing is our understanding of the existing law — the judges being, as Blackstone put it, “the depositary of the laws; the living oracles, who must decide in all cases of doubt, and who are bound by an oath to decide according to the law of the land.”(Hence the importance of something like Ronald Dworkin’s
“Right Answer Thesis” to rebut charges of retroactivity.)
I only mention all this because if there is a principle that prohibits retroactivity, presumably it applies to all law-making, not just law-making by the legislature. And I can imagine someone saying — not me, but I can imagine someone saying
— that it ill behooves the courts to go round giving the legislature lectures on retroactivity when everyone knows that a fair proportion of decisions by the judiciary amount to changes in the law, with retroactive effect, at least so far as the losing party is concerned. We can now finally face up to the issue directly and consider what, exactly, is wrong with the various kinds of retroactive or retrospective legislation? What did the great American jurist Lon Fuller mean, for example, when he wrote in his book The Morality of Law (1964) that a retroactive statute is a “monstrosity” and a violation of what he called “the internal morality of law”?
The usual gloss on Fuller’s condemnation of retroactivity has to do with the proposition that it is impossible to use rules to guide people’s conduct if the rules are not prospective. How can a rule guide our conduct if our conduct is today and the rule is made tomorrow? As Fuller put it, “[l]aw has to do with the governance of human conduct by rules. To speak of governing . . . today by rules that will be enacted tomorrow is to talk in blank prose.” A retrospective law is like a secret law or an unintelligible law; one cannot use it as a point of orientation. This is why secrecy and unintelligibility, along with contradictions in the laws, and too frequent changes in the law, and the absence of generality, and so on — are all also condemned by the principles of Fuller’s internal morality of law-making. If there is too much retroactivity, or two many flaws along these other dimensions of the internal morality of law, the legal character of the system of governance begins to dissolve. A collapse in legality, a decline in respect for what we would call the principles of the Rule of Law, must lead ultimately to the determination that what we have here is not really a legal system at all. Certainly it undermines the principle of fidelity to law, i.e. the idea that law is something to which the citizen owes respect and obedience. As Fuller put it:
there can be no rational ground for asserting that a man can have a moral obligation to obey a legal rule that does not exist, or is kept secret from him, or that came into existence only after he had acted, or was unintelligible, or was contradicted by another rule of the same system, or commanded the impossible, or changed every minute.
In some ways though this argument about it being impossible to guide conduct using retroactive laws is too narrow. For one thing, it focuses I think too sharply on penal laws, that is on the prescriptions and prohibitions of criminal law, where arguably the function of the law is the direct guidance of conduct. But the principle of prospectivity has also been thought important in the area of vested rights, and there the issue is not so much guidance as reliance. Aperson in whom current law vests a right is not guided by the law, because the law which vests the right in him may not require him to do anything. It requires others to do or refrain from doing certain things, and it offers the right-holder certain protections which he can count on, in calculating the pursuit of his interests. But it is the protection, the guarantee, that he counts on, rather than any guidance. And his complaint about retroactivity is that the provision guaranteeing him that security has been withdrawn by the retroactively operating amendment.
In Harry Duynhoven’s case, the element of guidance by rules has to do with the position and action of officials rather than the position and action of the person most immediately affected. The effect of the old s. 55 (1) (c) of the Electoral Act is to guide the conduct of those responsible for running the House of Representatives — primarily the Speaker of the House — and also those responsible for organizing elections. It tells those officials what the legal effect of certain action by a sitting M.P. will be, and what they are to do about it. No doubt this instruction to officials — to declare the seat vacant — will be of interest to Mr. Duynhoven, and had he been given proper legal advice in June he would no doubt have been guided in his actions by his desire to avoid a situation in which Parliamentary officials would be obligated to act in this way. But that is not the point of the provision. The point of the provision is to guide the conduct of the officials, in the interest of the integrity of the system of representation, as the original authors of the provision understood it. So section 55(1) (c) purports to guide the Speaker and other officials; and the retroactive amendment purports to reverse that guidance and indicate that they do not have to do what the Act previously told them to do (as a result of Mr. Duynhoven’s action after all). If anyone is going to be dazed and confused by the retroactivity, it is Jonathan Hunt, the Speaker of the House of Representatives. But Mr. Hunt was able to avoid confusion by delaying the action required of him under the old provision in anticipation of its retroactive suspension. Actually, even in the case of penal laws — where there does seem to be some issue of direct guidance for the citizen — there is still the question whether the law should be read primarily as an effort to guide the behaviour of citizens or as an effort to guide the behaviour of officials. Obviously every penal law has a subsidiary norm attached to it, instructing officials what to do in case the primary norm subject does not comply with the norm. In the case of a retroactive criminal law, that subsidiary instruction is almost always prospective: that is, in a statute promulgated today, we tell the courts and the jailers what to do tomorrow about an action that was performed yesterday. (It is just a matter of logic. If the act were retroactive with regard to the behaviour of all its norm-subjects, officials as well as citizens — if it had no prospective element at all, even at the subsidiary level — there would be no practical point enacting it.) And when you think about it, the idea of direct guidance for citizens by the criminal law begins to look rather odd. Is it the function of the prohibition on murder to guide our conduct? Well not if you look at the letter of the law: the homicide sections of the CrimesAct contain no provision corresponding to “Thou shalt not kill”. All they offer is a definition of murder and a specification of the punishment. You will remember that Hans Kelsen made much of this in his argument that all law was addressed primarily to officials.
