Federal Law Review
|
|
Home
| Databases
| WorldLII
| Search
| Feedback
Federal Law Review |
|
In an illuminating analysis of the concept of justiciability, Geoffrey Marshall distinguishes two senses of the term. In the first, descriptive sense, justiciable issues are issues which have, in fact, been committed by parliaments to a judicial forum. Any question which has been made subject to adjudication is, in the descriptive sense of the word, a justiciable question. In the second, prescriptive sense, justiciable issues are issues which are suitable to be resolved judicially.[2] In the United Kingdom, there is no obstacle to Parliament requiring courts to resolve questions which are not suitable to be resolved judicially. By contrast, in terms of the Australian Constitution, the Chapter III courts are limited to answering questions which are justiciable in the prescriptive sense of the term.
This conclusion follows in part from the finding in R v Kirby; Ex parte Boilermakers' Society of Australia[3] that, by virtue of the separation of judicial power, Chapter III courts cannot be asked to exercise functions 'foreign to the character and purpose of the judicature',[4] except to the extent that such functions are ancillary or incidental to the exercise of judicial power. Chapter III courts can as a result be asked only to decide issues which are of a judicial character. In addition, the fact that the jurisdiction of the Chapter III courts is confined to 'matters' implies that their jurisdiction is confined to judicially cognisable issues. A matter will not exist unless there is 'some immediate right, duty or liability to be established by the determination of the Court'.[5] In these ways, the Constitution confines Chapter III courts to the performance of a specialised set of tasks — judicial tasks. But what exactly makes a task judicial? What makes an issue suitable or unsuitable to be resolved by a court of law? This question was at the heart of the case of Thomas v Mowbray,[6] which involved a challenge to the constitutional validity of Division 104 of the Criminal Code Act 1995 (Cth) ('Criminal Code'), which provides for the making of control orders as one of a new range of counter-terrorism measures.
Terrorism is increasingly perceived as a risk control issue in what has aptly been called the 'world risk society'.[7] President Bush expressed this view when he said: 'If we wait for threats to fully materialize, we will have waited too long'.[8] The control order regime contained in Division 104 of the Criminal Code exemplifies an anticipatory response of just this kind. It enlists federal courts in what Clive Walker calls the tactics of prevention, disruption and countering,[9] by authorising such courts to impose potentially very onerous obligations, prohibitions and restrictions on people who are not charged with a criminal offence 'for the purpose of protecting the public from a terrorist act'.[10]
The prohibitions and restrictions that may be imposed by a control order include prohibitions and restrictions on being at specified areas or places; on communicating or associating with specified individuals; on using certain forms of telecommunication, such as the internet; and on carrying out certain activities in respect of the person's work or occupation. The obligations include the wearing of a tracking device; remaining at a specified place for certain periods; and reporting at specified times and places.[11]
In terms of subdivision B of Division 104, an interim control order may be sought only by senior members of the Australian Federal Police (AFP) after having received the written consent of the Attorney-General.[12] The officer can request an interim order from the Federal Court, the Family Court or the Federal Magistrates Court, all of which derive their jurisdiction from Chapter III of the Constitution. The provision assumes that the order will be made on the officer's ex parte application. Before making an interim order, the court must be satisfied, on the balance of probabilities, that the order would 'substantially assist in preventing a terrorist act' or 'that the person [subject to the order] has provided training to, or received training from, a listed terrorist organisation'.[13] The court must also be satisfied on the balance of probabilities that 'each of the obligations, prohibitions and restrictions to be imposed by the order is reasonably necessary, and reasonably appropriate and adapted, for the purpose of protecting the public from a terrorist act'.[14] In making this determination, the court must take into account the impact of the control order on the person's circumstances, including financial and personal matters.[15] If the criteria are satisfied, the court 'may' make an interim order.[16]
Subdivision D of Division 104 provides for a further hearing — as soon as practicable but at least 72 hours after the making of the interim order — at which the subject of the order may attend a court for it to confirm, vary, revoke or declare void the interim order.[17] The court must be satisfied of the same matters as when an interim control order is sought. The senior AFP officer must provide relevant documents to the subject of the control order so that he or she can understand and respond to the matters raised.[18] At the hearing, both sides can call evidence and make submissions, although the person who is the subject of the control order may be denied access to material if the disclosure of the information is likely to prejudice national security within the meaning of the National Security Information (Criminal and Civil Proceedings) Act 2004 (Cth).[19] A control order, once confirmed, can last for up to a year, with the possibility of renewal.[20] There is a severe penalty — up to five years' imprisonment — for contravening a control order.[21]
In 2006, the Victorian Court of Appeal overturned the conviction of Jack Thomas on charges of receiving funds from a terrorist organisation and possessing a falsified passport, on the basis that certain admissions he had made were not voluntary and should not have been admitted in the trial. Soon after Thomas was released from custody, he was subjected to an interim control order based on the same allegations as had been dismissed in the criminal proceedings. In the control order proceedings, however, the government had to meet only the civil standard of proof as explained above. In Thomas's case, he was required to remain in his house from midnight to 5 am every day and report to police three times a week. He was prevented from leaving Australia, from using any telephone or email service not approved by the AFP, and from communicating with specified individuals.
Before the interim order could be confirmed, Thomas challenged the validity of Division 104 in the High Court, arguing that it was not supported by a head of legislative power and that, if made within federal legislative power, it was repugnant to Chapter III of the Constitution, either because the powers conferred were non-judicial or because they were required to be exercised non-judicially.
This article is concerned only with the Chapter III challenge. This was rejected by a majority of 5:2. The majority (in separate judgments written by Gleeson CJ, Gummow and Crennan JJ, Callinan J and Heydon J)[22] found that the powers conferred by Division 104 are judicial and that issuing courts are not required to act in a way which is inconsistent with the nature of judicial power. Kirby J and Hayne J dissented. Kirby J found that Division 104 invalidly attempts to confer non-judicial power on federal courts and, furthermore, that several features of the legislation are incompatible with the way in which judicial power may be exercised under Chapter III. Hayne J confined himself to the former issue, finding that the jurisdiction which the legislation purports to give to federal courts is not jurisdiction to decide a matter and that the task assigned is not to exercise the judicial power of the Commonwealth.
Some commentators have objected to the majority decision from an explicitly human rights perspective, criticising the majority judges for their focus on the 'technicalities' of the notion of judicial power,[23] their 'cautious conservatism … in human rights matters',[24] and their 'positivistic' refusal to exploit the human rights implications of the fact that the Constitution is framed on the assumption of the rule of law.[25] Although I am sympathetic to this line of argument, my approach is different. I criticise the majority's decision on its own terms. Instead of relying on unwritten constitutional principles, I focus on the concept of a question which is suitable to be resolved judicially and the purposes of confining the judiciary to such questions. In my view, the resources of Boilermakers are sufficient to cast doubt on the constitutionality of the control order regime, making it unnecessary to invoke the more contested human rights arguments.
