Human Rights and International Law (22): The Usefulness of Retroactive Laws

George Steiner, in his magnificent novel about the fate of an escaped Adolf Hitler in post-war South America, describes rights as an ontological totality, or, in other words, a reality that encompasses the whole world and all phenomena in it. Rights are not just a set of local commands, only valid in some corner of the world or at some moment in history, or only applicable to a certain class of people. It would be unacceptable to have one part of reality – either a part of the world or a part of time – that is free from the moral power of rights.

The Israeli secret agents who, in Steiner’s book, captured the old Hitler in the forests of South America rightly believed that he could be tried by their makeshift jungle court and that their present-day norms dealing with genocide and persecution could be used to judge him, even though these norms did not exist at the time when Hitler committed his crimes or in the places where he committed them.

Human rights are therefore an exception to the otherwise very sensible rule that laws should not be applied retroactively or “ex post facto”. Certain actions that take place in a certain country at a certain moment in time, and that are not illegal in the context of the law as it is valid in the country and at the time, may afterwards – after the facts – be judged as violations of human rights, even if human rights were not part of positive law at the time. Otherwise, a tyrannical legislator such as Hitler may make it forever impossible to judge his deeds of oppression and to punish him, even after he and his regime are defeated. We shouldn’t be willing to accept an absolute definition of the prohibition on retroactive laws that leads to impunity. If violations of human rights can only be punished according to the laws that are in force in the country in which the violations occur, and that are in force at the moment that they occur, then a bit of creative legislation will lead to total freedom of action for the most brutal dictators. And no change of regime or military defeat will ever harm them. They will have a life-long insurance against justice.

However, the prohibition on retroactive laws is an important achievement, and is even part of the internationally accepted corpus of human rights. See for example article 15 of the International Covenant on Civil and Political Rights:

No one shall be held guilty of any criminal offence on account of any act or omission which did not constitute a criminal offence, under national or international law, at the time when it was committed. Nor shall a heavier penalty be imposed than the one that was applicable at the time when the criminal offence was committed.

The fact that human rights laws are an exception to the general rule is justifiable on two grounds:

  • First, it would be unreasonable to require that the system of human rights contain the seeds of its own destruction. Requiring the system to incorporate an absolute prohibition on retroactive laws, one that also makes retroactive human rights laws impossible, would mean introducing one rule in the system that can undo all other human rights rules. The absolute rule on retroactive laws would allow malevolent legislators to neutralize all other human rights rules in the system.
  • Second, the main rationale for the rule on retroactive laws is the fact that we must be able to know in advance whether our actions are or are not permitted by the law. Otherwise we are unable to plan our actions in a manner that is appropriate for law-abiding citizens and we run the risk of inadvertently violating the law. A law that is unknown to citizens is plainly absurd, and retroactive laws are by definition unknown before they are introduced. If laws are enacted that punish actions after they have occurred, then we risk being harmed by the unknown penal consequences of our actions, actions that we believe are legal. However, this rationale is absent in the case of human rights. Even if human rights are not part of existing law, most people will not inadvertently violate human rights. Human rights, even if they are not part of the law, are known to most citizens, and those who violate them know that even if their law allows them to do so, they can one day be held to account. Punishing someone on the basis of human rights that were not part of the law when the punishable act was committed, is clearly not the same thing as punishing someone for driving in a pedestrian zone when this zone was accessible for cars at the moment of the “infraction”.

Rights violations must always be punishable, even if the law that makes them punishable only comes into force after the violations have occurred, for example after the overthrow of a dictatorship or after the military defeat of the violators. All other acts that do not imply a violation of human rights, can only be punishable if they are a crime according to the law at the moment that these acts are committed. Generally, one cannot punish someone for an act that is not a crime and only becomes a crime afterwards, because otherwise this person is unable to know whether his act is legal or illegal and is unable to plan his actions in a way that fits a law-abiding citizen.

The exception to the general rule – this general rule being itself a particular case of the even more general rule of “nullum crimen sine lege“, no crime without a law – was introduced by the Nuremberg Tribunal.

[C]rimes against humanity were made punishable even if perpetrated in accordance with domestic laws … In so doing, it [the tribunal] indubitably applied ex post facto law; in other words it applied international law retroactively”. A. Cassese, International Law in a Divided World, p. 291-292.

[C]rimes against humanity … were defined in the Tribunal’s Charter as follows: “murder, extermination, enslavement, deportation and other inhumane acts committed against any civilian population before or during the war, or persecutions on political, racial or religious grounds in execution of or in connection with any crime within the jurisdiction of the Tribunal, whether or not in violation of the domestic law of the country where perpetrated”. In some respects, crimes against humanity are wider than war crimes; they can be committed before a war as well as during a war, and they can be directed against “any civilian population”, including the wrongdoing state’s own population. The prohibition of “crimes against humanity” thus constituted an exception to the old rule that a state was entitled to treat its nationals as it pleased; and it is fairly clear that this prohibition was not accepted as part of international law before 1945″. M. Akehurst, A Modern Introduction to International Law, p. 278-279.

Why do I discuss all this? The obvious temporal aspect of retroactive laws can have spatial consequences. If we are not bound by the rule on retroactive laws where human rights are concerned – which is, I admit, a somewhat stretched interpretation of the Nuremberg exception that only mentions some types of rights violation – then we can do as if human rights are law and we can always punish rights violations, even if these violations were not illegal at the moment that they were committed. We can apply this principle in a spatial dimension rather than a temporal dimension. If we can act as if human rights were law in a previous period of history in which they evidently were not law, then we can also act as if they are law in another place in the world, a place in which they are not yet part of the local law.

