Gordon, Jeffrey Steven --- "Imprisonment and the Separation of Judicial Power: A Defence of a Categorical Immunity from Non-Criminal Detention" [2012] MelbULawRw 2; (2012) 36(1) Melbourne University Law Review 41
IMPRISONMENT AND THE SEPARATION OF JUDICIAL POWER: A DEFENCE OF A CATEGORICAL IMMUNITY FROM NON-CRIMINAL DETENTION
JEFFREY STEVEN GORDON[*]
[The fundamental principle that no person may be deprived of liberty without criminal conviction has deteriorated. Despite a robust assertion of the principle by Brennan, Deane and Dawson JJ in Chu Kheng Lim v Minister for Immigration, subsequent jurisprudence has eroded it and revealed stark division amongst the Justices of the High Court. This article clarifies the contours of the disagreement and defends the proposition that, subject to a limited number of categorical exceptions, ch III of the Constitution permits the involuntary detention of a person in custody only as a consequential step in the adjudication of the criminal guilt of that person for past acts. This article proposes a methodology for creating new categories of permitted non-criminal detention and applies that methodology to test the constitutionality of the interim control orders considered in Thomas v Mowbray.]
I INTRODUCTION
The fundamental principle that no person may be deprived of liberty without criminal conviction has deteriorated. Evidence of the deterioration pervades judicial decisions and academic articles. In Al-Kateb v Godwin (‘Al-Kateb’), the High Court validated the indefinite detention of a stateless asylum seeker who had been refused entry into Australia yet could not be deported.[1] Hayne J declared as ‘open to doubt’ the assumption that there is ‘only a limited class of cases in which executive detention can be justified.’[2] In Re Woolley; Ex parte Applicants M276/2003 (‘Re Woolley’), a case concerning the immigration detention of four Afghan children, McHugh J said that it goes ‘too far’ to maintain that involuntary confinement can only be achieved as the result of the exercise of judicial power.[3] The Aboriginals Ordinance 1918 (NT), which authorised the removal of Indigenous Australian children to institutions and reserves,[4] was validated in Kruger v Commonwealth (‘Kruger’).[5] Gaudron J held that a law authorising detention is not, of itself, offensive to the separation of judicial power.[6] These arguments hold sway in academe. George Winterton doubted the ‘assumption that all involuntary detention (except the recognised exceptions) is necessarily punitive.’[7] Geoffrey Lindell found the suggestion that ch III prevents laws being passed to authorise executive detention ‘strained and unconvincing’.[8]
Faced with such trenchant opposition, this article defends a general prohibition of non-criminal detention, subject to a small number of precisely limited exceptions. In Chu Kheng Lim v Minister for Immigration, Local Government and Ethnic Affairs (‘Chu Kheng Lim’), Brennan, Deane and Dawson JJ recognised a constitutional immunity from being imprisoned without conviction.[9] The rationale for the immunity was that involuntary detention is punitive and ‘exists only as an incident of the exclusively judicial function of adjudging and punishing criminal guilt’.[10] In subsequent cases, the High Court has ‘dismantled the Chu Kheng Lim opinion, dismissing its rule of constitutional immunity from detention outside the criminal process’.[11] The plain-and-simple denial of the immunity implicates several subsidiary questions, but these can be gathered under an inquiry concerning the proper role of ch III of the Constitution when the federal legislature purports to authorise the involuntary confinement of a person, divorced from the criminal guilt of that person for past acts.
In the cases subsequent to Chu Kheng Lim, no dominant methodology for evaluating non-criminal imprisonment has emerged. In fact, there is near-chaotic division amongst the High Court Justices on every subsidiary issue. Some Justices believe that there is no ‘immunity’ from non-criminal detention;[12] another believes that an immunity stems from the limited enumeration of legislative power in s 51 of the Constitution;[13] and still others argue that the immunity derives from the strict separation of judicial power embodied by
ch III.[14] In addition, there is dispute over the proper role of ch III: some Justices argue that ch III has little or no role,[15] and another retains a substantial role for ch III (but one that does not extend so far as to guarantee an immunity).[16] Any discussion of the separation of judicial power engages the punitive/non-punitive distinction, and whether the operation of ch III is characterised by the prohibition of laws conferring penal powers on non-judicial organs. The Justices who accept the punitive/non-punitive distinction advocate differing approaches to that distinction,[17] and another proposes discarding the distinction altogether.[18] Finally, if it is accepted that there is a constitutional immunity, there is disagreement over precisely how that immunity should be implemented: proportionally (weighing the burden of imprisonment against the pursued governmental objective)[19] or categorically (permitting non-criminal detention only if it falls within a predetermined class of permissible detention).[20]
This article disentangles the jumble of questions. Its project is to reorganise and answer them by asserting that, subject to certain categories of permissible non-criminal detention, ch III permits the involuntary detention of a person only as a consequential step in the adjudication of criminal guilt. It invokes the tradition of the common law and draws from the United States (‘US’) experience to establish this position. It develops a principled methodology for admitting new exceptional categories, thereby meeting the criticism that a methodology sensitive to both the historical and analytical approaches to judicial power is unworkable and defeats the expectation of consistency.[21] As far as the author’s research has disclosed, this article is the first sustained academic defence of a categorical immunity from non-criminal imprisonment.
In the US, there has been a parallel decline of the principle that imprisonment can only be effected via the criminal process. In Padilla ex rel Newman v Bush, Judge Mukasey (later Attorney-General) held that the assumption that indefinite non-criminal confinement is per se unconstitutional ‘is simply wrong’.[22] In Hamdi v Rumsfeld (‘Hamdi’), the Supreme Court held that the executive could detain a citizen captured on a foreign battlefield and subsequently brought to US soil without engaging the criminal process.[23] Academic commentators are convinced that the criminal process is a poor weapon to combat the threat of terrorism.[24] The notion that enemy combatants must either be criminally charged or released ‘is simply inconsistent with US law’, because ‘we confine persons against their will for reasons other than punishment in a variety of circumstances’.[25] Adherence to the practice of charging or releasing detainees is ‘not a realistic response’.[26]
This article tracks the modern Australian doctrine of non-criminal detention. Commencing with Chu Kheng Lim, it argues that while legislative power in s 51 generally authorises non-criminal imprisonment, ch III prohibits any law from effecting involuntary non-criminal confinement unless such confinement falls within a categorical exception: detention for the purposes of determining a person’s eligibility to enter Australia; detention for the purposes of preventing and containing the spread of a specific and currently threatening infectious disease; the civil commitment of the mentally ill; detention for contempt of Parliament; and detention resulting from conviction by a court-martial or service tribunal. This article also develops a methodology for admitting new exceptional categories, one that is sensitive to analytical and historical approaches to the concept of judicial power.