It is generally accepted that the paradigm case of judicial power involves the authoritative determination of disputes about existing legal rights according to determinate legal standards on the basis of past events or conduct.[26] It is clear, however, that this is not the only judicial function and that there are other tasks which courts can legitimately be asked to do. In particular, in addition to allowing some limited exceptions to the Boilermakers principle, such as the legislative power of superior courts to make rules of procedure, the High Court has for pragmatic reasons permitted a considerable degree of overlap between the powers that may be entrusted to the different branches of government. It has repeatedly held that some governmental powers are not distinctively legislative, executive or judicial and that such 'innominate' powers can be assigned to any of the three branches of government. These powers take their character from the repository of the function and the way in which and purpose for which the powers are to be exercised. This is the so-called 'chameleon doctrine'.[27] The High Court has also consistently stated that judicial power is an 'amorphous' notion,[28] and that there is no single formulation by which it can be captured.[29]
An example of the chameleon doctrine at work is the case of R v Quinn; ex parte Consolidated Foods Corporation.[30] In this case, the High Court held that the power to cancel the registration of trade marks was judicial when exercised by the High Court but non-judicial when exercised by the Registrar of Trade Marks. This was despite the fact that the decision affected rights (albeit of a statutory kind), was dependent on objective standards, resolved a controversy and was final. In the Consolidated Foods Corporation Case, the chameleon doctrine was used to uphold the conferral of seemingly judicial power on an administrative agency but it can also be used in reverse, to uphold the conferral of seemingly non-judicial power on a court. Thus, as Kitto J explained in R v Spicer; Ex parte Australian Builders' Labourers' Federation,[31] a grant of a power 'not insusceptible of a judicial exercise [may be] understood as a grant of judicial power because the recipient of the grant is judicial'.[32] This is frequently found to be the case in respect of powers which are not exclusively judicial but historically have been exercised by the courts.[33]
The chameleon doctrine has been criticised for leading to contrived and artificial reasoning,[34] and for making the idea of the separation of judicial power virtually meaningless,[35] but even it has its limits: a power which is 'foreign' to the judicial function cannot be converted into judicial power by reposing it in the judiciary. For example, the power of deciding whether a person is entitled under legislation to preference in employment has been found to be invalidly reposed in the judiciary.[36] Conversely, a core judicial function cannot be turned into a non-judicial function by reposing it in the political branches of the government. For example, punishment for breach of the law is an exclusively judicial function. Thus a legislative judgment of guilt which inflicts punishment without a judicial trial is a breach of the separation of judicial power.[37]
In Chu Kheng Lim v Minister for Immigration, Local Government and Ethnic Affairs,[38] Brennan, Deane and Dawson JJ went further. They stated that, apart from certain exceptional cases, 'the involuntary detention of a citizen in custody by the State is penal or punitive in character and … exists only as an incident of the exclusively judicial function of adjudging and punishing criminal guilt.'[39] It would follow from this, first, that federal legislation could not authorise courts to detain individuals for other than punitive reasons, and, second, that administrative detention will ordinarily amount to punishment and therefore be in breach of the separation of federal judicial power. Whether these propositions can be described as established doctrine is, however, uncertain.
In so far as the first proposition is concerned, one possible qualification concerns sentences imposed for protective purposes in addition to punitive purposes. It appears that, in exceptional circumstances, indefinite sentences, imposed in part to protect the public from the commission of future crimes, may not be in breach of Chapter III.[40] It is, however, unclear whether courts exercising federal jurisdiction could be empowered other than at the time of sentencing to order the incarceration of individuals not for punitive purposes but on the basis of their dangerousness to the community.
Fardon v Attorney-General (Qld)[41] upheld State legislation which conferred the power to detain in custody prisoners who had served their sentences on the basis that they posed an unacceptable risk of serious sexual offence. The case was decided, however, with reference to the principle that the State legislature had not attempted to confer a function on a State court which impaired the institutional capacity of that court to exercise federal jurisdiction. The High Court left open the question whether analogous federal legislation would be valid. Furthermore, since those subject to the preventive detention regime upheld in Fardon had committed an offence as determined by normal judicial processes and it was thought probable that they would offend again, it must be a fortiori uncertain whether federal legislation could empower courts to detain in custody for preventive reasons individuals who have not previously been convicted of an offence.
In so far as the second proposition is concerned, the Lim protection against administrative detention of persons who have not breached the law, as determined by a court, seems to have been wound back in light of recent cases which suggest that the focus of the enquiry should be on the purpose of the law authorising the detention and its connection with a head of power, not an 'a priori proposition that detention by the Executive other than by judicial order is, subject to recognised or clear exceptions, always punitive or penal in nature.'[42] However, even on this more deferential approach, which would appear to give more scope to the executive to detain individuals provided Parliament can package the purpose as 'protective', the notion of core judicial functions which cannot be reposed in the other branches of government is preserved. For it remains the case that only courts can adjudicate on and punish criminal guilt.
If these various limits to the chameleon doctrine did not exist, the legislature would be entirely at large to bypass the separation of powers. This point was expressly made by the dissenting judges in Thomas,[43] and the other judges implicitly accepted it by focusing on the question whether the powers conferred by Division 104 are susceptible of judicial exercise. It therefore seems that Boilermakers has not passed its use-by date. Although the Court's flexible approach to the concept of judicial power means that judicial power is not as strictly quarantined from non-judicial power as might have first appeared from Boilermakers, the Commonwealth was wrong to argue in Thomas that Boilermakers 'does not matter much any more'.[44] Blurred and overlapping though the distinction between judicial power and non-judicial power may be, there are some principled limits on the powers that can be conferred on Chapter III courts.[45]
There are a number of ways in which the powers conferred by Division 104 depart from the central case of judicial power explained above. In considering whether the legislation crossed the line, the majority judges approached their task by examining the impugned elements individually. Proceeding in this way, they were able to find historical precedents or analogies for each element considered in isolation from the others. This led the majority to conclude that there was nothing essentially non-judicial in the package. I will give reasons in Part V for thinking that this approach and the conclusion to which it led are wrong — that the majority ought to have considered the cumulative effect of the departures from the core exercise of judicial power — but I will begin by outlining their analysis of each element of the legislation.
It is clear that the task of the courts in making a control order is not to resolve a dispute between contesting parties as to existing rights and duties but to impose new rights and duties — a function usually associated with either legislative or administrative power. It has, however, been accepted that the creation of new rights and duties may be compatible with the exercise of judicial power. Gleeson CJ made reference in Thomas to cases of matrimonial causes, bankruptcy, probate and the winding up of companies.[46] The fact that control orders create new rights and duties was therefore not, in his view and that of the other members of the majority, an insuperable obstacle.
A further Chapter III issue related to the imprecision of the standards contained in the legislation — the fact that the court has to be satisfied that the obligations, prohibitions, and restrictions to be imposed are reasonably necessary and reasonably appropriate and adapted for the purpose of protecting the public from a terrorist act. The majority judges conceded that the standards confer a wide discretion on courts but they nevertheless considered them to be objectively ascertainable and therefore judicially manageable. They variously pointed out that courts are often asked to make judgements as to whether a measure is reasonably necessary or reasonably appropriate and adapted;[47] to work with broadly expressed standards such as unreasonableness and injustice;[48] and to consider the protection of the public — for instance, in the sentencing context.[49] The matters which the courts are required to consider before making a control order are therefore not, they found, non-justiciable or insusceptible of judicial determination.