Whatever you think of that suggestion, you might want to at least concede this — that most of us are not guided by the criminal law in avoiding serious offending like murder and rape. We are guided by morality and our consciences, and law operates mainly as a fall-back position to provide for what we hope is the extraordinary situation in which there is nothing else affecting an individual’s reasoning about murder or rape but the threats which the law has issued. And in that case, it is just the threats that do their work and not any normative guidance.
So what exactly do we mean when we say that a law-change retroactively establishing that something is an offence is objectionable because it cannot guide conduct?
Well I think we mean to focus on those cases where law is attempting to change how people think about things morally, where ex hypothesi there is no pre-existing moral prohibition on the conduct anyway, and the law is trying to act independently dragging morality along with it. Or — and this is the more likely situation — there has been a partial sea-change in the way we think about the morality of some action, and law is now seeking to complete it. In this case, what is objectionable about retroactive law is that, in a situation of moral dissensus, people are not in a position to be guided safely and unequivocally by their own conscience. They have to count on the law to provide the only secure guidance that is possible, and the trouble with retroactive law-making is that it undermines the possibility of this guidance.
Here an observation of Lon Fuller’s is very important, when he says of retroactive legislation, that it “not only cannot itself guide action, but undercuts the integrity of rules prospective in effect, since it puts them under the threat of retrospective change”.In a situation of moral ferment, when we know that there may be a rule-change soon, and where people’s consciences are in disagreement with one another, what is the citizen to do? He can be guided by existing law. But if he knows that Parliament is not opposed in principle to retroactive legislation — if he knows that Parliament has already got a taste for it — then his only safe course is to chill his behaviour with the prospect or the possibility of retroactive change; for with the spectre of retroactivity looming that is the only safe thing to do.
It is interesting that this was the suggestion of the European Commission when it dealt with the issue about the retroactivity of the abolition of the marital rape exception. The majority of the European Commission of Human Rights, in holding that there had not been a violation, appeared to say that in the context of the progressive development of the law, the applicant ought to have anticipated that a court might embark upon the legitimate adaptation of the ingredients of the crime and that this was reasonably foreseeable by the applicant.Blackstone in his famous condemnation of ex post facto laws writes that in the case of the retroactive definition of an offence:
it is impossible that the party could foresee that an action, innocent when it was done, should be afterwards converted to guilt by a subsequent law: he had therefore no cause to abstain from it; and all punishment for not abstaining must of consequence be cruel and unjust.
The European Commission seemed to be saying that in certain cases, it is not impossible, and that the applicant could have reasonably foreseen at the time he used violence to have sex with his wife that this would be regarded in future as having been unlawful. Something similar was said also in the Australian case I mentioned — where the Attorney-General said that citizens had been given fair notice by the Prime Minister’s announcement of his intention to legislate, and the legislation was accordingly back-dated to the moment of that announcement. And I take it that something like this is going on with regard to the usual war crimes or crimes against humanity exceptions to most prohibitions on retroactivity in international human rights instruments.
As for retrospective increases in sentencing, here I have to confess that I am unsure what exactly the objection is supposed to be. In the case I mentioned that came before the New Zealand Court of Appeal
(Pora), and in an earlier case, addressing the same provision but resolved on a slightly different basis (Queen v. Poumako), the Court of Appeal said that the proposition that a person convicted of a crime must not be subjected to a heavier penalty than the law allowed at the time the crime was committed is “a long established basic rule of criminal justice” and that violations of this rule are “inconsistent with a fundamental right declared in both national law and international law”.[23 ]“The strength and importance of the principle enunciated in these provisions cannot be doubted ... The principle is one which is at the forefront of a criminal justice system which is fair and just.”And “the reasons for the principle ... are long established and impregnable”.But there was not much discussion of what these reasons were. This is not a criticism of the court. They had to deal with the matter largely as an issue of positive law, for there was in the very same statute a provision prohibiting the imposition by a court of any sentence (or the making of any order in the nature of a penalty) “that it could not have imposed on or made against the offender at the time of the commission of the offence”,and the hellishly difficult question the court faced was whether this general principle had priority over a later and more specific amendment. That issue did not require them to interrogate the values underlying the anti-retrospectivity principle. Legal scholars have not been much help either. Janet McLean talks about the retrospectivity complained of in Pora as being a “pernicious” form of retrospectivity; but she does not say much about what the perniciousness consists in.
Some of the judges and some of the scholars simply cited the Blackstone passage that I mentioned earlier, about the cruelty and injustice of punishing someone for something that was not an offence at the time.But what needs to be explained is how this applies to the quite separate case of a modest increase in penalty to a serious offence — murder — that is already very well-established. Sir Kenneth Keith made the interesting suggestion in his opinion in Pora that perhaps the reasons for the rule prohibiting retrospective increases in penalty can be traced to “Thomas Hobbes’ brilliant discussion in The Leviathan (1651).”
In Chapter 28 of that work, Hobbes argued as follows:
if a punishment be ... prescribed in the law itself, and after the crime committed there be a greater punishment inflicted, the excess is not punishment, but an act of hostility. For seeing the aim of punishment is not a revenge, but terror; and the terror of a great punishment unknown is taken away by the declaration of a less, the unexpected addition is no part of the punishment.