II LEGISLATIVE POWER, JUDICIAL POWER AND IMPRISONMENT
During 1989–90, 35 Cambodian nationals arrived in Australia by boat.[27] The Migration Act 1958 (Cth) (‘Migration Act’) created a category of ‘designated persons’ comprising non-citizens who arrived in Australian territorial waters during a specified period and who had neither presented a visa nor been granted an entry permit.[28] Section 54L of the Act imposed an obligation on officials that ‘a designated person must be kept in custody’.[29] This section, as the pivotal provision of the statutory scheme creating mandatory detention, was challenged in Chu Kheng Lim by the Cambodians on the basis that it usurped the judicial power of the Commonwealth which ch III vests exclusively in courts exercising federal jurisdiction.
Section 51(xix), in ch I of the Constitution, provides that the Commonwealth Parliament ‘shall, subject to this Constitution, have power to make laws for the peace, order, and good government of the Commonwealth with respect to ... aliens’. The phrase ‘subject to this Constitution’ permitted the ch III challenge. The Migration Act was defended on the basis that s 54L and related provisions were laws clearly within the head of power, and, moreover, that the power to detain an alien for the purposes of determining his or her entitlement to enter Australia is an executive rather than judicial function. This defence succeeded in the High Court.
Brennan, Deane and Dawson JJ (with whom Mason CJ relevantly agreed) delivered the lead judgment.[30] On the question whether s 54L and related provisions fell within the scope of legislative power conferred by s 51(xix), they held that:
[The provisions] constitute, in their entirety, a law or laws with respect to the detention in custody, pending departure or the grant of an entry permit, of the class of ‘designated’ aliens to which they refer. As a matter of bare characterization, they are, in our view, a law or laws with respect to that class of aliens. As such, they prima facie fall within the scope of the legislative power with respect to ‘aliens’ conferred by s 51(xix).[31]
The answer to the next question was less obvious. Did ch III, by exclusively vesting the judicial power of the Commonwealth in those courts exercising federal jurisdiction, operate to limit Parliament’s capacity to authorise the executive detention of aliens under s 51(xix)? No, according to Brennan, Deane and Dawson JJ, so long as an important rider was added: the detention must be for the purposes of deportation or expulsion or consideration of an alien’s entry permit.[32]
In a crucial passage, Brennan, Deane and Dawson JJ articulated a ‘constitutional immunity’ from executive detention.[33] With reference to Coke and Blackstone, their Honours held:
There are some functions which, by reason of their nature or because of historical considerations, have become established as essentially and exclusively judicial in character. The most important of them is the adjudgment and punishment of criminal guilt under a law of the Commonwealth. That function appertains exclusively to and ‘could not be excluded from’ the judicial power of the Commonwealth. That being so, Ch III of the Constitution precludes the enactment, in purported pursuance of any of the sub-sections of s 51 of the Constitution, of any law purporting to vest any part of that function in the Commonwealth Executive.[34]
In emphasising that such a limitation on legislative power is a matter of substance and not mere form, Brennan, Deane and Dawson JJ said:
It would, for example, be beyond the legislative power of the Parliament to invest the Executive with an arbitrary power to detain citizens in custody notwithstanding that the power was conferred in terms which sought to divorce such detention in custody from both punishment and criminal guilt. The reason why that is so is that, putting to one side the exceptional cases to which reference is made below, the involuntary detention of a citizen in custody by the State is penal or punitive in character and, under our system of government, exists only as an incident of the exclusively judicial function of adjudging and punishing criminal guilt.[35]
Brennan, Deane and Dawson JJ noted that ‘[t]here are some qualifications’ to be made to the general constitutional prohibition on executive detention.[36] One is the detention of an alien for the purpose of deportation.[37] Three reasons were offered to carve out this exception. First, the constitutional immunity from executive detention was expressed to apply prima facie to citizens only. The Constitution maintains a distinction between citizens and aliens, and so an alien’s status, rights and immunities under Australian law differ from a citizen’s status, rights and immunities. In particular, an alien is vulnerable to exclusion or deportation.[38]
Second, the power to exclude and deport an alien is ‘prima facie executive in character’:[39]
In this Court, it has been consistently recognized that the power of the Parliament to make laws with respect to aliens includes not only the power to make laws providing for the expulsion or deportation of aliens by the Executive but extends to authorizing the Executive to restrain an alien in custody to the extent necessary to make the deportation effective.[40]
Support for this was gleaned from international law, where the power to exclude and deport is seen as a concomitant of territorial sovereignty.[41] When conferred upon the executive, the power to detain an alien for the limited purpose of determining an entitlement to remain is incidental to, and takes its character from, the executive powers to exclude, admit and deport.[42]
Finally, the detention authorised by the Migration Act was not punitive, that is, was not for the purpose of punishing an alien for past conduct constituting a breach of Australian law. Brennan, Deane and Dawson JJ referred to a line of authority for the proposition that the executive power to deport an alien is not ‘punitive in nature’.[43] The ancillary power to detain an alien for the purposes of deportation is similarly non-punitive.
Having carved out the exception, Brennan, Deane and Dawson JJ concluded that a law (enacted purportedly pursuant to s 51(xix)) authorising detention will be valid if the detention that it requires and authorises ‘is limited to what is reasonably capable of being seen as necessary for the purposes of deportation or necessary to enable an application for an entry permit to be made and considered’.[44] The Migration Act passed this test by virtue of a provision that sufficiently limited the power of detention that was conferred. A designated person was to be removed from Australia as soon as practicable if the person requested such removal in writing from the Minister.[45] This provision was necessary for the constitutionality of the legislation because it conferred a perpetual power on a designated person to bring his or her detention to an end, and so the continuation of detention was seen as ultimately voluntary: ‘It is only if an alien who is a designated person elects, by failing to make a request under s 54P(1), to remain in the country as an applicant for an entry permit that detention ... can continue.’[46] As a law authorising detention of an alien that was ‘reasonably capable of being seen as necessary’ for the purposes of admission or expulsion, the Migration Act authorised detention that was incidental to executive power and not part of the judicial power of the Commonwealth.[47]
The articulation of a ‘constitutional immunity’ from executive detention cast the onus on the Commonwealth to show that the detention authorised by the Migration Act was lawful. The successful argument proceeded in two stages. First, a general exception to non-criminal detention exists if such detention is incidental to the executive power to admit, exclude and deport; and second, the impugned legislation was properly characterised as facilitating the exercise of this executive function. Clearly, Brennan, Deane and Dawson JJ were of the view that any exception to the immunity must be strictly circumscribed.