Turning next to the fact that the rights and obligations created by control orders depend on the prediction of future circumstances, and involve, in particular, interfering with liberty on the basis of an assessment of future risk rather than as an incident of judging criminal guilt — the more traditional judicial function — the majority pointed to other instances in which it is accepted that courts may exercise powers of this nature, such as the granting of bail, the making of apprehended violence orders and the traditional power to bind persons over to keep the peace.[50] The majority judges also emphasised the point that, in so far as Lim suggests that legislation cannot require federal courts to detain individuals for preventive reasons, control orders involve restrictions on liberty which fall short of detention in custody.[51]
Finally, the majority judges rejected the argument that issuing courts are required to act in a way which is inconsistent with the essential character of a court or the nature of judicial power.[52] Gleeson CJ dealt with the matter in the most detail, saying that the 'purpose of conferring this function on a court is to submit control orders to the judicial process, with its essential commitment to impartiality and its focus on the justice of the individual case',[53] and he added that there is 'nothing to suggest that the issuing court is to act as a mere instrument of government policy'.[54] Applications are made in open court, subject to the power to close the court under the court's general statutory powers; the rules of evidence apply; the burden of proof is on the applicant; prior to the confirmation hearing, the subject of a control order is provided with the documents necessary to respond; the confirmation hearing involves evidence, cross-examination and argument; the court has a discretion whether to revoke, vary or confirm the order; an appeal lies in the ordinary way; and the court must take into account the impact of the order on the subject's personal circumstances.[55]
One troubling aspect of Subdivision D of the Criminal Code is the fact that information may be withheld from the controlled person on ground of national security in terms of s 104.12A(3). The majority did not rule on the validity of this provision. It confined itself to upholding the validity of the interim control order regime as contained in Subdivision B, noting that Subdivision D had not been engaged in respect of Thomas, who had challenged the legislation before the further hearing for which Subdivision D provides. The possibility was therefore left open that s 104.12A(3) of the legislation might raise Chapter III issues.[56]
I turn now to evaluate the majority's views about the compatibility of the interim control order provisions with Chapter III. I will argue that there are two respects in which the task imposed on judges by the control order regime is not suitable for courts of law. Firstly, it is not sufficiently constrained by pre-existing law. It is therefore closer to a political task than a legal task. Secondly, it involves enlisting courts in projects of social control which are at odds with their character and purpose. I will suggest that the combination of these failings should have been seen as fatal.
In elucidating these failings, it is helpful to draw on Lon Fuller's analysis of the nature of law. Fuller explains that law is distinguished from the exercise of arbitrary power by its purpose. It does not aim merely to achieve public order.[57] That could be done by 'frighten[ing] us into impotence'.[58] Instead, law's distinctive aim is to subject our conduct to the governance of rules.[59] One implication of Fuller's analysis is that if the legislator wishes to control behaviour by legal means it must enact rules which are clear enough to be capable of constraining the reasoning of the judges who have to apply them. If the legislator's rules are so vague that it is impossible to predict how courts will apply them, then the courts will not have been enlisted in the task of guiding behaviour according to pre-existing standards. Judges empowered to act on their own views have not been empowered to apply law. Instead, they have been given a political task — that of deciding which outcome they think best.
Of course, the line between legal and political decision-making is not a bright-line distinction. This does not, however, gainsay the claim that legal decision-making involves a distinctive kind of reasoning which is more objective and consequently more predictable than political decision-making. Legislators make their own, independent moral and political assessments, unconstrained by the decisions of their predecessors. Judges are in a different position. Although they frequently need to rely on moral and political reasoning, they are constrained by what John Finnis calls 'the contingencies of the existing posited law'.[60] These existing standards limit the extent to which judges' personal views can influence their decisions. Adjudication therefore necessarily involves 'some submerging of the judge's own mind … in favour of the community's'.[61] If this were not the case, it would be impossible to distinguish adjudication as a method of dispute-settlement from arbitration on the merits of the case, or from what Joseph Raz calls 'a system of absolute discretion'.[62] Only those who think the law is radically indeterminate and entirely incapable of generating answers to legal problems will disagree.
If it is in the nature of adjudication that it is constrained by external standards, one key question about the control order regime is whether it submerges the judge's own mind sufficiently to entitle us to call the task it gives judges an act of 'judging'. The problem, here, of course, is that the standards contained in the legislation are very imprecise. As we have seen, the Thomas majority did not regard the imprecision of the standards as an insuperable problem, referring to other cases in which conferrals of power in very broad terms had been upheld. But were these cases really similar?
Consider, for instance, the powers granted to the State Supreme Courts under the Matrimonial Causes Act 1959 (Cth) which were upheld in Cominos v Cominos.[63] Section 84(1) of the Act empowered such a court 'to make such [maintenance] order as it thinks proper, having regard to the means, earning capacity and conduct of the parties to the marriage and all other relevant circumstances'. It also empowered such a court, in terms of s 86(1), to require the parties to the marriage to make settlements of property which it considered 'just and equitable in the circumstances'.
Although it is true that these provisions not only enabled such a court to create new rights and duties but also conferred wide discretionary powers on it, the Cominos Court pointed out a number of respects in which the powers in question had a judicial character and did not set the court free to make social policy. Firstly, the powers were to be exercised in the course of a controversy between parties. They were ancillary to the jurisdiction to hear and determine matrimonial causes, which are incontestably judicial proceedings and involve determination of existing rights in accordance with pre-existing law.[64] Secondly, the provisions did not free the court to give effect to its own idiosyncratic conceptions of justice or to act on broad policy considerations unrelated to the facts of the case or to what was appropriate to the means, interests and needs of the parties to and children of the marriage.[65] In so far as the apparently broad standard of 'all other relevant circumstances' was concerned, this had to be read with reference to the purpose of the power, the criterion of relevancy, and the matters specifically mentioned by Parliament, such as the means of the parties to the marriage. These significantly cut down on the considerations to which a court could have regard in interpreting the phrase.[66] Likewise, the phrase 'just and equitable' did not confer a discretion of an arbitrary kind. It was not, Gibbs J said, quoting Windeyer J, 'a matter of unfettered individual opinion', but a traditional and relatively determinate judicial concept.[67]
Wide discretionary powers were also embedded in a traditional judicial context in R v Joske; Ex parte Shop Distributive and Allied Employees' Association,[68] which upheld a power given to the Commonwealth Industrial Court to grant relief if it was satisfied that such relief would not do 'substantial injustice'. As Hayne J pointed out in Thomas, this power was predicated upon a finding that there had been an invalidity in the affairs of an industrial organisation, which presupposed a controversy between parties about whether past action or inaction accorded with identified legal standards.[69] In his judgment in the Shop Distributive Employees Case, Barwick CJ had similarly observed that 'the relief which a court is authorized to give consequentially upon its judicial determination of some situation can rarely, if ever, be denied the quality of an exercise of judicial power'.[70]
It seems, then, that in cases such as Cominos and the Shop Distributive Employees Case, wide discretionary powers were upheld because other, more traditional features of judicial power were present to offset the imprecision of the legislative standards. But there are no such countervailing features contained in Division 104. First, the power conferred by it is not ancillary to the exercise of power which is clearly judicial. Secondly, the courts must, in deciding whether to exercise the power, consider the protection of the public — a very different standard from standards such as 'reasonableness', 'fairness', 'justice' and 'equity'. Though the latter standards are broad, they are susceptible to incremental elaboration, lending themselves to the formulation of more specific rules to guide and confine the exercise of discretion in future cases. Furthermore, they are used to resolve a dispute between identifiable individuals, not to make a decision for the good of society. Finally, they are not as imprecise as they seem because judges have a shared understanding of their evaluative content. They therefore have a legal character. The protection of the public, by contrast, is a much more indeterminate and empirical standard which is less suited to judicial decision-making and allows much more scope for subjective, unpredictable and arbitrary judgments. Both Kirby J and Hayne J made these points in their dissenting judgments in Thomas.[71]
Although, as Leslie Zines points out, the principle that a court can only apply standards that are ascertainable is a matter of degree,[72] there must exist a point at which the line between the ascertainable and the unascertainable has been crossed. In his analysis of the concept of discretion, Kent Greenawalt describes the spectrum of possibilities in the following way. At one end, there is simple factual judgement. In the middle, although difficult judgements are required, we still want to say that only one decision can be considered proper. At the other end, the standards are so open-ended that more than one choice is justified. Greenawalt gives the example of a teacher told to pick the ten pupils to watch a Shakespeare play 'who would get the most out of it'.[73] Although this precludes many grounds of decision — for instance, choosing the pupils the teacher likes best — one teacher could justifiably decide to take the pupils who will best understand the play, while another might equally justifiably decide to take the pupils who have had least exposure to such plays. There is no basis for saying that one choice is right and the other wrong.[74] As Greenawalt explains:
We do not usually speak of someone as being under a duty to reach a particular decision when we think the grounds of testing the correctness of that decision so complex that we cannot say with confidence that a decision is correct, when we do not even know practically how to test correctness and when we do not think that the actor himself has a solid basis for choosing one decision rather than another.[75]
DJ Galligan provides a similar analysis of discretionary judgement, remarking that it is a matter of degree, and that how much discretion a decision-maker has is dependent 'on the degree of choice there is in applying, weighting or creating standards.'[76] He emphasises the fact that even the bearers of the widest discretion are not entitled to act without reasons. What distinguishes discretionary judgement is therefore not the fact that it is unreasoned, but the fact that it is in large degree a matter of personal judgement and assessment.[77]
Obviously, courts charged with the decision whether to impose a control order must make their decision on the basis of reasons. But to what extent is their decision controlled by the standards contained in Division 104? Although these standards are less open-ended than Greenawalt's example of the standard the teacher is instructed to apply, it seems clear that they are quite close to that end of the spectrum. Certainly they are much closer to that end than standards such as those considered in Cominos.