But notice how this depends on a very specific theory of punishment — deterrence, as opposed to retribution; and indeed, special deterrence as opposed to general deterrence.If our approach to punishment is general deterrence, we find no objection here because the retrospectively increased penalty for the offender may be justifiable inasmuch as it deters others. Or if our approach is in terms of retribution or revenge, then there is no inherent objection to the retrospective increase of penalty, provided that the penalty is what the offence deserves. (That is an important analytic point when you are thinking about retrospectivity: you should assume that the penalty would be justified if it were imposed prospectively, and then ask what exactly is the objection based on retrospectivity alone.) So let us assume that the enhanced penalty in Pora is appropriate in retributive terms. What exactly is wrong with imposing the enhancement retroactively? We know that the felon himself can only be deterred by the threat that has been communicated to him. But it is a big step from saying that to saying that he is entitled — as a matter of justice — to rely on the exact amount of the sentence known to him at the moment he decides to perform the act. Consider the position:
In keeping with the rule of law, the punitive consequences of actions should be known at the outset and fixed once and for all; to vary the consequences later when it is too late to do anything to avoid them would be arbitrary and unfair. ... A person who might drive 5 miles per hour over the speed limit where the fine is
$50 might think twice about doing so if being caught would entail jail time or the loss of one’s driving licence.
I can see this with regard to some mala prohibita. But is this really what we want to say about mala in se — like murder or rape? Are we to view the penalty as some sort of tariff agreed upon, some sort of deal between the legislature and the potential criminal. So we imagine Mr. Pora saying to himself:
Well I was going to kill them, and then I remembered the penalty for murder. But
I thought, “Well, It is worth it. I want to kill them so much It is worth ten years in prison for me. I did that calculation and it was worth it.” So I killed them. And now you tell me I have to serve, not ten years without parole, but thirteen or fourteen. Isn’t that unfair?
Are we inclined to view crime in terms of something like what contract theorists call “efficient breach”? And if we are not, if we do not see punishment as simply a price or a tax that must be negotiated by anyone who wants to perform the forbidden act, then I repeat my question: what exactly is wrong with retrospective legislation imposing what would, from a prospective point of view, be a just or appropriate penalty for the offence? Well of course there is more to say than this. If, in a non-retrospective context, someone is sentenced for a serious offence for a prison term longer than a court has the legal authority to impose then we do say something has gone wrong, even if in theory the higher sentence is appropriate? (Some recent decisions of the Court of Appeal held that the High Court had exceeded the limits of the mandatory non-parole periods they were entitled to impose.) Still, I think in a case like this the complaint is not so much about injustice to the individual, but more about the integrity of the penal system as such. Then there are considerations about the way we determine what a fair penalty would be. From one point of view, the later Parliament encroaches on the sovereignty of its predecessor by legislating retrospectively. No Parliament can control its successors, so far as prospective enactments are concerned, but they ought at least to have control over their own time-frame. Moreover, retrospective legislation of this kind often has a dangerously individualized feel to it; it is not general in the way that we want legislative reasoning to be. Justice Thomas observed that the retrospective provision complained of in Poumako:
is dangerously close to being, if it is not in fact, a bill of attainder. Bills of attainder, or bills of “pains and penalties” where a lesser penalty than death was imposed, have an infamous history in English law. ... The iniquity of such bills lies in the fact that the legislature prescribes by legislative act the punishment for a named or ascertainable individual or group of people. When this is done the law ceases to be of general application and citizens cease to be equal before the law.
And there are considerations about the conduct of a defence once a crime has been committed and indicted. Defendant and his counsel have to figure out how to respond to plea bargaining and so on, and for that they do need a sense of the determinate penalty on offer. So I am probably wrong about Pora. I suspect most people think that the offence against prospectivity in Pora was egregious and that this is the paradigm of what is wrong with backward-looking legislation. Compared to this paradigm the Duynhoven rescue seems trivial.
Certainly such a restrictive view of the prohibition corresponds to what happens in America. In the United States, the general issue of backward-looking legislation is addressed by certain provisions of the Constitution which prohibit what the Framers called “ex post facto” laws. Section 9 (3) of Article I provides that
“[n]o Bill of Attainder or ex post facto Law shall be passed” by Congress and section 10 applies that also to state legislatures. Now the Supreme Court of the United States has consistently held — since Calder v Bull  USSC 10; 3 U.S. 386 (1798) — that these prohibitions apply only to penal law. They do not operate to protect private rights in civil law; this is seen as the domain of the quite separate constitutional prohibition on taking property.
Now this may be changing. There was a recent case, Eastern Enterprises v Apfel, 524 US 428 (1998), in which legislation (the Coal Industry Retiree Health Benefit Act) required certain companies, which had long since left the coal- mining business, to fund health benefits for retired miners and their widows, as though that obligation had arisen out of or had been applied by statute to the employment contracts entered into by these companies when they employed miners many years ago. The case was disposed of under the Takings Clause, but Justice Clarence Thomas said that he, for one, would be open in future to argument that Calder v. Bull should be overruled:
this Court has considered the Ex Post Facto Clause to apply only in the criminal context. I have never been convinced of the soundness of this limitation, which in Calder was principally justified because a contrary interpretation would render the Takings Clause unnecessary. ... In an appropriate case, therefore, I would be willing to reconsider Calder and its progeny to determine whether a retroactive civil law that passes muster under our current Takings Clause jurisprudence is nonetheless unconstitutional under the Ex Post Facto Clause.
But for the time being, the position in America is that only penal statutes are caught under this provision. Without a specific takings clause, we might expand the prohibition slightly. We might say — and there is considerable authority for this — that retrospective and retroactive legislation are objectionable only in the context of penal laws or only when in some other way they affect vested rights. In his submission to the Privileges Committee, Sir Geoffrey Palmer quoted Professor Burrows as saying:
In essence, retrospective legislation is only objectionable if it takes away existing rights or defences, renders unlawful things that were lawful when they were done, or attaches a tax or other liability to something done in the past.