By virtue of Mason CJ’s agreement,[48] the judgment of Brennan, Deane and Dawson JJ constituted the majority of the High Court on the question whether a law authorising the imprisonment of a class of aliens violates ch III. Of the other Justices, Toohey J held that the requirement of detention in custody pending removal ‘accords with judicial recognition of the power of the Parliament to authorize the Executive to hold an alien in custody in order to ensure his or her deportation’.[49] Gaudron J was ‘not presently persuaded that legislation authorizing detention in circumstances involving no breach of the criminal law and travelling beyond presently accepted categories is necessarily and inevitably offensive to Ch III’.[50] Rather, the detention regime was valid simply because it was ‘capable of being seen as appropriate or adapted’ for the purposes of deportation or for the making and consideration of an entry application.[51] McHugh J accepted that ch III would invalidate a law authorising the detention of an alien if it went beyond what was reasonably necessary to effect deportation or consider an entry application,[52] and he agreed that ch III did not have that operation in this case.[53] This was because the imprisonment was not punitive in character and did not go beyond what was reasonably necessary to achieve the non-punitive object of preventing an alien from entering the Australian community until a determination on his or her application for entry is made.[54]
The diversity of opinion reflected in the various judgments of Chu Kheng Lim raises more questions than it settles. Subsequent cases have significantly eroded the methodology embodied by the judgment of Brennan, Deane and Dawson JJ, and this has done unforeseen violence to Australia’s Constitution. The above discussion reveals three critical questions thrown up by Chu Kheng Lim, each of which has been elaborated in subsequent case law:
a) From which section does the so-called ‘constitutional immunity’ from executive detention originate?
b) What is the nature of the relationship between valid detention and the constitutional immunity?
c) What is the typology of the so-called ‘exceptional cases’?
A From Which Section Does the So-Called ‘Constitutional Immunity’ from Executive Detention Originate?
The stream of High Court authority emanating from Chu Kheng Lim has debated the question whether a constitutional immunity derives from ch III (as Brennan, Deane and Dawson JJ held) or from the terms of s 51 (as Gaudron J maintained). However, care is required in asking whether a constitutional immunity should be sourced in ch III or in s 51. There are three related questions that need to be separated:
1 Does the legislative power of the Commonwealth conferred by s 51 extend to authorising imprisonment generally?
2 Does ch III have any operation when Parliament purports to enact a law authorising imprisonment?
3 Is there a constitutional immunity from executive detention?
An answer to one of the questions does not foreclose an answer to another, except that an affirmative answer to question 1 and a negative answer to question 2 combine to compel a negative answer to question 3. Questions 1 and 2 are anterior to question 3, and influence the final form that question 3 takes. Where questions 1 and 2 are both answered in the affirmative,
question 3 becomes, ‘Does ch III create a constitutional immunity from executive detention?’ Where both questions 1 and 2 are answered in the negative, question 3 becomes, ‘Does s 51 create a constitutional immunity from executive detention?’ Finally, where question 1 is answered in the negative and question 2 in the affirmative, question 3 becomes, ‘Do s 51 and ch III combine to create a constitutional immunity from executive detention?’
It is the thesis of this article that questions 1, 2 and 3 should all be answered ‘Yes’.
1 Does the Legislative Power of the Commonwealth Conferred by Section 51 Extend to Authorising Imprisonment Generally?
In Chu Kheng Lim, Brennan, Deane and Dawson JJ held that the provisions of the Migration Act mandating detention, as a ‘matter of bare characterization’, fell within the scope of s 51(xix).[55] Evidently, their Honours felt that this did not require extensive justification, contenting themselves with the observation that the provisions constituted laws with respect to the detention in custody, pending departure or the grant of an entry permit, of the class of ‘designated’ aliens.[56]
Gaudron J was not so easily convinced. She held that the legislative power of the Commonwealth generally does not extend to imposing mandatory detention on aliens which is unconnected with the aliens’ entitlement to remain in Australia and which is not appropriate and adapted to regulating entry or facilitating departure, because
[a] law of that kind does not operate by reference to any matter which distinguishes aliens from persons who are members of the community constituting the body politic, nor by reference to the consequences which flow from non-membership of the community.[57]
In Kruger v Commonwealth, she elaborated: ‘the true constitutional position is that, subject to certain exceptions, a law authorising detention in custody, divorced from any breach of the law, is not a law on a topic with respect to which s 51 confers legislative power’.[58]
This approach, which reads into s 51 a general limitation on the competence of Parliament, misconceives the nature of the conferral of legislative power by the Constitution. Section 51 confers plenary legislative power with respect to the subject matters named therein. In Bank of NSW v Commonwealth, Dixon J explained:
The purpose of the enumeration of powers in s 51 is not to define or delimit the description of law that the Parliament may make upon any of the subjects assigned to it. Speaking generally, the legislative power so given is plenary in its quality. The purpose of the enumeration is to name a subject for the purpose of assigning it to that power.[59]
It follows that it is inappropriate to limit, in advance, the scope of the various heads of power in s 51 by holding that most of them do not allow Parliament to authorise detention divorced from any breach of the law. Such an approach seeks to ‘delimit the description of law that the Parliament may make upon any of the subjects assigned to it’.[60] The framers of Australia’s Constitution ‘were not prepared to place fetters upon legislative action, except and insofar as it might be necessary for the purpose of distributing between the states and the central government the full content of legislative power’.[61]
Accordingly, free from any ch III restrictions and considered on its own (or ‘[a]s a matter of bare characterization’),[62] s 51 plainly permits the authorisation of executive detention. The force of this argument was conveyed by Gummow J in Al-Kateb:
it could not seriously be doubted that a law providing for the administrative detention of bankrupts in order to protect the community would be a law with respect to bankruptcy and insolvency (s 51(xvii)), or that a law providing for the involuntary detention of all persons within their homes on census night would be a law with respect to census and statistics (s 51(xi)). If such laws lack validity, it is not by reason of any limitation in the text of paras (xvii) and (xi) but by the limitation in the opening words of s 51, ‘subject to this Constitution’, which attract any limitation required by Ch III.[63]
Stephen McDonald has made the same point.[64] The confusion of the contrary view is captured by Gaudron J’s cryptic comment that ‘[t]here is, however, no decision of this Court that compels the conclusion that a law which operates on or by reference to aliens ... is, on that account, a valid law with respect to aliens’.[65] This ignores that ‘[t]here is no limitation to the power in the words of the placitum; and unless the limitation can be found elsewhere in the Constitution, it does not exist at all’.[66]
Aside from erroneously maintaining that s 51 itself did not extend to authorise executive detention, Gaudron J favoured the s 51 approach over the competing ch III approach because the exceptions contemplated by ch III are indeterminate.[67] However, if clarity and certainty are the constitutional virtues that Gaudron J seeks, then it is far from obvious that the approach grounded in s 51 advances those virtues. The primary question, for Gaudron J, is whether the head of power under consideration permits or forbids executive detention. There are 39 heads of legislative power in s 51, including powers to legislate with respect to defence (s 51(vi)), quarantine (s 51(ix)), aliens (s 51(xix)), and the people of any race for whom it is deemed necessary to make special laws (s 51(xxvi)). To advocate a separate approach for each head of power would invite a real lack of uniformity in respect of executive detention and would undermine every citizen’s legitimate constitutional expectation that he or she will not be arbitrarily imprisoned.
