HEYDON J. The appeal should be allowed.
Background
Counsel for the first respondent referred to the "general antipathy in international law … against retrospectivity" - an antipathy which can be somewhat selectively displayed. Analysis should not be diverted by that antipathy in this case. Nor should analysis be diverted by the characteristic vagueness with which war crimes are defined. Finally, analysis should not be diverted by reflections upon the zeal with which the victors at the end of the Second World War punished the defeated for war crimes. The victors were animated by the ideals of the Atlantic Charter and of the United Nations. The Universal Declaration of Human Rights was about to peep over the eastern horizon. But first, they wanted to have a little hanging.
At least in its application to the first respondent, the Prime Minister's Decree No 81 of 1945 ("the Decree") is not open to criticisms of retroactivity, vagueness or excessive zeal. The provision of the Decree that creates the crime in respect of which Hungary seeks the first respondent's extradition sets up three categories of offender. One is "[a] person who seriously violated international legal rules applicable to war in respect of the treatment of the population of the occupied territories or prisoners of war". The second category refers to a person who "treated the population of the reannexed territories barbarously, misusing the power granted to him". The third category refers to "an instigator, perpetrator or accomplice of the unlawful execution or torture of persons … in Hungary". The case Hungary puts against the first respondent places him in the third category, not the first two categories. Those first two categories are replete with indeterminate and even uncertain phrases. At least in its application to the facts alleged against the first respondent, the third category does not share those vices.
The facts that Hungary alleges against the first respondent are as follows. On 8 November 1944, the first respondent:
"provided patrol service in the territory of Budapest …, with the aim of capturing hiding persons of Jewish origin and presenting them at their place of service. In the course of his patrol service, [the first respondent], reserve ensign, recognized, on a tram line, [the victim], a young man of Jewish origin, also [like the first respondent] residing at Budafok, and known by him earlier, who did not wear the yellow star, which was otherwise prescribed for him on a mandatory basis. Thereafter, [the first respondent] dragged [the victim] to the army post of his Unit located at Budapest … Thereafter, [the first respondent] and his two other accomplices [both also officers], assaulted [the victim] in one of the offices of the Unit from about 3.00 p.m. until the evening hours continuously and so badly that [the victim] died of his injuries late in the evening. [The first respondent] delivered the dead body of [the victim] from the territory of the Unit on the same night together with his several fellow-soldiers, and then, after having fixed ballast to the dead body, they threw it into the Danube."
The first respondent was a Hungarian citizen in 1944. He is alleged to have carried out this conduct on Hungarian soil. The alleged victim appears to have been a Hungarian citizen. Unlike the instruments relied on to try the major war criminals at Nuremburg and Tokyo after the Second World War, the Decree is legislation enacted by a competent legislator. The Decree is supported by the territorial and active personality principles of jurisdiction under international law. Although the Decree is retrospective in form, in substance, as it applies to this case, it merely repeats an existing prohibition against murder. Indeed, the crime the Decree created requires antecedently "unlawful" conduct. In this case, that conduct was murder.
In 1946 the Allies hanged the former German Foreign Minister for crimes against peace. No doubt he was an unlovely character. No doubt other things that he may have done were repellent. But critics find the way he was forced to the scaffold unattractive because the existence of crimes against peace before 1945 was at best questionable. They also find it unattractive because the prosecution comprised four countries with very large empires, significant parts of which had been acquired by starting wars. That was an example of behaviour which gave war crimes trials a bad name. On the other hand, post-1949 German governments could not be criticised for punishing Germans who committed war crimes by murdering or physically maltreating Germans in Germany during the Second World War contrary to German law at that time. The same is true of post-1945 Hungarian governments seeking to punish Hungarians who murdered Hungarians in Hungary in 1944 - a time when the conduct alleged against the first respondent, though the Decree now calls it a "war crime", was criminal under Hungarian law.