Although, again, we have to be clear about the distinction between treating this as a principle of interpretation and treating it as a principle for legislators. Stated in a positive way, the suggestion we are considering is that legislation which confers benefits ex post facto, or lifts burdens retroactively, or removes prohibitions, or cures disqualifications is not nearly so objectionable as penal or digestive legislation. That was the essence of the submissions on behalf of Harry Duynhoven — that retroactive legislation would not be objectionable in this case because it was curative or beneficial.
I wonder about that. Let us begin with curative legislation. Is retroactive legislation which is curative always benign? Surely not. Lon Fuller cited a chilling example:
Throughout their period of control the Nazis took generous advantage of a device not wholly unknown to American legislatures, the retroactive statute curing past legal irregularities. The most dramatic use of the curative powers of such a statute occurred on July 3, 1934, after the “Roehm purge.” When this intraparty shooting affair was over and more than seventy Nazis had been — one can hardly avoid saying — “rubbed out,” Hitler returned to Berlin and procured from his cabinet a law ratifying and confirming the measures taken between June 30, and July 1,
1934, without mentioning the names of those who were now considered to have been lawfully executed.
The pointSub ism
notsion tos tocomparthe Priv eiletheges LabourCommittgovernmentee on behalf of ofH Newarry D Zealanduynhoven withMP,” theJuly
Nazis, but just to make us pause in our assumption that curative retroactivity is always benign. Sometimes it is, sometimes it is not. Often it is a way of covering up or avoiding the embarrassment of administrative irregularity — pretending it did not happen, and depriving the citizen of the remedies that would otherwise be associated with its occurrence.
What about the proposition that there can be no objection to retroactive legislation that operates beneficially? In paragraph 54 of his submission to Parliament’s privileges committee, Sir Geoffrey Palmer argued in favour of retroactive legislation in the following terms:
Since this legislation would be conferring a benefit on Mr. Duynhoven, probably other MPs and indeed the general public by saving the cost of a by-election, it could be considered beneficial. Beneficial [retroactive] legislation is not constitutionally objectionable.
Is this true? Is there no objection to retroactive legislation when it is beneficial? It sounds plausible — after all, if it is pure benefit, where is the harm? We need to look at benefits, however, in a slightly more sophisticated way.
One thing we have to be sure about is that when the law lifts a burden or confers a benefit, we are not dealing with a competitive situation. If we are dealing with a competitive situation, then the issue of benefit is a red herring, especially if the competition is intense or what game-theorists call “zero-sum”. In a commercial or industrial context, where there is competition for business or market share, conferring a benefit on company A is way of disadvantaging company A’ s competitors. Now this by itself is enough to raise issues about the Duynhoven rescue right away. It is not matter of commercial competition: but it is a matter of electoral competition. Some political parties said they would welcome the opportunity to contest a by-election for Mr. Duynhoven’s seat. The government confidently predicted they would hold the seat if there were a by-election, but that does not mean there would be no competitive issue in its contestation. A second point worth remembering is that the issue in Duynhoven’s case is not just a benefit to Mr. Duynhoven. I guess he gets to keep his salary and his franking privileges. More importantly, though, he retains what a legal theorist would call the Hohfeldian powers associated with the position of MP and indeed Minister. The provision that occasioned the difficulty — s. 55(1)(c) of the Electoral Act — does not just take away benefits. Its most important effect is to disqualify the sitting MP from exercising his powers as a Member of Parliament. The power to vote on legislation and other motions before the House; the power to sustain a government in office with his vote along with sixty or so others; not to mention the very considerable powers of a Minister of the Crown, which Mr. Duynhoven could not exercise at all if he were not an MP. (I assume someone stopped him exercising ministerial powers as soon as his little difficulty became apparent.) This issue of his powers is no small matter. The exercise of an MP’s powers determines who we are governed by, what laws are enacted, and in the case of a Minister, what regulations are enforced and how public money is spent. Moreover, these powers exist in an intensely competitive context. People spend a great deal of time, wealth and energy competing for a Parliamentary seat. They do not do it for the money, but because these legal powers matter to them, and they care how they are exercised. The game is high-stakes, and the outcomes are important. But the integrity of the game matters too, because we value our electoral law not just as a way of getting a government, but because it embodies or purports to embody extraordinarily important principles of democracy and equality. Each person’s vote counts the same; and we are to see to it that votes count and that nothing but votes, organized in a fair competitive structure, determines ultimate political outcomes, like who governs the country. Now, this is where I want to return to the point I made at the beginning of this article about retroactivity and the systematic character of laws. In democracies, we do not just have a mad scramble for political power, with a winner emerging
— a strong man — every so often, in Kremlin fashion, by goodness knows what processes. We have a complex and highly elaborate system to frame and house the high-stakes political competition associated with the struggle for control of Parliament and control of the government and administration of the country. It is a highly elaborate net of rules, doctrines, and precedents, which define a game
— in which all sorts of acts and all sorts of situations have special legal status and myriads of legal consequences. It matters whether a seat is vacant or not: we cannot run the electoral system unless that is governed by rules. It matters whether someone is qualified as a candidate or not, and whether someone is qualified as an elector. That has to be governed by rules. It matters how list seats are allocated. These all have to be governed by rules, and not just a set of rules
— but a system of interlocking rules, where the legal designation of an action or situation has legal consequences for the application of all sorts of rules.