Under the s 51 approach, each head of power is potentially an exception to the general proposition that the heads of power in s 51 do not authorise executive imprisonment. The advocates of the s 51 approach contemplate the defence, quarantine, aliens and race powers as exceptions;[68] but on the current state of authority, only the aliens power has received detailed attention and can lay claim to the status of exception to the general rule.[69] When the time comes to decide whether the other candidates are in fact exceptions, what precedent or consistent historical practice may proponents of the s 51 approach appeal to? They may not look to Blackstone, Dicey, or any other traditional common lawyer who considered Parliament supreme and who would have never dreamt of instituting a general limitation on its capacities.[70] Nor may they look to American precedent, where the power to commit to prison is generally seen as punitive and an exclusively judicial function by virtue of art III of the United States Constitution.[71] With each head of power, they must start anew. The s 51 approach denies Parliament power to detain not because arbitrary imprisonment offends the deep and historical common law values in which the Constitution is embedded, but merely because the text of the Constitution contains ‘no general power to make laws with respect to imprisonment or detention’.[72]
Therefore s 51 is the last provision to which an a priori limitation should be applied. Parliament’s ability to authorise imprisonment cannot be generally denied by reference to broad heads of power addressing vastly different subject matters. Who is to know, in the infinite variety of future human experience, whether executive detention will become truly necessary in order to properly give effect to a grant of legislative power in s 51? Or, conversely, whether the connection between a head of power and a law authorising executive detention will at some subsequent stage become so ‘insubstantial, tenuous or distant’[73] that it cannot be considered a law with respect to that head of power?
2 Does Chapter III Have Any Operation When Parliament Enacts a Law Authorising Imprisonment?
Parliament is given power to make any law ‘for the peace, order, and good government of the Commonwealth’ on the subject matters of any of the heads of power in s 51. The qualifier in s 51, ‘subject to this Constitution’, imports any limitation arising out of the strict separation of judicial power embodied by ch III. The question is whether Parliament, in authorising executive detention, impermissibly reposes judicial power in the executive. The answer to this question depends on the relationship between imprisonment and judicial power.
In Chu Kheng Lim, Brennan, Deane and Dawson JJ held that the adjudgment and punishment of criminal guilt is the most important of those ‘functions which, by reason of their nature or because of historical considerations, have become established as essentially and exclusively judicial’.[74] Since the involuntary detention of a citizen exists only as an incident of the exclusively judicial function of adjudging and punishing criminal guilt, a law authorising executive detention is a legislative attempt to confer judicial power on the executive and will be invalidated on that basis.[75]
The reference to ‘nature’ and ‘historical considerations’ indicates two broad approaches to analysing the concept of judicial power.[76] First, although it is generally doubted that a complete formal or analytic definition of judicial power is possible,[77] there are indicia, including: whether the exercise of the power involves a conclusive, binding order and authoritative decision on rights; a declaration of existing rights under existing law; a discretion based on ascertainable legal standards; and a pattern of judicial characteristics.[78] A verdict of guilt and the imposition of imprisonment following a criminal trial meet these criteria. The judicial nature of the power to imprison springs from the effect that the exercise of the power has upon a person’s legal right to liberty.[79]
The second, historical, approach to analysing judicial power recognises that
we have inherited and were intended by our Constitution to live under a system of law and government which has traditionally protected the rights of persons by ensuring that those rights are determined by a judiciary independent of the parliament and the executive.[80]
Those rights ‘are the basic rights which traditionally, and therefore historically, are judged by that independent judiciary which is the bulwark of freedom’.[81] The ‘classic example’ is the ‘governance of a trial for the determination of criminal guilt’.[82] In Re Nolan; Ex parte Young, Gaudron J pronounced that
it is beyond dispute that the power to determine whether a person has engaged in conduct which is forbidden by law and, if so, to make a binding and enforceable declaration as to the consequences which the law imposes by reason of that conduct lies at the heart of exclusive judicial power.[83]
Thus Brennan, Deane and Dawson JJ are on solid ground when they say that the adjudgment and punishment of criminal guilt is, by reason of its nature and because of historical considerations, the most important instance of the exclusive judicial power.[84] Chapter III has an important role to play in evaluating the constitutionality of a law that authorises imprisonment.
Gaudron J resisted this conclusion. In Chu Kheng Lim, she was not persuaded that legislation authorising detention is ‘necessarily and inevitably offensive to Ch III’,[85] and in Kruger her Honour explicitly rejected the proposition that the power to deprive people of their liberty is necessarily judicial power.[86] In support of that conclusion, Gaudron J advanced two arguments. The first was based on her position that s 51 does not extend to authorising laws conferring a power of detention divorced from criminal guilt, unless they are laws with respect to topics relating to defence, quarantine, aliens or the people of any race for whom it is deemed necessary to make special laws.[87] This argument fails, for the reasons set out above.