When courts are required to make such difficult factual assessments as to whether a control order would 'substantially assist' in preventing a terrorist act, or is 'reasonably necessary and reasonably appropriate and adapted to protect the public from a terrorist act', taking into account the impact of the control order on the person's circumstances, we have moved some distance from the relatively objective realm of 'judgement', into a more subjective realm, based in policy, in which different answers in response to the same facts could well appear to be equally warranted. There is a large body of criminological research which confirms the unreliability of attempts to predict the risk that an offender may commit a further offence.[78] The predictions required of courts by the control order legislation are even more complex. The considerable subjectivity attached to assessing the risk of a terrorist act if certain precautions are not taken suggests that, on the same facts, one court could reasonably find that a control order was sufficiently connected to the goal of protecting the public, while another could just as reasonably disagree. If so, the criteria contained in Division 104 are not well suited to 'submerge the judge's mind' and the task imposed on judges by the control order regime is much closer to a political task than a legal task.
Lord Bingham reasoned along similar lines in A v Secretary of State for the Home Department,[79] in which the House of Lords had been invited to find that there was no 'public emergency threatening the life of the nation' entitling the British government to derogate from its obligations under the Human Rights Act 1998 (UK). Lord Bingham said that this was not a legal question, because it involved making factual predictions 'of what various people around the world might or might not do, and when (if at all) they might do it, and what the consequences might be if they did.'[80] Such predictions, he said, are 'necessarily problematical [because] reasonable and informed minds may differ, and a judgment is not shown to be wrong or unreasonable because that which is thought likely to happen does not happen.'[81]
Thus far I have given reasons for rejecting the Thomas majority's view that the criteria in Division 104 provide a significant source of constraint on a judge's personal assessments. I turn now to the preventive aspects of the control order regime. As explained above, the majority took the view that the function of restraining liberty in order to prevent predicted future harm is a judicial function, pointing to other instances in which it is accepted that courts may exercise preventive powers, such as the imposition of conditions in a bail order, the making of apprehended violence orders, and the power to bind persons over to keep the peace.
Accepting that it may be permissible for federal judges to restrain liberty for reasons which are at some remove from the central judicial function of imposing restraints in order to punish past conduct, the key question is whether the control order power occupies the same position on the spectrum as these examples. Once again, the fact that we are dealing with matters of degree does not mean that at some point the line does not get crossed: the concept of judicial power must exert some constraints on the circumstances in which federal judges can be directed to participate in the government's strategies for protecting society. Suppose, for instance, that in turbulent times it would help to avert a riot if innocent members of a disliked minority were to be subjected to restrictions on their liberty simply to placate an incensed majority. Or suppose that it would conduce to the protection of the public to restrict some people's liberty solely in order to send a warning to others. Could federal judges be directed to restrict liberty for these sorts of reasons? This seems most unlikely. We reject the idea that these could be judicial tasks because it is not the accepted function of the courts to restrict liberty for arbitrary reasons — reasons which have nothing to do with a danger posed by those whose liberty is to be restricted, let alone one created by their own past conduct. The only consideration on the basis of which the individuals in my examples are required to forfeit their liberty is the protection of the public and it is difficult to accept that judicial power could be enlisted in such a frankly utilitarian project — one which is entirely disconnected from anything individuals have done or are likely to do and is willing, in a matter so fundamental as their liberty, to use people as a mere means to the ends of others, to use Kant's well-known phrase.[82]
If we now return to the examples given by the majority in Thomas of courts restricting liberty without a finding of criminal guilt, it is noteworthy that those made subject to the relevant orders are quite unlike the victims of the utilitarian exercises hypothesised above. As Kirby J observed in Thomas, in these cases the order is often ancillary to other proceedings and/or is directed at protecting an identifiable individual, not society at large. Most importantly, it is the past conduct of the person which leads to the order being made and the order is issued to restrain a breach of the law by that person or is at least directed against what that person is likely to do in the future.[83] Since the restrictions on their liberty are triggered by their own criminal or at least dangerous past conduct — conduct which they would have known could lead to the use of state force against them and which they could have avoided — it cannot be said that they have been 'used' to protect the safety of the public or that the public has been protected at their expense. This consideration offsets the fact that judges are restricting liberty for non-punitive purposes and helps to justify the conclusion that the task is sufficiently judicial in character to amount to the exercise of judicial power.
The question is whether the making of a control order is more like my hypothetical examples or more like the examples discussed in Thomas. I suggest it is more like the former. Before it makes an order a court must be satisfied either that making the order would substantially assist in preventing a terrorist act or that the person has provided training to, or received training from, a terrorist organisation. If satisfied on either score, it must then be satisfied that the restrictions on liberty are reasonably necessary and reasonably appropriate and adapted for the purpose of protecting the public from a terrorist act.
One implication of these provisions is that a control order may be imposed on individuals who pose a statistical risk of danger to society, although not one created by their own past behaviour. It therefore cannot be said that it was in their power to avoid the order. Suppose that the authorities have evidence that there is a cell of terrorists operating from a mosque but do not know who the terrorists are. They also have evidence that the cell is planning an attack. Might it not be thought necessary to impose control orders on all the people who worship at the mosque in order to prevent the attack? If this seems far-fetched, we should not forget that similar reasoning was used to justify wartime detentions. Japanese Americans, for instance, were detained not on the basis of an individualised assessment of the risk they posed to society but on the basis that their membership of a particular ethnic group made them more likely to commit acts of espionage and sabotage.[84] The control order legislation opens the door to using membership of a social group as a predictor of risk in this kind of invidious way, and not even in the circumstances of war. By contrast, the United Kingdom control order legislation provides that there must be some evidence that the controlled person has been involved in 'terrorism-related activity'.[85]
It is true that there may be circumstances in which it is reasonable to subject individuals to restraints on the ground that they pose a statistical risk of danger to society for reasons unrelated to their past conduct, such as the quarantining of persons who have been exposed to an infectious disease. But the 'dangerous group' in such cases is not defined along troubling grounds such as ethnicity or religion. Furthermore, such restraints are reasonable because they are directed to countering a threat posed by the potentially infectious person. By contrast, a control order can be made not only against individuals whose membership of a group increases the probability of their committing dangerous acts but even against individuals who pose no risk at all. This is because a control order need not be directed at preventing the controlled person from causing harm but may be directed, as Kirby J observed in Thomas, at what possibly unknown third parties not subject to the order might do.[86] Once again, a comparison with the UK legislation is instructive. In the UK, the obligations imposed by a control order must be considered 'necessary for purposes connected with preventing or restricting involvement by that individual in terrorism-related activity'.[87]
In failing to include a comparable requirement, the Australian control order regime makes it possible, as Kirby J said, to 'deprive individuals of their freedoms because doing so conduces to the desired control of others,'[88] and, as he went on to note, this is not a function 'hitherto regarded as proper to the powers vested in the Australian judiciary'.[89] My analysis bears out this argument. There are limits to the extent to which federal courts can be enlisted in the risk-management schemes of government. A traditional aspect of our legal arrangements is the judiciary's special responsibility to protect the liberty of the individual. Although courts may in certain circumstances restrict liberty for protective purposes, it is not their function to do so arbitrarily — that is, without paying attention to the question whether the individual whose liberty is sought to be restricted is dangerous to society.