It even matters — as Americans found to their cost in the last Presidential election, in the Florida debacle — what counts as voting for one candidate or another, when the markings on a ballot paper are unclear. And I emphasize that example. We know what it looks like when electoral law begins to unravel. What happened in Florida was ugly and it was distressing to America’s reputation as a democracy. It did not unravel altogether, but the unravelling and the confusion, recriminations and accusations that accompanied it went further than most people would have expected. And a lot of how it began was piecemeal manipulation of electoral law for partisan advantage. The Florida case illustrated also the very tight link between the integrity of the electoral system and the ideal of the Rule of Law. Both of these were damaged in 2000 in Florida and in the Bush v. Gore litigation. On a smaller scale exactly the same mix of issues are at take here: the principle of non-retroactivity is key to the rule of law ideal, and here it is being fairly casually brushed aside in order to undermine the integrity of just a little bit of our electoral law.
The point I was making is that it was quite inappropriate to say, on Harry Duynhoven’s behalf, that the retroactive legislation that was being considered would do nothing but confer a benefit, and that, for this reason, there could be no objection to it. The law-change that was envisaged, the law change that actually took place, changed the rules of the game to confer a benefit on one party in a complex system of high-stakes competition, and that is not a trivial matter. Still, supporters of the Duynhoven legislation would respond that the change was a relatively trivial part of the system, and it corrected something that most people regarded as an anomaly. After all, a person can stand for election and hold a seat even though he holds the citizenship of another country, provided he is in other respects qualified as a New Zealand elector, so why should the event of his acquisition of such citizenship during his tenure of a seat trigger a vacancy. I really do not want to go into the merits of this except to say two things. Firstly, it is not hard to construct a scenario in which the citizenship of a candidate would matter. Suppose there were an ANZUS party composed of people with dual American and New Zealand citizenship who sought a return to our former subservience to that great country. People worry about Americans buying up large tracts of New Zealand land: I am not saying that they should worry, but they do. Are there not similar reasons for worry about dual citizenship of members of Parliament if, as in this case, there were a genuine conflict of interest between New Zealand and the country of second citizenship — and why call it “second”? — of these MPs? Secondly, the view that this provision is trivial and anomalous is not universally shared. When I attended some of the Committee stages of the Electoral Vacancies Amendment Bill, I heard a Maori member — an MP who was tangata whenua
— argue against the change. National Party members argued against it. Maybe they would have supported it, had the boot been on the other foot. But one has to take their opposition at face value.
And that leads me to a final point. The change in Duynhoven may have been minor. But it was pushed through under urgency, with the brute force of the government’s numbers and some of its allies, without all-party support.And that is a very dangerous thing for any constitutional change, let alone a retroactive change in the constitution, let alone a retroactive change in the constitution designed to avoid political embarrassment for the ruling party. I say this was a “constitutional” matter, and section 55 (1)(c) of Electoral Act was a “constitutional” provision. That is a hard case to make in New Zealand where there is no written constitution, and thus normal delineation of legal provisions that are constitutional and those that are not. It is something we have to sort out for ourselves. No official text tells us how to do it. But I think the issue is constitutional in the straightforward sense I mentioned earlier: it is part of the structure of political competition, and so it helps define our system of government. The corresponding provision — slightly different from ours
— is certainly part of the constitution in Australia. Section 44(i), Ch. 1, pt. 4 of the Constitution of Australia provides for the disqualification of a person from being chosen or of sitting as a Senator or House of Representatives member if they are “... under any acknowledgment of allegiance, obedience, or adherence to a foreign power, or is a subject or a citizen or entitled to the rights or privileges of a subject or a citizen of a foreign power.”Now of course, the fact that the Australians regard it as constitutional is not dispositive — and actually there has been a proposal inAustralia to have this provision removed from the constitution and just made part of ordinary statute law. So I do not want to give too much weight to this. But think of the matter along the following lines. In constitutional theory, there
“Section 44 of the Constitution has caused several problems for candidates whose right to stand for Parliament was challenged under this provision, and
is a certain sort of nightmare associated with political systems like that enjoyed by New Zealand. Where there is no entrenched constitution, where there is no power of judicial review, where everything is at the mercy of Parliamentary sovereignty, the nightmare is that a majority will use its power in the House of Representatives to change the electoral law and extend the life of the Parliament, thus rendering itself electorally unaccountable. Now, as some of you know, I hold no brief for the idea of a written constitution, and I have spent the best years of my life, here, and in the United Kingdom, and — thanklessly — in the United States arguing in favour of legislative supremacy and against judicial review of legislation.[38 ]I love and I respect the system of government in New Zealand, though I have lived for some time under a system of a quite different kind. But there is this one nightmare, and in Duynhoven’s case — on a fairly minor matter — we had a taste of it.
The government did not want to hold the election in New Plymouth that the law required. They could not see the point of it. It would be a foregone conclusion, they said. It would be too costly they said. And so they used their numbers to legislate under urgency — because they had to get in before the law actually took effect — without all-party support, to allow their view to prevail
— the view that a now legally required election was unnecessary. A minor case. But consider this. We now have an electoral system which some members of the unicameral legislature are prepared to mess with, using their majoritarian powers. We thought there were informal conventions restraining this sort of thing. Apparently there are not. No doubt the government thought it appropriate; and no doubt they will postpone elections in other cases to whenever they think it is appropriate and in the public interest. And so it begins: the old idea that we vote to determine whose view of the public interest should prevail has been cut back and qualified; now, in the new New Zealand the system for determining whose view of the public interest should prevail will on occasion be subordinated to one party’s view of the public interest. And notice that this argument against what happened in the Duynhoven case has very little to do with retroactivity as such: prospective legislation of the same kind, forced through under urgency, without all-party support would be just as bad. Retroactivity just aggravates the offence and highlights the point that this is interference with a systematic operation of our electoral rules, by one of the players acting to secure an advantage in the electoral game.