The second argument was an objection based upon the indeterminacy of the exceptions that Brennan, Deane and Dawson JJ contemplated in Chu Kheng Lim. Gaudron J attacked the notion that the power to authorise detention is exclusively judicial but for clear exceptions.[88] The only way of proving exclusivity is to prove that the exceptions are clear or fall within precise and confined categories.[89] The exceptions recognised in Chu Kheng Lim do not answer this description because, for example, the exceptions relating to mental illness and infectious disease ‘point in favour of broader exceptions relating, respectively, to the detention of people in custody for their own welfare and for the safety or welfare of the community.’[90] Accordingly, ‘it is not possible to say that, subject to clear exceptions, the power to authorise detention in custody is necessarily and exclusively judicial power’.[91]
There is no disputing that the burden of this article is to show that the exceptions contemplated by Chu Kheng Lim are clear and fall within precise and confined categories. However, when seeking to demonstrate that the exceptions neither are clear nor fall within precise and confined categories, it is not sufficient for Gaudron J to locate a justification relating to individual or community welfare and thereby stamp the exceptions as unclear and impossible to categorise.
For one thing, it would be very difficult to locate any legal rule or principle, and any exception contemplated by such a rule or principle, that could not be justified on the very broad basis of individual or community welfare. So why do the welfare-based justifications suddenly point to broader exceptions here? No-one contends that the legislature or executive can authorise detention without trial on the basis of generalised welfare concerns. This article will attempt to show that the recognised exceptions exist because they are based in history, tradition and necessity. There is nothing inconsistent in saying that they exist for the individual’s or the community’s welfare, and that they are clear and precisely confined. The mere fact that the exceptions are susceptible to that kind of abstraction or justificatory ascent does not rule out their clarity and precision.
More importantly, it is simply incorrect to say that the recognised exceptions of civil commitment of the mentally ill and quarantine detention point in favour of broader exceptions based on individual and community welfare. The proper justification for these exceptions may be partly welfare-based, but primarily it acknowledges that the criminal law is unsuitable for these classes of people.[92] A mentally ill person may pose a danger to himself or herself, or to the community, but can only be detained without a criminal conviction because he or she will never hold the subjective mental state necessary to sustain such a conviction. Undeniably, there is an element of individual and community welfare at play, but this is subordinate to the recognition of the inaptness of the criminal law. As Dixon J explained in R v Porter:
it is perfectly useless for the law to attempt, by threatening punishment, to deter people from committing crimes if their mental condition is such that they cannot be in the least influenced by the possibility or probability of subsequent punishment ...[93]
The carefully limited exceptions acknowledge that individual and community welfare is best protected when imprisonment can occur only by criminal process.
The judiciary is the traditional guardian of individual liberty and the countermajoritarian institution of constitutional democracies,[94] and the phrase ‘judicial power of the Commonwealth’ embeds the Constitution in the common law tradition. The careful quarantining of judicial power from the political branches acts as a structural and prophylactic device protecting individual liberty.[95] Under ch III, the debate over precisely what constitutes judicial power can only be argued by reference to fundamental and historical common law values, thereby fostering Australia’s basic political commitments.
3 Does Chapter III Create a Constitutional Immunity from Executive Detention?
Having determined that ch III is the primary actor in analysing the constitutionality of a law authorising executive detention, it remains to ascertain the exact nature of its role.
(a) The Dominant View: Punishment and Cognate Terms
In the High Court, the dominant view of the operation of ch III is that it does not permit laws authorising executive detention that is punitive or penal in character. A significant number of Justices have appealed to the punitive/non-punitive distinction when determining whether a law that authorises imprisonment violates ch III. In Chu Kheng Lim, Brennan, Deane and Dawson JJ relied heavily on the distinction, saying, ‘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’.[96] The power of Parliament to authorise the executive detention of aliens for the limited purpose of deportation or admission was justified largely on the basis that such detention was non-punitive.[97]
In subsequent cases, although the central question continued to be whether the impugned law was penal or punitive, the automatic implication of punishment from involuntary detention has been challenged. In Re Woolley, McHugh J said that it goes too far to say that executive detention ‘is always penal or punitive and can only be achieved as a result of the exercise of judicial power.’[98] The strongest indicator of whether detention is penal or punitive is the purpose of the law authorising detention.[99] Although it is true that, where nothing more appears, the fact that a law authorises or requires detention compels the inference that it is penal or punitive, this will rarely be the case:
The terms of the law, the surrounding circumstances, the mischief at which the law is aimed and sometimes the parliamentary debates preceding its enactment will indicate the purpose or purposes of the law ... [T]he issue of whether the law is punitive or non-punitive in nature must ultimately be determined by the law’s purpose, 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.[100]
McHugh J concluded that if the purpose of a law authorising or requiring detention is ‘purely protective’, then detention under that law will not be regarded as penal or punitive.[101]
In Al-Kateb, Hayne J ‘was most clear in expressing his opinion that in this context Chapter III had little scope’,[102] saying that the assumption that ‘there is only a limited class of cases in which executive detention can be justified ... is at least open to doubt’.[103] Nevertheless, he was also of the opinion that the ch III question must be answered by reference to the punitive/non-punitive distinction.[104] He adopted H L A Hart’s definition of the ‘standard or central case of punishment’:
(i) It must involve pain or other consequences normally considered unpleasant.
(ii) It must be for an offence against legal rules.
(iii) It must be of an actual or supposed offender for his offence.
(iv) It must be intentionally administered by human beings other than the offender.