This way we can elude the perpetual discussions about the application of international law, about the force of human rights treaties in domestic law, about the role of reservations, about the priority or superiority of international law, about the extent to which some or all human rights are part of international customary law, about whether human rights are part of ius cogens etc. For me, this is no fundamental modification of the Nuremberg-principle, but it has far-reaching consequences. And it is not as farfetched as it may seem at first glance. Most people will agree that it would be wrong to judge someone solely by the laws of his country. Laws, after all, can be incomplete or even immoral.

What Are Human Rights? (22): Part of the Rule of Law

The claim here is not the trivial one that human rights depend on the rule of law because they can’t be enforced without it. The more interesting question is the opposite one: whether there can be a rule of law without human rights. Or, in other words, is the rule of law a necessary but not a sufficient condition for human rights?

At first sight, the answer to both questions would be “yes”. Indeed, the law can be anything, and as long as it “rules” in some way – i.e. as long as the laws are consistently enforced and not superseded by frivolous and arbitrary commands of men – one could claim that there is some sort of “rule of law”, even if the laws in question violate human rights. Civilizations had the rule of law long before the concept of human rights even existed (the Roman Empire may be an example).

Joseph Raz has famously claimed that

the law may, for example, institute slavery without violating the rule of law. (source)

Nazi Germany was also very much a law based society. (See here for example). Indeed, it can be plausibly claimed that strong and authoritarian states are better able to impose rules. That would lead to an incompatibility between human rights and the rule of law.

The fact that many if not most dictatorships make a mockery of the rule of law and of the law itself, and govern in a totally arbitrary way based on the whims of a few men rather than laws and rules, doesn’t exclude the possibility that some dictatorships respect the rule of law, and that the rule of law can indeed be the rule of very bad law, viewed from the perspective of human rights. A prima facie conclusion has to be that dictatorships can respect the rule of law and that regimes based on human rights can inhibit the rule of law: privacy protection, rules on the determination of criminal guilt etc. can make the rule of law more difficult. Authoritarian regimes can easily lift the veil of privacy in order to check for violations of the law, and are not at risk of freeing guilty people because of the presumption of innocence and the burden of proof.

The rule of law, viewed in this manner, is a purely formal concept devoid of substance: as long as the laws “rule”, we have a rule of law, no matter what the substance of those laws may be. Laws are then viewed solely as rules that guide conduct, but the direction in which they guide is immaterial. The rule of law, according to this view, should not be confused with the rule of the right law. The rule of law as a concept deals not with the content of the laws but with the way in which they are enforced and formulated.

That last word is important: the rule of law should logically be more than a system of governance in which rules are imposed by force. Imposing rules by way of force can in itself not be viewed as a system of the rule of law. It would be far-fetched to claim, for example, that a government using force to impose completely arbitrary rules that change every day respects the rule of law. The rules in question have to be formulated in a certain way; there have to be rules of legislation in order to have a rule of law.

These rules usually include the following:

  • Laws should not be imposed retroactively: the rule of law implies respect for the laws, and citizens can’t be expected to respect laws if they are imposed retroactively.
  • Laws should be made public, for the same reason.
  • Laws should be relatively permanent, clear and intelligible, again for the same reason.
  • Laws should strive to be general rules applicable to everyone, rather than commands directed at certain persons or groups; the reason for this rule of legislation is the differentiation between rule of law and rule of man.
  • Laws should not contradict each other, again for reasons of respect.

These rules of legislation differentiate laws and the rule of law from an arbitrary set of rules imposed by force. The rules of legislation are formal and don’t, at first, impose content on the specific laws generated by these rules. However, once you take a closer look at these rules of legislation, it becomes increasingly difficult to maintain that the rule of law is a contentless concept that allows the law to be virtually anything, even abject oppression. Some of the values inherent in the rules of legislation are also inherent in human rights: publicity and equality for example.

The rules of legislation also create another link to human rights: they assume free will. If rules can’t be secret or can’t be applied retroactively it’s because we want to give people the choice to change their behavior so that it complies with the law. Secret and retroactive laws are impossible according to the rules of legislation, and hence also according to the rule of law, because they are an affront to freedom. (See the work of Lon L. Fuller for a more detailed version of this argument).

Hence, freedom is an important part of the rule of law, just like publicity and equality. So it would be strange to claim that a regime respects the rule of law if its laws violate people’s freedom, equality and public activity (such as speech). That would have to be a diminished kind of rule of law. Maybe the regime in question does respect the rules of legislation and does more than impose any arbitrary set of rules by way of force. But if it does so, it sets in motion a dynamic that will ultimately lead to freedom, equality and publicity because it uses these values in its legislation (although not in its laws). Violations of human rights are initially consistent with the rule of law – correctly understood as more than any arbitrary set of rules imposed by force – but not over time, since the dynamic of the rules of legislation uses values that are likely to infuse the laws themselves rather than merely the rules of legislation. And these values will direct the laws towards human rights since they are the same as the values inherent in human rights.

For example, if you have a law that imposes slavery, this law may initially have been created with respect for the rules of legislation (for instance, it may be a public law that doesn’t criminalize behavior that took place before the publication of the law). But since these rules imply the equality and freedom of all citizens, the law in question will ultimately come to be seen as inconsistent with the system of legislation. Over time, the rule of law will become the rule of the right law.