A preliminary point
The appellants submitted that to assault a person to the point of death was murder in Hungarian law in 1944. In contrast, the first respondent submitted:
"the alleged conduct as described in the Arrest Warrant … does not contain any allegation concerning mens rea, and the [Commonwealth] Director of Public Prosecutions had advised that under Western Australian law it might constitute one of several offences … While it must be accepted that the [first appellant] was entitled to conclude that the alleged conduct was criminal, he did not have a proper basis to be satisfied - if it were relevant - that it constituted the crime of murder."
Parties who offer this type of submission tend to prompt in the recipient the reflection: "if they see this as an argument which is necessary for success, they must regard the rest of their case as bad".
The submission has two difficulties. First, it is wrong. Secondly, even if it were right it would be immaterial. It is wrong because presuming Hungary succeeds in establishing the facts alleged, the circumstances described in the Arrest Warrant point strongly to murder, whatever mental element was required to establish murder in Hungary in 1944. In particular, the rank of the perpetrators, the way they behaved, the time their conduct took, and the first respondent's alleged conduct after the victim died suggest an intention to cause the victim's death. Further, Hungary advised Australia that it was murder. And even if it were not murder, it would not matter for the purposes of Art 2(5)(a) of the Treaty. All that matters under Art 2(5)(a) is that it was an offence - not necessarily murder - in Hungary in 1944. If the conduct was not murder, it was an offence. It has doubtless been an offence at least since the time of Árpád.
The first respondent's point is an extremely technical one. He appeals to the need to protect liberty. It is true that the protection of liberty often turns on technicality. But is the first respondent's technical point sound? No.
Reasoning
Some of the arguments the appellants advanced are less convincing than others. It is not necessary to deal with them all. The appeal should be allowed for the following reasons.
Was "the offence in relation to which extradition is sought" an offence in Hungary in 1944 within the meaning of Art 2(5)(a)? Hungary describes the offence for which extradition is sought as a "war crime" that falls within the third category outlined above. In the context of Art 2(5)(a), for the reasons given below, the word "offence" refers to the factual criteria necessary to establish criminal guilt. The elements of the offence in relation to which Hungary seeks extradition are assault on a person which is intentional and which causes that person's death. If the first respondent satisfied those criteria, he would have been an instigator of, perpetrator of, or accomplice in the unlawful execution of a person in Hungary. Intentionally assaulting a person and causing that person's death constituted an offence in Hungary in 1944. Even if that offence was not murder, it was still an offence. Therefore "the offence in relation to which extradition is sought" was an offence in Hungary in 1944. The criteria of liability for murder or for any lesser offence corresponded with the criteria of liability for the third category of war crime under the Decree. It is not necessary that the named offence, "war crime", should have existed in Hungarian law in 1944. It is sufficient that the alleged acts or omissions which Hungary contends amount to the named offence constituted an existing offence in 1944, even if that offence had another name.
This construction of the Treaty can be tested as a matter of ordinary English. Assume that the first respondent returns to Hungary. Assume that he is prosecuted for and convicted of the "war crime" alleged. If a questioner later asked him what he was convicted for, an accurate answer would be: "Beating a Jew to death in Budapest in 1944." The questioner could equally accurately answer: "That's murder. That was certainly an offence in Hungary in 1944." The accuracy of these answers is not diminished by the fact that it would have been equally truthful for the first respondent to say that he was convicted of a war crime, and for the questioner to say that there was no offence by that name in Hungary in 1944.
The majority of the Full Court of the Federal Court of Australia concluded that the "offence" referred to in Art 2(2) and (5)(a) for which the Requesting State seeks a person's extradition must be "a known, fixed, entity". Their Honours concluded that the search which sub-Arts (2) and (5)(a) call for is a search for factual correspondence between the "known, fixed, entity" charged and the criminal law of the Requesting State. The first respondent submitted that the word "offence" denoted "a legal construct rather than a set of acts or omissions." He submitted that it referred to "a particular, identified offence with which a person is to be charged (eg 'murder' or 'war crime')." But Art 2(1) refers to "offences however described", and these cannot be "particular, identified offences".