(v) It must be imposed and administered by an authority constituted by a legal system against which the offence is committed.[105]
Hayne J noted that it may be appropriate to identify treatment of persons as punitive where those persons are not offenders, ‘[b]ut punishment is not to be inflicted in the exercise of the judicial power except upon proof of commission of an offence’.[106] His Honour gave two reasons for holding that the mandatory detention of aliens is not punitive: first, ‘immigration detention is not detention for an offence’;[107] and second, the segregation of those aliens who make landfall without permission is of the same character as the prevention of non-citizens making landfall, which is only punitive ‘in the most general sense’.[108]
McHugh J and Hayne J employed differing methodologies in approaching the question whether a law is punitive (‘the punitive question’), and both methodologies are flawed. First, consider McHugh J, who advocated an all-of-the-circumstances approach to the punitive question. He offered a catalogue of four factors and concluded that the punitive question ‘must ultimately be determined by the law’s purpose’.[109]
In the US, courts have struggled for half a century with the standard set down by the Supreme Court in Kennedy v Mendoza-Martinez[110] to settle the punitive question (expressed as the question of whether a law is punitive or regulatory). Goldberg J set out seven factors for determining whether a statute is penal or regulatory:
Whether the sanction involves an affirmative disability or restraint, whether it has historically been regarded as a punishment, whether it comes into play only on a finding of scienter, whether its operation will promote the traditional aims of punishment — retribution and deterrence, whether the behavior to which it applies is already a crime, whether an alternative purpose to which it may rationally be connected is assignable for it, and whether it appears excessive in relation to the alternative purpose assigned ...[111]
The seven factors will yield, however, to ‘conclusive evidence of congressional intent as to the penal nature of a statute’.[112]
The seven Mendoza-Martinez factors, as well as the paramount consideration of congressional intent inferred from members’ consensus, bear a striking similarity to McHugh J’s four indicia. Unfortunately, this similarity is not a virtue. The Mendoza-Martinez factors have proved notoriously uncertain and capable of easy manipulation. Commentators have variously described them as ‘so “open-ended” as to be meaningless’,[113] ‘misleading and impoverished’,[114] ‘vacuous, circular’,[115] and ‘a seemingly inappropriate standard’.[116] In application, Mendoza-Martinez has been ‘manipulated to the point of “lack[ing] any real content”’[117] and ‘inconsistent at best’,[118] and provides such a highly deferential standard[119] that ‘punishment is, with few exceptions, that which the legislature classifies as punishment’.[120] The jurisprudence has been plagued by ‘the complexities, intracacies [sic], and contradictions involved when courts attempt to set standards or evolve a formula to differentiate between punishment and regulation’.[121]
A cursory examination of the heads of power in s 51 furnishes several examples of the complexities, intricacies and contradictions that would surely follow were the punitive question to centre on the law’s purpose. Consider s 51(xxx) (the relations of the Commonwealth with the islands of the Pacific), and suppose that Parliament determined that the extradition of doctors to urgently counter a chronic shortage in Nauru would greatly further relations with that nation. On McHugh J’s view, ch III would not invalidate legislation authorising the detention of doctors to effect this purpose (it would be a severe distortion to say that such a law punishes doctors for qualifying as doctors). Or, to take another example, if a special law is enacted pursuant to s 51(xxvi) (the race power), then what constitutes a legitimate non-punitive purpose? The same question arises in respect of many other heads of power in s 51.
There is perhaps a more fundamental constitutional difficulty. The constitutionality of the law authorising detention cannot be upheld purely on the basis that the terms of the law and parliamentary debates evidence a non-punitive purpose.[122] Such an approach runs afoul of Australian Communist Party v Commonwealth.[123] A legislative power to authorise non-punitive detention does not enable Parliament to authorise detention which is, in its opinion, non-punitive.[124] For example, a law may authorise the imprisonment of all people suspected of bank robberies where the parliamentary debates justify the law on the basis of ensuring the liquidity of financial institutions, and where the statute provides that it is enacted solely for that purpose.[125] The statute may further provide that ‘any burden, liability or disability imposed by this Act is not punitive’. Despite its absurdity, on McHugh J’s approach it seems conceivable that such a law would pass constitutional muster.
McHugh J gave too little weight to the fundamental constitutional principle that the validity of a law cannot depend on the opinion of the lawmaker.[126]
Hayne J, on the other hand, appealed to Hart’s five-element definition of punishment, noting that the imprisonment of Al-Kateb was not for an offence against legal rules and therefore was not punitive.[127] This is a triumph of form over substance, a literalist approach easily circumvented by a statute’s form.[128] By contemplating a role for ch III, Hayne J concedes that the separation of judicial power is to have some real operative effect;[129] so to say that its application can be avoided by omitting reference to an offence against legal rules seems to mock the separation of powers.[130]
It is a bold claim that punishment for the purposes of the Australian Constitution means precisely what Hart announced to a meeting of the Aristotelian Society in 1959.[131] In any case, Hart anticipated the ‘abuse of definition’ that Hayne J committed.[132] Although Hayne J conceded that it might be ‘appropriate to identify treatment of persons as punitive where those persons are not offenders’,[133] the observation that ‘immigration detention is not detention for an offence’ was crucial to the conclusion that Al-Kateb’s imprisonment was non-punitive.[134] This ‘definitional stop’ is especially tempting when use is made of the condition that there must be an offence against legal rules,[135] and Hart warned specifically against its use.[136]
This misuse of Hart is a philosophical error with serious constitutional consequences: it supports the view that ‘detention camps where aliens are held indefinitely, without having been convicted of an offence, have nothing to do with the judicial power of the Commonwealth’.[137] The criminal law affords a bundle of procedural and substantive protections to those who stand accused. Imprisonment is justified only where a jury in a fair trial ascertains conduct prohibited by existing law, meriting a sentence of imprisonment. The existence of an offence prior to the accused’s act bolsters the claim that imprisonment is justified. Taking the offence away provides no argument that imprisonment is consequently more justified; indeed, the opposite is true. Executive imprisonment cannot be rendered lawful by reason of the absence of an offence. That would turn the rule of law on its head.
Even though Hayne J’s reference to Hart is superficial and misconceived, it does not follow that his broader methodology — appealing to a philosophically complete account of punishment to answer the punitive question — is flawed. Here, the reason that Hayne J’s approach fails is because Hart consciously restricted his attention to ‘the institution of criminal punishment’.[138] Ex hypothesi, executive detention is not criminal punishment, so of course it would never satisfy Hart’s definition. Hayne J’s approach amounts to accepting the absurd proposition that any sanction imposed without criminal trial is for that reason non-punitive.