The arbitrary interferences with liberty countenanced by the control order regime are even more troubling when one takes into account the potentially onerous nature of the restrictions, which may be as severe as a criminal penalty,[90] and the fact that breach of a control order is a criminal offence and can be punished with a very severe penalty — up to five years' imprisonment. Although the High Court in Thomas did not subject to scrutiny the government's claim that control orders are protective, one could argue that the combination of severe restrictions and a severe penalty for breach suggests a measure which is punitive in substance. Andrew Ashworth makes a similar argument about anti-social behaviour orders in the United Kingdom. Drawing on the approach of the European Court of Human Rights, which is willing to look behind an ostensibly protective purpose for punishment in disguise, Ashworth argues that where the penalty for breaching a preventive order is so great as to be higher than the penalty for many crimes, and where the restrictions imposed by an order are very onerous, the measure should be regarded as punitive in character.[91]
But if it is true that control orders have a punitive dimension, then the argument for thinking that Division 104 confers non-judicial power is even stronger, at least in respect of its conferral of power to make an order without regard to whether the individual whose liberty is restrained is a threat to society. If it is not a judicial function to order arbitrary interferences with liberty for the good of society, arbitrary interferences approximating punishment are even more foreign to the character and purpose of the judiciary.
In summary: accepting that the judicial function cannot be defined with precision, that there are many tasks which are difficult to categorise as legal or non-legal, that a certain amount of indeterminacy in legal standards can be desirable even at the cost of some loss in certainty,[92] and that a task need only by and large satisfy the requirement of being a judicial task in order to amount to the exercise of judicial power, it is nevertheless clear that at a certain point the line between judicial power and non-judicial power will be crossed. In the borderline cases referred to in Thomas as providing support for the validity of Division 104, there were countervailing features of the context which justified the conclusion that the task was overall sufficiently judicial. By contrast, the functions which judges exercise under the control order regime are at a greater remove from the paradigmatic exercise of judicial power. Furthermore, they depart from the paradigm in more than one way, combining a high degree of discretionary judgement with preventive powers which can be enlivened without an assessment of the risk posed by the person whose liberty is sought to be restricted. These factors work together to magnify the unfamiliarity of the task, as Kirby J observed,[93] so that it can fairly be said that the whole is greater than the sum of its parts.
Finnis observes that 'law brings definition, specificity, clarity, and thus predictability into human interactions'[94] and that it provides 'a normative direction to citizens'.[95] The control order regime does not provide such direction. It not only requires the courts to create new rights and obligations, but requires them to do so by reference to criteria which are at the far end of the spectrum in so far as lack of legislative guidance is concerned. The notions of a terrorist act, and what is reasonably necessary to prevent one, operate as minimal constraints on the court's decision-making and a citizen would be hard pressed to predict how they will be applied in practice. The legislation therefore does not give fair warning to individuals as to the circumstances in which state force will be used against them.[96] This concern is heightened given the importance of the interest at stake: the largely subjective and unpredictable empirical judgements which Division 104 asks of judges are the basis for the imposition of potentially draconian restrictions on a person's liberty coupled with a severe penalty in cases of breach. Clarity would normally be regarded as paramount when the consequences are so serious. And matters are further aggravated by the fact that, by contrast with other preventive orders, control orders are not necessarily based on a person's prior conduct. Since a person's voluntary conduct is not necessarily the basis for the making of a control order, avoiding a control order may be outside a person's power. As Richard Bellamy observes, one can avoid 'falling foul' even of a tyrant's law if the tyrant's rules are known and can be followed.[97] The same is unfortunately not true of the control order regime. Finally, control orders can be made simply to counter threats potentially posed by third parties. We have seen that Fuller distinguishes law from the arbitrary exercise of power on the basis that law does not aim merely to achieve public order — that could be achieved by a reign of terror. Yet the control order legislation permits interferences with liberty purely to achieve public order. In light of the cumulative effect of these failings, I conclude that the legislation does not provide for the legal ordering of conduct and the High Court should have found that federal courts cannot be utilised in an attempt to confer a 'veneer of legality'[98] on the arbitrary form of social control for which it provides.
My argument thus far has been based on abstract analysis of the concept or meaning of judicial power. But there are additional reasons for thinking that the legislation infringes the separation of judicial power. This is because it threatens the purposes which lie behind and explain the doctrine of the separation of judicial power.
The Constitution does not expressly forbid the conferral of non-judicial power on Chapter III courts. Why, then, did Boilermakers insist that the Constitution prevents them from resolving non-justiciable questions or questions which are not suitable for judicial determination? One reason relates to efficiency. Some questions should not be given to courts to resolve because courts are not institutionally well equipped to answer them. NW Barber provides an illuminating account of the way in which form and function are interrelated in this way.[99] He is mainly concerned with the unsuitability of the courts to resolve significantly 'polycentric' issues or issues which have complex and unforeseeable repercussions for people other than the parties appearing before the court.[100] Polycentric issues cannot be satisfactorily resolved by courts because a court's understanding of any given case is largely confined to the submissions of the rival parties before it and the court is as a result unlikely to have knowledge of the future consequences of its decision for other people.[101] Polycentric issues are therefore non-justiciable.[102] For related reasons, as Peter Cane points out, issues which courts do not have the skill, expertise or experience to resolve can be regarded as non-justiciable.[103] In preventing courts from resolving questions which are non-justiciable by virtue of being beyond their institutional capacity, the separation of powers doctrine serves the purpose of allocating governmental tasks to the institutions which can most competently execute them.
Efficiency is not, however, its only purpose. There are also a variety of normative considerations which explain why the Constitution should have wished to place law and politics in separate hands. First, judges should be confined to answering legal questions because they do not have the legitimacy to make the evaluative and policy decisions which politicians are required to make. Secondly, the effective resolution of controversies — especially disputes about the legality of governmental action — would not be possible if judges were not insulated and perceived to be insulated from manipulation by the political branches.[104] Thirdly, a judiciary restricted to resolving disputes in accordance with determinate standards known in advance serves the rule of law. If judges were empowered to decide disputes on the basis of their own views as to what would be the most desirable outcome — if the law were to exert no constraint on their exercise of power — the outcome would be the rule of men, not of laws. As Blackstone wrote: 'were the judicial power joined with the legislative, the life, liberty and property of the subject would be in the hands of arbitrary judges, whose decisions would then be regulated only by their own opinions, and not by any fundamental principles of law'.[105]
We should now test the control order legislation by asking how likely it is to give rise to the dangers against which the Boilermakers prohibition was designed to guard. I have previously made reference to the complex factual predictions, requiring knowledge of the possible conduct of many parties, which a court is obliged to make in deciding whether a control order is likely to assist in preventing a terrorist act. If courts do not have the skill, expertise or experience to make the required predictions, then the Constitution's goal of efficient government has been undermined. I have also argued that the standards contained in the legislation are substantially indeterminate. Such indeterminacy gives rise to the danger of legally uncontrolled and unpredictable judgments on the part of judges, which is a threat to the rule of law. The only remaining question therefore is whether the use of the courts to make control orders is likely to compromise the reputation of the judiciary for independence and impartiality.