You may think all this slightly over-wrought, and no doubt it is. The Duynhoven incident was relatively insignificant. It was not as though the future of the government hung in the balance, and the constitutional impact was probably small. But throughout this article I have been trying to emphasize the point of systematicity. Our law works as a system; and it works to the extent that the integrity of the system can be held together. The principle of prospectivity — that we should make law in a forward-looking way, not retroactively or retrospectively
— along with other Rule-of-Law values is key to that systematic integrity. Yet when retroactive law-making is defended — for example by the Government
— it is always defended as an occasional matter, relating only to the particular issue under discussion. Who could possibly object in this case? But I submit
— and this was Lon Fuller’s point — Ageneral increase in the resort to statutes curative of past ... irregularities represents a deterioration in that form of legal morality without which law itself cannot exist.
The threat of such statutes hangs over the whole legal system, and
robs every law
on the books of some of its significance.
When it does that in any legal context there is a problem. But it is particularly a problem with the integrity of the electoral system, which in a constitutional system like New Zealand’s rests on nothing more resilient than the uncompelled willingness of the major parties to respect it.
[* ]B.A., LL.B., Otago; D.Phil. (Oxon). Maurice & Hilda Friedman Professor of Law, Columbia University, New York. An earlier version of this paper was given as a faculty seminar to the University of Otago Faculty of Law and as a public lecture to the Law School, Victoria University of Wellington in August 2003.
The analysis that follows is taken from J. Paul Salembier, “Understanding
Retroactivity: When the Past Just Ain’t what it Used to Be,” Hong Kong Law Journal,
33 (2003) 99.
For “strong and weak” retroactivity, see Robert G. Natelson, “Statutory Retroactivity: the Founders’ View,” Idaho Law Review 39 (2003) 489. For “primary” and “secondary,” see Salembier, op. cit., fn. 1.
Section 2 (4) of the Criminal Justice Amendment Act (no. 2) 1999.
Criminal Code Amendment (Anti-hoax and Other Measures) Bill 2002, inserting into the Australian Criminal Code, s. 471(10): (1) Aperson is guilty of an offence if:
(a) the person causes an article to be carried by a postal or similar service; and (b) the person does so with the intention of inducing a false belief that: (i) the article consists of, encloses or contains an explosive or a dangerous or harmful substance or thing; or (ii) an explosive, or a dangerous or harmful substance or thing, has been or will be left in any place. Penalty: Imprisonment for 10 years.
The date that was chosen for its retroactive effect was the date of Prime Minister Howard’s announcement that such legislation would be introduced. As the Australian Attorney-General said to the Select Committee that considered the bill —
see http://www.aph.gov.au/Senate/committee/scrutiny/bills02/b02.doc The Government agrees that the retrospective creation of an offence is a serious matter. However, in the case of the new hoax offence there are exceptional circumstances justifying retrospectivity. During October 2001, hoaxes were causing significant concern and disruption. Following the terrorist attacks of
11 September 2001, police investigated over 3000 incidents involving suspicious packages of which over 1000 involved anthrax hoaxes. As a result of these hoaxes, mail centres and offices had to be decontaminated, security measures enhanced and emergency services diverted from other duties. These false alarms cost the community both in terms of unnecessary use of public resources and in terms of increased fear and anxiety.
As stated in the Explanatory Memorandum, it was necessary to ensure that such conduct was adequately deterred in the period before the resumption of Parliament. The Prime Minister’s announcement of 16 October 2001 provided this deterrence. The Prime Minister’s announcement was in very clear terms, and received immediate, widespread publicity. The amendments operate only from the time of that announcement.
It has been accepted that amendments to taxation law may apply retrospectively where the Government has announced, by press release, its intention to introduce a Bill to amend taxation law, and the Bill is introduced within 6 months after the date of the announcement (Senate Resolution of 8 November 1988). The new hoax offence was introduced within 4 months after the date of the Prime Minister’s announcement.
An additional consideration is that there is no circumstance in which the perpetration of a hoax that a dangerous or harmful thing has been sent could be considered a legitimate activity in which a person was entitled to engage pending these amendments. The amendments do not retrospectively abrogate a legitimate right or entitlement. For all these reasons, the retrospective application of these amendments is not considered to contravene fundamental principles of fairness or due process.
Ibid: “[W]hile it is undeniable that perpetrating a hoax cannot be considered a
‘legitimate’ activity, what this bill proposes to do is retrospectively declare it to be ‘criminal’ activity – again, a different, and more serious, issue of principle. Not every ‘illegitimate’ activity is ‘criminal’ activity. Declaring something ‘illegitimate’, and then retrospectively declaring it to be a crime, would seem to establish an unfortunate and undesirable precedent.”
Blackstone, Commentaries on the Laws of England, Vol. I, p. 46, cited by Gault J. at para  of The Queen v. Poumako  NZCA 69;  2 NZLR 695.
For the importance of systematicity as an answer to some Legal Realist skepticism, see Jeremy Waldron, “Transcendental Nonsense and System in the Law,” Columbia
Law Review (Centennial Issue), 100 (2000), 16-53.