Paragraphs (a) and (b) of Art 2(5) are directed to "determining whether an offence is an offence against the law of" Australia and Hungary. Here, they are directed to determining whether the first respondent's alleged conduct in assaulting a person until he died was an offence against the law of both Australia and Hungary. Article 2(2)(a) provides: "[I]t shall not matter whether the laws of the Contracting States place the acts or omissions constituting the offence within the same category of offence or denominate the offence by the same terminology". That is, it does not matter whether the acts or omissions that Hungary alleges constitute a war crime are placed in the same category as murder under Hungarian law. It does not matter whether the acts or omissions that Hungary alleges constitute a war crime are placed under the category of murder under Australian law. And it does not matter whether Hungary denominates the intentional assault of a person until he dies by the terminology of a war crime or by the terminology of a murder.
Article 2(2)(b) provides that in determining whether the first respondent's alleged conduct in assaulting a person until he died is an offence against the law of both Australia and Hungary: "the totality of the acts or omissions alleged against the person whose extradition is sought shall be taken into account". It also provides: "it shall not matter whether, under the laws of the Contracting States, the constituent elements of the offence differ." Proving some types of war crime under Hungarian law may call for proof of more than must be proved to establish murder. But the totality of the acts or omissions alleged must be taken into account in determining whether the allegation of a "war crime" is an allegation of the crime "murder". It does not matter that the constituent elements of a "war crime" may be greater in number than those of "murder". It does not matter that they may otherwise be different. The language of the Treaty directs attention to "the totality of the acts or omissions alleged". Complying with that direction is inconsistent with concentrating on a "legal construct", a "particular identified offence" or a "known, fixed, entity". Article 2(2)(b) directs that "offence" in Art 2(5) is not used in those latter senses. In fact, in the Arrest Warrant issued against the first respondent the allegation that he committed a "war crime" does not differ from an allegation that he committed a murder.
It is possible to draw attention to incongruities and infelicities in Art 2(5). It is also possible to point to problems which would have been solved if its terms were different. And it is possible to contend that the absence of certain language points to a construction adverse to the appellants' position. There is a certain disconformity in the use of the words "offence", conduct "constituting the offence" and "acts or omissions" in Art 2(5). An analyst could seek to draw conclusions adverse to the appellants from that disconformity. That type of reasoning is common in linguistic construction. It can be overdone. It is easy for counsel to conduct a minute and leisurely examination of a document years after it was drafted and ingeniously detect flaws in the drafting if it is read one way. If those flaws could easily have been remedied, it may be contended that it should be read another way. This is a hypercritical approach. It is reminiscent of the approach criminal defence counsel often take to a summing-up when drafting a notice of appeal.
So far as that type of reasoning has merit, its merit certainly depends on context. The present context concerns a bilateral treaty entered by Australia and Hungary and negotiated by State representatives who, most probably, did not share fluency in a common language. It is true that the drafting might have been better adapted to achieve the result for which the appellants contend. But that fact does not negate the conclusion that the drafting actually employed achieved that result. As Deane J said in Commonwealth v Tasmania (Tasmanian Dam Case): "International agreements are commonly 'not expressed with the precision of formal domestic documents as in English law'."
Notice of Contention: failure to give reasons
The first respondent contended that even if the appellants' arguments in support of the appeal were accepted, the decision of the Full Court of the Federal Court of Australia should nevertheless be affirmed. He submitted that the first appellant made a jurisdictional error by failing to provide a statement of reasons for his 12 November 2009 decision. That decision determined that the first respondent be extradited to Hungary. It was made under s 22 of the Extradition Act 1988 (Cth) ("the Act").
The first respondent alleged in his Further Amended Grounds of Review that the first appellant's failure to supply reasons made his decision "a nullity and of no legal effect".