Hayne J failed to recognise that he and Hart were engaged in different tasks. On the one hand, as noted above, Hart examined ‘the rational and moral status of our preference for a system of punishment under which measures painful to individuals are to be taken against them only when they have committed an offence’.[139] On the other hand, Hayne J attempted to satisfy or vindicate that preference by ascertaining whether Parliament enacted a law that exacted punishment without criminal process, and, if so, to mobilise ch III against it. Hayne J did not have the philosophical luxury that Hart had in examining the rational and moral status of the preference. Rather, Hayne J was obligated to accept the preference as binding. He should have regarded with extreme suspicion the imposition of measures painful to individuals who had not committed a prior offence. Leslie Zines argued that ‘the close traditional connection of detention with punishment should result in at least a prima facie assumption’ that a law authorising detention is punitive.[140] At its root, that is why executive detention must be scrutinised: it detaches a universally acknowledged punishment (imprisonment) from the commission of a prior offence.
(b) The Correct Approach: Effects and Impacts
Under ch III, the constitutionality of legislative or executive action must turn upon the effect or impact of that action rather than the intent or purpose of the actor.[141] The virtue of this approach is its requirement that the validity of state action must depend on the actual outcome of the act, and it attributes no credibility to a legislature or executive that crosses its heart and promises that, while it does imprison people, it does not mean to punish them. The intention animating the executive or legislature may be venomous or it may be compassionate,[142] but in Australia neither branch should be entitled to give effect to this intention by imprisoning anyone without criminal trial:
in a democracy, where safeguards are built in to protect human dignity, the effect of a sanction rather than the reason for imposing it must necessarily be the criterion as to whether an accused is entitled to safeguards provided by the Constitution.[143]
This approach properly focuses on the individual who is the subject of the burden or sanction, ensuring consistency with the constitutional tradition guarding against government encroachment on individual liberty.
The focus on effect has received curial support from Gummow J. In Al-Kateb, citing Blackstone, he held that ‘[i]t is primarily with the deprivation of liberty that the law is concerned, not with whether that deprivation is for a punitive purpose’.[144] He quoted the statement of Scalia J in Hamdi that ‘[t]he very core of liberty secured by our Anglo-Saxon system of separated powers has been freedom from indefinite imprisonment at the will of the Executive’;[145] and reproduced Brennan, Deane, Toohey and Gaudron JJ’s statement in Witham v Holloway that ‘there can be no doubt that imprisonment and the imposition of fines, the usual sanctions for contempt, constitute punishment’.[146] In Fardon v Attorney-General (Qld), Gummow J reaffirmed his position by quoting a remark of Kirby J that loss of liberty is ‘ordinarily one of the hallmarks reserved to criminal proceedings conducted in the courts, with the protections and assurances that criminal proceedings provide’.[147] In turn, Kirby J supported this reasoning in Thomas v Mowbray.[148]
Reference to punishment and cognate terms is unlikely to be helpful at all. At best, the punitive/non-punitive distinction furnishes an inexact and emotive test,[149] because it is ‘heavily charged with subjective emotional and intellectual overtones’.[150] Moreover, it is difficult to imagine any imprisonment — including criminal imprisonment — that could not be characterised as non-punitive. All forms of detention are directed at meeting some social need by segregating persons from the community. Therefore, all detention will have the non-punitive purpose of protecting society, and
[o]nce it is accepted that many forms of detention involve some non-punitive purpose, it follows that a punitive/non-punitive distinction cannot be the basis upon which the Ch III limitations respecting administrative detention are enlivened.[151]
The upshot is that, when confronted with a ch III challenge to a law on the basis that it inflicts non-criminal imprisonment, the first question for a court to resolve is whether the law actually effects a deprivation of liberty or confinement of the person. Blackstone declared that ‘[t]he confinement of the person, in any wise, is an imprisonment’.[152] Thus, house arrest, putting someone in the stocks, arresting or forcibly detaining someone in the street, all count as imprisonment.[153] Therefore, in Al-Kateb, Hayne J would have been perfectly correct to restrict his attention to Hart’s first criterion: does the law involve pain or other consequences normally considered unpleasant? It is true that this may not always be an easy matter to decide, but (in the case of imprisonment) guidance may be taken from a number of discrete areas of the law including, for example, the tort of false imprisonment or, in the US, the question of custody under the Fifth Amendment.[154]
(c) Is ‘Immunity’ the Correct Characterisation?
Having established that the ch III issue turns on the effect or consequences of the government action, question 3 remains unanswered: does ch III create a constitutional immunity from executive detention? In Chu Kheng Lim, Brennan, Deane and Dawson JJ used the word ‘immunity’ to characterise the operation of ch III, and said that, putting to one side the exceptional cases, ‘the citizens of this country enjoy, at least in times of peace, a constitutional immunity from being imprisoned by Commonwealth authority except pursuant to an order by a court in the exercise of the judicial power of the Commonwealth’.[155] Gaudron J argued for ‘a broad immunity similar to, but not precisely identical with that enunciated by Brennan, Deane and Dawson JJ in [Chu Kheng] Lim’,[156] but she erroneously derived the immunity from s 51 of the Constitution.
Generally speaking, ch III prohibits Parliament from enacting legislation that results in or effects imprisonment without trial. This can be expressed as a lack of power, that is, that ch III deprives Parliament of the power to authorise detention outside the criminal process. Alternatively, it can be said that citizens are not generally or continuously liable or susceptible to imprisonment otherwise than by criminal trial. In a Hohfeldian analysis, these alternatives correspond to saying that Parliament is under a disability (the ‘jural correlative’ of immunity) to impose non-criminal imprisonment, and that citizens are not subject to a liability (the ‘jural opposite’ of immunity) to non-criminal imprisonment.[157] In other words, a citizen is exempt from the power of Parliament to imprison him or her without a criminal trial, and so it makes perfect sense to say that the citizen thereby enjoys an immunity.[158]
The primary objection to this characterisation takes issue with the proposition that the power to authorise detention is exclusively judicial. George Winterton identified the ‘weak point’ of the reasoning of Brennan, Deane and Dawson JJ in Chu Kheng Lim as ‘the assumption that all involuntary detention (except the recognised exceptions) is necessarily punitive’.[159] In Al-Kateb, Hayne J expressly disputed that detention is ‘a step that can never be taken except in exercise of judicial power.’[160] That is because
once the step is taken, as it was in Chu Kheng Lim, of deciding that mandatory detention of unlawful non-citizens can validly be provided without contravention of Ch III, it is plain that unlawful non-citizens have no general immunity from detention otherwise than by judicial process.[161]
McHugh J echoed this position in Re Woolley, commenting that ‘it is going too far to say that, subject to specified exceptions, detention by the Executive is always penal or punitive and can only be achieved as a result of the exercise of judicial power.’[162]
No-one has articulated precisely why involuntary detention (other than the exceptions) is not necessarily punitive. At most, it is alleged that Brennan, Deane and Dawson JJ went ‘too far’ or stumbled on a ‘weak point’.[163] This insistence is curious, especially since tradition regards imprisonment as unlawful unless it falls within a recognised exception. Moreover, the objection is based on a caricature of the judgment of Brennan, Deane and Dawson JJ in Chu Kheng Lim. It overlooks the judgment’s distinction between judicial power and powers incidental thereto: putting to one side the exceptional cases, the involuntary detention of a citizen in custody ‘exists only as an incident of the exclusively judicial function of adjudging and punishing criminal guilt’.[164] The nature of the power to imprison as incidental to judicial power, as well as the acknowledgment of a small number of categories of imprisonment which are not incidental to judicial power, plainly indicate that the judgment did not say that involuntary detention can never be achieved other than as a result of the exercise of judicial power. Rather, Brennan, Deane and Dawson JJ appear to rely on the uncontroversial assumption that ‘imprisonment is historically the dominant social response to behavior defined and spoken of as criminal’, and that it stands ‘at the core of the historically developed notion of criminal punishment’.[165] The effect of imprisonment is as if an authoritative decision on the detainee’s rights had already been made. Subject to certain exceptions, ch III prohibits non-criminal imprisonment because imprisonment is generally correlated with, and attendant on, a conviction of guilt.