A.V. Dicey, Introduction to the Study of the Law of the Constitution, 1885 (Eighth edition of 1915 (Indianapolis: Liberty Classics, 1982), Part II, Ch. 4 — cited by Thomas J. in Poumako  NZCA 69;  2 NZLR 695, para .
Joseph Raz, Practical Reason and Norms, New edition (Oxford: Clarendon Press, 1999), p. 114.
Think of the retroactive effect on the losing parties, respectively, in Donoghue v. Stevenson  UKHL 100;  All ER Rep 1;  AC 562 or Hedley Byrne v. Heller  AC
Bentham, “Truth versus Ashhurst”
(http://www.law.mq.edu.au/Units/law420/LAW203S/Ashhurst.doc): “It is the judges (as we have seen) that make the common law. Do you know how they make it? Just as a man makes laws for his dog. When your dog does anything you want to break him of, you wait till he does it, and then beat him for it. This is the way you make laws for your dog: and this is the way the judges make law for you and me. They won’t tell a man beforehand what it is he should not do—they won’t so much as allow of his being told: they lie by till he has done something which they say he should not have done, and then they hang him for it.” See also Gerald Postema, Bentham and the Common Law Tradition (Oxford: Clarendon Press, 1986), p. 277.
Analytically speaking, this is a good case because it is a rule change most of us approve of — which ought to enable us to focus on the issue of retroactivity by itself. Does retroactivity as such undermine the legitimacy of an otherwise desirable change in the law? Cf., however, the opinion of Ralph Beddard, in “Retrospective Crime,” New Law Journal, 145 (1995) 663: “In many ways it seems unfortunate that the questions appertaining to retrospective criminal law-making by the UK courts should be tested in relation to the sensitive subject matter of marital rape where the anachronisms of the earlier law are well recognized and where strong desire for legal change is indicated by everyone except the accused.”
Blackstone, Commentaries, I, p. 69.
Dworkin, Taking Rights Seriously, Chs. 3-4.
Lon L. Fuller, The Morality of Law Revised Edition (New Haven: Yale University Press, 1969) originally 1964. Fuller’s discussion, along with a few proverbial observations by Bracton, Coke, Hobbes, Blackstone, and Madison, give us more or less all we have in our jurisprudence in the way of a full elaboration of the values underlying the principle of prospectivity — which is itself a pretty scandalous state of affairs.
Fuller, The Morality of Law, p. 39.
Fuller, Morality of Law, p. 39.
Beddard, op. cit. fn. 14 above.
Blackstone, Commentaries, Vol. 1, p. 46.
Article 7 of the European Convention on Human Rights provides that no one shall be held guilty of a penal offence made so retrospectively. Article 7 includes the important proviso that it “... shall not prejudice the trial and punishment of any person for any act or omission which, at the time when it was committed, was criminal according to the general principles of law recognized by civilized nations.” Article 15 of the International Covenant on Civil and Political Rights includes an identical proviso except that the phrase “civilised nations” is replaced by “the community of nations”.
Per Keith J. in Pora (paras 61 and 63)
Henry J. in Poumako at para .
Gault J. in Poumako  NZCA 69;  2 NZLR 695 at para .
Section 4(2) of the Criminal Justice Act 1985. In addition, see section 25(g) of the New Zealand Bill of Rights Act 1990, which provides: “Everyone who is charged with an offence has ... [t]he right, if convicted of an offence in respect of which the penalty has been varied between the commission of the offence and sentencing, to the benefit of the lesser penalty,” and also perhaps the retrospectivity provisions of the Acts Interpretation Act.
Janet McLean, “Legislative Invalidation, Human Rights Protection and s 4 of the New Zealand Bill of Rights Act,” NZ Law Review 421. But cf. James Allan, “Oh That I Were Made Judge in the Land,” 30 (2002) Federal Law Review 561 at pp. 570 ff.
Blackstone, Commentaries, Vol. 1, p. 46: “[W]hen after an action is committed, the legislator then for the first time declares it to have been a crime, and inflicts a punishment upon the person who has committed it; here it is impossible that the party could foresee that an action, innocent when it was done, should be afterwards converted to guilt by a subsequent law: he had therefore no cause to abstain from it; and all punishment for not abstaining must of consequence be cruel and unjust.”
Keith J. in Pora, para : “Some of the features of the fundamental rule may be emphasised. We have already mentioned that it is very widely accepted. It is also long accepted. That appears for instance from the frequent reference in scholarly writing to Thomas Hobbes’ brilliant discussion in The Leviathan (1651) chs 27 and 28. That account and many others, along with the legal texts themselves, also emphasise that the rule applies to criminal penalties as well as to criminal liability. The commentaries in addition tie the rule back to principle, especially to justice and effective deterrence.”
Thomas Hobbes, Leviathan, Revised edition, ed. Richard Tuck (Cambridge: Cambridge University Press, 1996), Ch. 28.
Cf. Gault J. in Poumako para 6: — he talks of “the reasons for the principle in terms of prior direction or deterrence” and the consequent possibility of knowing compliance.
R. Sullivan, Driedger on the Construction of Statutes (Toronto: Butterworths, 3rd edn,
1994), p. 514 (quoted by Salembier, op. cit., at p. 127).