Gleeson CJ and Callinan J expressed the view in Thomas that it is better from the perspective of protecting human rights that the power to make control orders be given to independent and impartial courts, acting in accordance with judicial methods, than to members of the executive.[106] Assuming that the power to make control orders could be given to the executive,[107] their Honours' view is attractive on the surface and it is certainly true that the public is likely to feel reassured, at least initially, that the task has been given to judges. The benefits may, however, prove short-lived. For one thing, it is by no means certain that judges are more likely to make an accurate assessment of risk. To assume they will is to overlook the heightened atmosphere of fear which the threat of terrorism has created. As Lucia Zedner remarks, 'it is well documented that risk occupies a larger space in our collective psyche than rational calculation would dictate. Precisely because it is unknowable, prospective risk always threatens to outweigh present interest'.[108] Their Honours' reasoning may also underestimate the deference judges tend to accord the government's judgement on issues of national security. Indeed, the difficulty of testing intelligence information might make deference inevitable. Hayne J made this point in Thomas. He thought that the courts will in practice be forced to accept the government's view of the intelligence material. This, he said, will lead the judiciary to be seen as a compliant instrument of the political branches of government and it will put the courts' reputation as independent arbiters of the law at risk.[109]
More generally, the judiciary enjoys a reputation for independence and impartiality because judges are confined to specialised, legal functions. In using the judiciary to perform a function which is not only politically controversial,[110] but, at best, marginally legal, the control order regime blurs the institutional differences between the judiciary and the political branches of government. In the long run, the use of the courts to restrict the liberty of individuals for the purpose of protecting the public may therefore kill the goose that lays the golden egg.
When judges depart radically from the settled legal framework to achieve social goals they think desirable, or subject administrative decisions to an inappropriately intensive standard of review, or review decisions of a political nature which it would have been better to treat as non-justiciable, there is a justifiable reaction, based on the desirability of keeping law and politics in separate hands. But the policy reasons for separating law and politics apply not only when the political branches of government object to courts intervening in political matters. They also apply when — as under the control order regime — the political branches encourage such intervention. This is because, as Alan B Morrison points out, the separation of powers doctrine is not only about the three governmental 'teams', their territorial interests and what power they happen to be interested or uninterested in exercising.[111] More fundamentally, it is about securing certain interests, such as those in an independent and impartial judiciary, and these interests are threatened not only when the courts fail to show sufficient respect for the political branches of government but also when they make political decisions at Parliament's behest.
This article has argued that the control order legislation does not confer judicial power. This is in part because of the indeterminacy of the legislative standards and in part because of the circumstances in which it permits control orders to be made. The standards are so indeterminate that the question whether a control order should be made is not a question which can be settled in a legal way. And even if the standards were crystal clear and perfectly foreseeable in their application, the legislation permits judges to restrict a person's right to liberty not because that person's behaviour has given rise to an apprehension of harm, and not even to protect the public from a risk of future harm posed by that person, but merely because such restrictions are expected to assist in countering threats posed by other people. This is very close on the spectrum of preventive powers to judicial participation in sacrificing or using individuals for the good of society. Viewed as a package, the legislation therefore amounts to an attempt to confer non-judicial power on federal courts.
I have also argued that the same conclusion follows if we ask whether the power to make control orders is a threat to the purposes behind the separation of judicial power. The separation of judicial power ensures that courts answer only questions that they are competent to resolve, that they have the legitimacy to resolve, that do not impair their capacity or perceived capacity to apply the law impartially and that do not give rise to the risk of judicial lawlessness. I have argued that the complex assessments of risk required of courts by the control order legislation are beyond their institutional capacity. Furthermore, the legislation fails to subject judges to the law and the task it gives them is likely to jeopardise their perceived independence from the political branches of government. For these reasons as well, then, the control order legislation breaches the doctrine of the separation of judicial power.
[*] BA (Witwatersrand), LLB (Cape Town), B Phil (Oxon), D Phil (Oxon); Professor in Law, Division of Law, Macquarie University.
[1] I am grateful to Peter Radan and Alex Reilly for very helpful comments.
[2] Geoffrey Marshall, 'Justiciability' in A G Guest (ed), Oxford Essays in Jurisprudence: A Collaborative Work (1961) 265, 267-8.
[3] [1956] HCA 10; (1956) 94 CLR 254 ('Boilermakers').
[4] Ibid 272 (Dixon CJ, McTiernan, Fullager and Kitto JJ).
[5] Re Judiciary and Navigation Acts [1921] HCA 20; (1921) 29 CLR 257, 265 (Knox CJ, Gavan Duffy, Powers, Rich and Starke JJ). In Thorpe v Commonwealth (No 3) [1997] HCA 21; (1997) 144 ALR 677, 692 Kirby J observed that if a question is not of its nature apt to a court performing court-like functions, 'it matters little in practical terms whether the court … rules that it lacks jurisdiction for want of a "matter" engaging its powers, or … [whether] it says that any such "matter" would be non-justiciable'.
[6] [2007] HCA 33; (2007) 237 ALR 194 ('Thomas').
[7] Ulrich Beck, World Risk Society (1999).
[8] George W Bush, 'Remarks at West Point: "New Threats Require New Thinking"' in M L Sifrey and C Cerf (eds), The Iraq War Reader (2003) 268, 269, quoted in Keith Spence, 'World Risk Society and War Against Terror' (2005) 53 Political Studies 284, 289.
[9] Clive Walker, 'Keeping Control of Terrorists without Losing Control of Constitutionalism' (2007) 59 Stanford Law Review 1395, 1400. See also Andrew Ashworth and Lucia Zedner, 'Defending the Criminal Law: Reflections on the Changing Character of Crime, Procedure and Sanctions' (2008) 2 Criminal Law and Philosophy 21, 40.
[10] Criminal Code s 104.1. A terrorist act is defined as an action or threat of action with certain characteristics. First, it must be done or made with the intention of 'advancing a political, religious or ideological cause' (s 100.1(1)). Second, the intention must be to coerce or influence by intimidation either an Australian or foreign government, or the public, including the public of a country other than Australia (s 100.1(1)). Third, the action which is committed or threatened must satisfy one or more of six criteria. These are causing death, or serious physical harm, or serious damage to property, or endangering life, or creating a serious risk to public health or safety, or seriously interfering with or disrupting certain vital systems (s 100.1(2)).
[11] Criminal Code s 104.5(3).
[12] Criminal Code s 104.2(1).
[13] Criminal Code s 104.4(1)(c).
[14] Criminal Code s 104.4(1)(d).
[15] Criminal Code s 104.4(2).
[16] Criminal Code s 104.4(1).
[17] Criminal Code s 104.5(1)(e).
[18] Criminal Code s 104.12A(2).
[19] Criminal Code s 104.12A(3).
[20] Criminal Code s 104.16(1)(d); s 104.16(2).