The first respondent submitted that the first appellant was obligated to give reasons when the decision was made, independently of any request for them. He conceded that there was no common law duty to give reasons. But he submitted that in the federal sphere a duty to give reasons arises from the limited nature of the Commonwealth Parliament's powers. He submitted that s 22 of the Act should be construed as conditioned by an obligation to give reasons. In the alternative, if s 22 could not be so construed, the first respondent submitted that it was invalid. Section 22 cannot be so construed. Accordingly, the first respondent's alternative submission becomes relevant. The first respondent submitted that the Commonwealth Parliament could not confer an unlimited power on the Executive. The power to make an unreasoned decision is an unlimited power. The power to make a decision without giving reasons is a power to make an unreasoned decision. It is equivalent to the power to give an unexaminable decision, contrary to an implication from s 75(v) of the Constitution. The first respondent also argued that the enforcement of limits on statutory power cannot be achieved unless decision-makers give reasons. Therefore, he contended, the provision of reasons was essential to the validity of the provision conferring the statutory power.
These arguments must be rejected. It does not follow from the fact that a decision-maker has not provided reasons that the decision-maker's decision is unreasoned. Nor does it follow that it is unexaminable. The publication of reasons certainly helps those who wish to challenge administrative decisions. But it is not essential to a challenge. A decision-maker can be compelled to produce documents revealing the reasons for a given decision, whether by a subpoena duces tecum or a notice to produce. That decision-maker can be compelled by interrogatories to reveal those reasons in writing, and by a subpoena ad testificandum to reveal those reasons in the witness box. It is true that judicial review proceedings cannot be commenced on an entirely speculative basis. But non-speculative inferences can be drawn from the nature of the decision and from the dealings between the decision-maker and the affected person before the decision was made. It is also true that it would be difficult for a person challenging the decision to frame non-leading questions capable of eliciting answers that would reveal the decision-maker's reasons. But the person challenging the decision can question the decision-maker as though on cross-examination where the decision-maker is not making a genuine attempt to give evidence on a matter of which that decision-maker may reasonably be supposed to have knowledge: Evidence Act 1995 (Cth), s 38(1)(b). Reluctance on the decision-maker's part to give reasons would support an inference that there were no reasons, or no convincing reasons. It would be likely to stimulate close curial scrutiny. That is particularly so of adherence to a code of omerta in the witness box.
Further, it is not possible to derive from s 75(v) of the Constitution an implication that all decision-making powers subject to s 75(v) review must be construed as carrying with them a duty to provide reasons. Section 75(v) is extremely important. But it is a grant of jurisdiction. It is not a source of substantive law governing the conduct of Commonwealth officers in relation to their reasoning processes.
Another difficulty for the first respondent is that the Administrative Decisions (Judicial Review) Act 1977 (Cth), which in s 13 creates a duty in some cases to give reasons, assumes that his argument is flawed. As the first respondent conceded, s 13 of that Act is predicated on the fact that there is no underlying obligation to give reasons. The first respondent endeavoured to meet that difficulty by submitting that his argument was novel, but that novelty was no bar to success. However, his argument cannot overcome the fact that Sched 1 par (r) of that Act specifically relieves the first appellant from the s 13 duty to give reasons. The first respondent's argument can scarcely stand with that express provision about the specific topic at hand.
The first respondent relied on passages in Cunliffe v The Commonwealth. Those passages do not support the first respondent's argument. They record the difficulties of challenging administrative decisions if reasons are not provided. They lament the consequential inadequacy to that extent of curial control over the exercise of administrative powers. But they do not state that a decision for which reasons are not provided is void.
For those reasons the contention the first respondent advanced must be rejected.
Orders
The appeal should be allowed. By reason of their undertaking recorded in the argument on the special leave application, the appellants should pay the first respondent's costs of the appeal. The appellants seek their costs of the Notice of Contention, however. No argument was advanced against that application. The appellants are entitled to those costs. Consequential orders should be made, including the dismissal of the Further Amended Application.