Chase J. in Calder v Bull said the following: “I shall endeavour to show what law is to be considered an ex post facto law, within the words and meaning of the prohibition in the Federal Constitution. The prohibition, ‘that no state shall pass any ex post facto law,’ necessarily requires some explanation; for, naked and without explanation, it is unintelligible, and means nothing. ... [T]he plain and obvious meaning and intention of the prohibition is this; that the Legislatures of the several states, shall not pass laws, after a fact done by a subject, or citizen, which shall have relation to such fact, and shall punish him for having done it. The prohibition considered in this light, is an additional bulwark in favour of the personal security of the subject, to protect his person from punishment by legislative acts, having a retrospective operation. I do not think it was inserted to secure the citizen in his private rights, of either property, or contracts. The prohibitions not to make any thing but gold and silver coin a tender in payment of debts, and not to pass any law impairing the obligation of contracts, were inserted to secure private rights; but the restriction not to pass any ex post facto law, was to secure the person of the subject from injury, or punishment, in consequence of such law. If the prohibition against making ex post facto laws was intended to secure personal rights from being affected, or injured, by such laws, and the prohibition is sufficiently extensive for that object, the other restraints, I have enumerated, were unnecessary, and therefore improper; for both of them are retrospective.
“I will state what laws I consider ex post facto laws, within the words and the intent of the prohibition. 1st. Every law that makes an action, done before the passing of the law, and which was innocent when done, criminal; and punishes such action. 2nd. Every law that aggravates a crime, or makes it greater than it was, when committed. 3rd. Every law that changes the punishment, and inflicts a greater punishment, than the law annexed to the crime, when committed. 4th. Every law that alters the legal rules of evidence, and receives less, or different, testimony, than the law required at the time of the commission of the offence, in order to convict the offender. All these, and similar laws, are manifestly unjust and oppressive. ... The expressions ‘ex post facto laws,’ are technical, they had been in use long before the Revolution, and had acquired an appropriate meaning, by Legislators, Lawyers, and Authors. The celebrated and judicious Sir William Blackstone, in his commentaries, considers an ex post facto law precisely in the same light I have done.”
J.F. Burrows, Statute Law in New Zealand (3rd edition, 2003), p. 403, quoted in “Legal 29, 2003, para. 55.
Cf. Lord Denning in Blyth v Blyth  AC 643, p 666: “The rule that an Act of Parliament is not to be given retrospective effect only applies to statutes which affect vested rights.”
Lon L. Fuller, “Positivism and Fidelity to Law — a Reply to Professor Hart,”
Harvard Law Review, 71 (1958) 630, at 650.
See Gianni Zappala and Stephen Castles, “Citizenship and Immigration in
Australia,” Georgetown Immigration Law Journal, 13 (1999) 273, at 298-300 were generally found to be in breach of section 44 by the High Court. For instance, in 1992 the High Court found by a majority of five to two that two candidates in a House of Representatives by-election were ineligible due to section 44(i) and therefore a recount could not be conducted. (See Sykes v. Cleary (1992) 109 A.L.R.
577) The two candidates found to be ineligible were John Charles Delacretaz and Bill Kardamitsis. (The case had been brought by Mr. Sykes, a candidate in the by- election against Phillip Cleary, arguing that the latter’s election to the seat was void under section 44(iv) of the Constitution which prohibits candidates or members from holding an “office of profit under the Crown.” Mr. Sykes’ petition then also challenged the eligibility of three other candidates under section 44(i) of the Constitution.) [Delacretaz] was born in Switzerland, had migrated to Australia in
1951, and lived in Australia since that time. He became a naturalized Australian in
1960. [Kardamitsis] was born in Greece, migrated toAustralia in 1969, and has lived in Australia since that time. Kardamitsis also became a naturalized Australian in
1975, at which time he renounced all other allegiance. Upon becoming anAustralian citizen, he had surrendered his Greek passport, and his two subsequent visits to Greece were undertaken with his Australian passport.
“All the High Court judges agreed that the relevant test was whether a person had taken “all reasonable steps” to renounce the nationality of the other country. The majority found that although Delacretaz and Kardamitsis were both naturalized Australians, they had not taken “reasonable steps” to renounce their Swiss and Greek citizenship respectively. As one Justice argued:
It is not sufficient ... for a person holding dual citizenship to make a unilateral declaration renouncing foreign citizenship when some further step can reasonably be taken which will be effective under the relevant foreign law to release that person from the duty of allegiance or obedience. So long as that duty remains under the foreign law, its enforcement — perhaps extending to foreign military service — is a threatened impediment to the giving of unqualified allegiance to Australia
... [Delacretaz and Kardamitsis] each failed to take steps reasonably open under the relevant laws of his native country ... to renounce his status as a citizen of that country and to obtain his release from the duties of allegiance and obedience imposed on citizens by the laws of that country
... accordingly, neither ... was capable of being chosen as a Member of the
House of Representatives. Two justices dissented, arguing that both candidates had renounced their previous nationalities when they had sworn their oath of allegiance and renounced any other allegiance during their naturalization ceremonies. They also argued that the test to determine whether “reasonable steps” had been taken to renounce previous citizenships should be with respect to Australian law rather than the law of the other country concerned.”
See Jeremy Waldron, The Dignity of Legislation: The 1996 Seeley Lectures (Cambridge: Cambridge University Press, 1999); Law and Disagreement (Oxford: Clarendon Press, 1999); “Moral Truth and Judicial Review,” American Journal of Jurisprudence, 43
(1998), 75; “Judicial Review and the Conditions of Democracy,” Journal of Political Philosophy, 6 (1998), 335; “ARight-Based Critique of Constitutional Rights,” Oxford Journal of Legal Studies, 13 (1993), 18.
Fuller, op. cit., fn. 36 above, p. 651.