[21] Criminal Code s 104.27.
[22] Callinan J generally agreed with Gummow and Crennan JJ's judgment and Heydon J stated that he agreed with all of the other majority judges on the Chapter III issues.
[23] Paul Fairall and Wendy Lacey, 'Preventative Detention and Control Orders under Federal Law: The Case for a Bill of Rights' [2007] MelbULawRw 40; (2007) 31 Melbourne University Law Review 1072, 1088.
[24] Ibid 1092.
[25] Ibid.
[26] The classic statement is Griffith CJ's observation in Huddart, Parker & Co Pty Ltd v Moorehead [1909] HCA 36; (1909) 8 CLR 330, 357:
the words "judicial power" … mean the power which every sovereign authority must of necessity have to decide controversies between its subjects, or between itself and its subjects, whether the rights relate to life, liberty or property. The exercise of this power does not begin until some tribunal which has power to give a binding and authoritative decision (whether subject to appeal or not) is called upon to take action.
See also Re Cram; Ex parte Newcastle Wallsend Coal Co Pty Ltd [1987] HCA 29; (1987) 163 CLR 140, 148-9 (Mason CJ, Brennan, Deane, Dawson and Toohey JJ) quoting Kitto J in R v Gallagher; Ex parte Aberdare Collieries Pty Ltd (1963) 37 AJLR 40, 43: 'the power of judicial determination … includes … "the giving of decisions in the nature of adjudications upon disputes as to rights or obligations arising from the operation of the law upon past events or conduct."'
[27] See, for instance, Gaudron J's exposition of the doctrine in Re Dingjan; Ex parte Wagner [1995] HCA 16; (1995) 183 CLR 323, 360: 'some powers are essentially judicial … while others take their character from the tribunal in which they are reposed and they way in which they are to be exercised and, thus, may be conferred on courts or other tribunals as the Parliament chooses'.
[28] R v Trade Practices Tribunal; Ex parte Tasmanian Breweries Pty Ltd [1970] HCA 8; (1970) 123 CLR 361 ('Tasmanian Breweries'), 396 (Windeyer J).
[29] See, for instance, Precision Data Holdings Ltd v Wills (1991) 173 CLR 167, 188–9 (the Court):
The acknowledged difficulty, if not impossibility, of framing a definition of judicial power that is at once exclusive and exhaustive arises from the circumstance that many positive features which are essential to the exercise of the power are not by themselves conclusive of it.
[30] [1977] HCA 62; (1977) 138 CLR 1 ('Consolidated Foods Corporation Case').
[31] [1957] HCA 81; (1957) 100 CLR 277.
[32] Ibid 305.
[33] See, for instance, R v Davison [1954] HCA 46; (1954) 90 CLR 353, 382 (Kitto J).
[34] John de Meyrick, 'Whatever Happened to Boilermakers? Part II' (1995) 69 Australian Law Journal 189, 190.
[35] P H Lane, 'The Decline of the Boilermakers Separation of Powers Doctrine' (1981) 55 Australian Law Journal 6, 6: 'If the two notions of judicial power and non-judicial power develop towards anonymity their separation becomes meaningless.'
[36] Queen Victoria Memorial Hospital v Thornton [1953] HCA 11; (1953) 87 CLR 144.
[37] See, for instance, Deane J's statement in Polyukhovich v Commonwealth [1991] HCA 32; (1991) 172 CLR 501, 608: 'There are some functions which … have become established as incontrovertibly and exclusively judicial in their character. One … of such functions is the adjudgment of guilt of a person accused of a criminal offence.'
[38] (1992) 176 CLR 1 ('Lim').
[39] Ibid 27.
[40] Veen v The Queen (No 2) [1988] HCA 14; (1988) 164 CLR 465; R v Moffat (1997) 91 A Crim R 557. See also Fardon v Attorney General (Qld) [2004] HCA 46; (2004) 223 CLR 575, 634 (Kirby J) ('Fardon').
[41] [2004] HCA 46; (2004) 223 CLR 575.
[42] Re Woolley; Ex parte Applicants M 276/2003 (2004) 225 CLR 1, 26 (McHugh J).
[43] Thomas [2007] HCA 33; (2007) 237 ALR 194, 288–9 [341]–[344] (Kirby J); 319 [462], 322 [472] (Hayne J).
[44] Ibid 288 [340] (Kirby J); 322 [472] (Hayne J).
[45] Cf Simon Evans, 'The Meaning of Constitutional Terms: Essential Features, Family Resemblance and Theory-Based Approaches' [2006] UNSWLawJl 43; (2006) 29 University of New South Wales Law Journal 207, 235: 'the absence of classical definitions that can be expressed in terms of necessary and sufficient characteristics does not mean that … terms cannot be given a stable, workable meaning.'
[46] Thomas [2007] HCA 33; (2007) 237 ALR 194, 205 [15] (Gleeson CJ, citing Fardon [2004] HCA 46; (2004) 223 CLR 575, 596–7 (McHugh J)).
[47] Thomas [2007] HCA 33; (2007) 237 ALR 194, 207–10 [19] – [27] (Gleeson CJ); 226–7 [100] – [103] (Gummow and Crennan JJ).
[48] Ibid 219–21 [73] – [76] (Gummow and Crennan JJ); 356 [596] (Callinan J).
[49] Ibid 228 [109] (Gummow and Crennan JJ); 356 [595] (Callinan J).
[50] Ibid 205–6 [16] (Gleeson CJ); 222 [79], 229–30 [116] – [121] (Gummow and Crennan JJ); 356 [595] – [596] (Callinan J).
[51] Ibid 206–7 [18] (Gleeson CJ); 229 [116] (Gummow and Crennan JJ). Before Thomas was decided, some commentators took the view that the restrictions on liberty imposed by a control order might be so severe as to amount to detention. If so, and if Gummow J's statement in Fardon were to be accepted that, exceptional cases aside, 'the involuntary detention of a citizen in custody by the State is permissible only as a consequential step in the adjudication of criminal guilt' ((2004) [2004] HCA 46; 223 CLR 575, 612), this would give reason to doubt the constitutionality of at least the more draconian aspects of the control order regime: Andrew Lynch and Alexander Reilly, 'The Constitutional Validity of Terrorism Orders of Control and Preventative Detention' [2007] FlinJlLawRfm 4; (2007) 10 Flinders Journal of Law Reform 105, 121–3. This argument has now been foreclosed by the view taken in Thomas that there is a qualitative difference between detention in custody and other restrictions on liberty. By contrast, the European Court of Human Rights has recognised that restrictions on liberty short of detention in custody may be functionally equivalent to detention. See, for instance, Guzzardi v Italy [1980] ECHR 5; (1981) 3 EHRR 333. The House of Lords took a similar view in the case of Secretary of State for the Home Department v JJ [2007] UKHL 45; [2008] 1 AC 385, in which a majority held that control orders with 18 hour curfews amounted to a deprivation of liberty, not merely a restriction of it.
[52] Thomas [2007] HCA 33; (2007) 237 ALR 194, 211 [30] (Gleeson CJ); 228–30 [112] – [121] (Gummow and Crennan JJ); 357 [598] – [599] (Callinan J).
[53] Ibid 211 [30] (Gleeson CJ)
[54] Ibid.
[55] Ibid.
[56] Ibid 230–1 [122] – [125] (Gummow and Crennan JJ).
[57] Lon L Fuller, The Morality of Law (1964) 107.
[58] Ibid 40.
[59] Ibid 106.
[60] John Finnis, 'Natural Law: The Classical Tradition' in Jules Coleman and Scott Shapiro (eds), The Oxford Handbook of Jurisprudence and Philosophy of Law (2002) 1, 11.
[61] Ibid 36.
[62] Joseph Raz, Practical Reason and Norms (1975) 137.
[63] [1972] HCA 54; (1972) 127 CLR 588 ('Cominos').
[64] Ibid 591 (McTiernan and Menzies JJ); 599 (Gibbs J); 605 (Stephen J); 608 (Mason J).
[65] Ibid 593 (Walsh J).
[66] Ibid 594–5 (Walsh J); 602 (Stephen J).
[67] Ibid 599 citing Sanders v Sanders [1967] HCA 33; (1969) 116 CLR 366, 379–80.
[68] [1976] HCA 48; (1976) 135 CLR 194 ('Shop Distributive Employees Case').
[69] Thomas [2007] HCA 33; (2007) 237 ALR 194, 322 [473].
[70] [1976] HCA 48; (1976) 135 CLR 194, 201.
[71] Thomas [2007] HCA 33; (2007) 237 ALR 194, 230–1 [317] – [319], 281–2 [321] – [322], 291–2 [354] (Kirby J); 323–4 [476] – [477], 332 [515] – [516] (Hayne J). In Tasmanian Breweries [1970] HCA 8; (1970) 123 CLR 361, 376 the fact that a tribunal had the power to determine whether restrictions and practices were contrary to the public interest was held to be a strong indicator that its powers were non-judicial. Kitto J stated that the Act did not require the Tribunal to decide whether the relevant restriction or practice satisfied an 'ascertained standard' but referred the Tribunal 'ultimately to its own idiosyncratic conceptions and modes of thought.'
[72] Leslie Zines, The High Court and the Constitution (4th ed, 1997) 195.
[73] Kent Greenawalt, 'Discretion and the Judicial Decision: The Elusive Quest for the Fetters that Bind Judges' (1975) 75 Columbia Law Review 359, 366
[74] Ibid 365–6.
[75] Ibid 374.
[76] D J Galligan, Discretionary Powers: A Legal Study of Official Discretion (1986) 14.
[77] Ibid 6–8. Cf Ronald Dworkin's analysis of 'strong' discretion — the discretion which exists when '[an official's] decision is not controlled by a standard furnished by the particular authority we have in mind' in Ronald Dworkin,Taking Rights Seriously (1977) 33.
[78] Lucia Zedner, 'Securing Liberty in the Face of Terror: Reflections from Criminal Justice' (2005) 32 Journal of Law and Society 507, 512.
[79] [2004] UKHL 56; [2005] 2 AC 68.
[80] Ibid 102 [29].
[81] Ibid.
[82] Immanuel Kant, The Moral Law: or, Kant's Groundwork of the Metaphysic of Morals (H J Paton trans, 1948 ed) 95.
[83] Thomas [2007] HCA 33; (2007) 237 ALR 194, 285 [331], 287 [338] (Kirby J).
[84] For discussion of this issue see Eugene V Rostow, 'The Japanese American Cases — A Disaster' (1945) 54 Yale Law Journal 489.
[85] Prevention of Terrorism Act 2005 (UK) c 2, ss 2(1), 4(7)(a).
[86] Thomas [2007] HCA 33; (2007) 237 ALR 194, 287 [338].
[87] Prevention of Terrorism Act 2005 (UK) c 2, s 1(3).
[88] Thomas [2007] HCA 33; (2007) 237 ALR 194, 292 [355].
[89] Ibid.
[90] The Council of Europe Commissioner for Human Rights made this point about the UK control order legislation. See Secretary of State for the Home Department v MB [2007] EWCA Crim 2016; [2008] 1 AC 440, 470–71 [16] (Lord Bingham).
[91] Andrew Ashworth, 'Social Control and "Anti-Social Behaviour": The Subversion of Human Rights' (2004) 120 Law Quarterly Review 263, 281.
[92] As H L A Hart believes: The Concept of Law (1994) 130. But contrast Tom Campbell, The Legal Theory of Ethical Positivism (1996) 64.
[93] Thomas [2007] HCA 33; (2007) 237 ALR 194, 291–3 [354] – [358].
[94] John Finnis, Natural Law and Natural Rights (1980) 268.
[95] Ibid 283.
[96] Justin Gleeson makes a similar point, saying: 'under s 104.4, a person could end up being made the subject of a control order … in circumstances where the person could not have known in advance by inspection of the statute book or other reasonable enquiry that the person was engaging in conduct which would or might lead to such an order' ('Thomas v Mowbray' (paper delivered at Twelfth Annual Public Law Weekend, Australian National University, 10 November 2007, available at: <http://law.anu.edu.au/CIPL/Conferences & SawerLecture/2007/PLW%202007/07%20PLW%20Proceedings.htm> ).
[97] Richard Bellamy, 'Introduction' in Richard Bellamy (ed), The Rule of Law and the Separation of Powers (2005) xi, xvii.
[98] See Thomas [2007] HCA 33; (2007) 237 ALR 194, 293 [358] (Kirby J).
[99] N W Barber, 'Prelude to the Separation of Powers' (2001) 60 Cambridge Law Journal 59, 59.
[100] Lon L Fuller developed the concept of polycentricity in 'Forms and Limits of Adjudication' (1978) 92 Harvard Law Review 353.
[101] Barber, above n 99, 75–77. See also Frederick Schauer's comments on 'informational differentiation' — the distinctive way in which courts obtain their information — and its implications for the decisions they should be assigned in 'Legal Positivism and the Contingent Autonomy of Law' in Tom Campbell and Jeffrey Goldsworthy (eds), Judicial Power, Democracy and Legal Positivism (2000) 215, 222-3.
[102] John Allison, 'The Procedural Reason for Judicial Restraint' (1994) Public Law 452, 455.
[103] Peter Cane, An Introduction to Administrative Law (1996) 38.
[104] Cheryl Saunders, 'The Separation of Powers' in Brian Opeskin and Fiona Wheeler (eds), The Australian Federal Judicial System (2000) 3, 10.
[105] Sir William Blackstone, Commentaries on the Laws of England (17th ed, 1830) vol 1, 269.
[106] Thomas [2007] HCA 33; (2007) 237 ALR 194, 206–7 [18] (Gleeson CJ); 355, [592], 356 [595], 357 [599] (Callinan J).
[107] The dissenting judges thought otherwise: ibid 290 [349] (Kirby J); 329–30 [506] (Hayne J). See also Lynch and Reilly, above n 51, 110–1.
[108] Zedner, above n 78, 516.
[109] Thomas [2007] HCA 33; (2007) 237 ALR 194, 331 [512]. See also 296–7 [367]–[369] (Kirby J). For a contrary view, see David Dyzenhaus and Rayner Thwaites, 'Legality and Emergency — The Judiciary in a Time of Terror' in Andrew Lynch, Edwina Macdonald and George Williams (eds), Law and Liberty in the War on Terror (2007) 9, 21.
[110] Andrew Lynch discusses the public outcry which followed the Victorian Court of Appeal's quashing of the conviction of Thomas in 'Maximising the Drama: "Jihad Jack", the Court of Appeal and the Australian Media' [2006] AdelLawRw 9; (2006) 27 Adelaide Law Review 311, 324–30.
[111] Alan B Morrison 'A Non-Power Looks at Separation of Powers' (1990) 79 Georgetown Law Journal 281, 285, 299.
AustLII:
Copyright Policy
|
Disclaimers
|
Privacy Policy
|
Feedback
URL: http://www.austlii.edu.au/au/journals/FedLawRw/2008/8.html