(d) Refining the Expression
Drawing from the above, the ch III immunity from non-criminal detention should be expressed in the form favoured by Gummow J: ch III permits the involuntary detention of a citizen or lawful alien in custody only as a consequential step in the adjudication of the criminal guilt of that person for past acts.[166]
This formulation emphasises the necessity of criminal proceedings, as distinct from requiring merely a judicial order. Thus, for example, the ch III immunity would presumptively invalidate detention of a citizen on the basis of that citizen’s status as determined by an administrative tribunal, where such determination required confirmation by a ch III court. It is also subject to certain limited exceptions, which do not form a closed set.[167] The formulation subsumes pre-trial incarceration to secure a defendant’s presence at trial into the general rule, because such detention is consequential to the adjudication of criminal guilt.
B What is the Nature of the Relationship between Valid Detention and the Constitutional Immunity?
To summarise, the legislative power of the Commonwealth conferred on Parliament by s 51 is plenary and, considered on its terms alone, extends to authorising executive detention. However, the power to imprison a person is incidental to the judicial power of the Commonwealth. Therefore, a law that effects or results in involuntary imprisonment will usually be invalidated because ch III requires that the judicial power of the Commonwealth be exercised by courts alone.
The use of the word ‘usually’ in the last sentence indicates an ambiguity in these propositions. The ch III prohibition is not absolute, and the purpose of this section is to implement ch III’s usual invalidation of laws effecting imprisonment. Typically, when a prohibition exists that can be overcome in narrow circumstances, the appropriate jurisprudential tool is the rebuttable presumption. Being a burden-shifting device, it does not provide a substantive answer, but it does furnish a suitable framework of analysis.[168]
In the executive detention context, there are two rebuttable presumptions at play. The first is the presumption of constitutionality, that is, that the burden is on the plaintiff to prove that a law is unconstitutional. The analysis in the foregoing sections shows that this initial presumption is rebutted where a citizen proves that the impugned legislation effects imprisonment. The burden then shifts to the Commonwealth to prove that the legislation does not offend ch III. In this section, a reference to ‘the presumption’ is a reference to the latter presumption.
The substantive question to be confronted, then, is this: how is the presumption to be rebutted? There are two competing answers. First, the case law discloses what can broadly be called a proportionality or balancing approach. It consists in determining whether the detention is proportionate to a legitimate governmental objective. It has been expressed as whether the detention is ‘reasonably necessary’ for securing a legitimate governmental objective,[169] or whether it is ‘reasonably capable of being seen as necessary’ for securing the objective.[170] Another variation on the formula is the ‘appropriate and adapted’ test (although that phrasing was adopted by Gaudron J in respect of the erroneous s 51 immunity).[171] Leslie Zines observed that ‘proportionality would appear to be a useful criterion’ when deciding the punitive question.[172] Stephen McDonald also defended a proportionality approach, validating detention that is ‘reasonably capable of being seen as necessary, appropriate or proportionate to the fulfilment of a legitimate end unconnected with punishment’.[173] The general proportionality approach weighs imprisonment against the legitimate governmental objective in order to determine whether, in pursuit of the objective, imprisonment is excessively burdensome. It balances ‘the relative intensity of the interference with the importance of the aim sought’.[174]
If a court is tasked with balancing the burden of imprisonment against a legitimate governmental objective, it has to ascertain the justification of the imprisonment and the desirability of the governmental objective, attach weights to each, and decide which is weightier.[175] This approach involves a constant reference to background principles and justifications.[176] It is highly fact-specific, requiring balancing in each individual case, and so is not conducive to setting precedent; it places interpretive power in the precedent-following courts. It is well-equipped to deal with situations where a plurality of values is at play and where underlying values are disputed.[177] In the constitutional sphere, it is especially suited to the more fluid norms that require a high degree of adaptability and sensitivity to changing circumstances.
The second methodology, adopted by Brennan, Deane and Dawson JJ in Chu Kheng Lim, relies on categorisation.[178] In contrast to the proportionality approach, categorisation does not permit the court to weigh on a case-by-case basis the relative harms and benefits of imprisonment as compared to the governmental objective. Rather, the question is restricted to whether the detention authorised by the statute falls within a set of preordained categories of permissible detention. Thus, in Chu Kheng Lim, Brennan, Deane and Dawson JJ noted the existing categories[179] and justified the category that was under consideration (the detention of aliens for the purposes of determining their eligibility for entry into Australia).[180]
The categorisation approach prohibits balancing in each case and forbids reference to any background justifications.[181] It places interpretive power in the precedent-setting court, with the lower courts’ discretion limited to deciding whether the facts fall within certain ‘outcome-determinative lines’.[182] All the fundamental interpretive work is done by the superior court at the outset.[183] It is best equipped to deal with core constitutional values which are constant and generally accepted, usually to the point of consensus. This has been expressed as a requirement of close proximity to deeply embedded constitutional values.[184]
The choice of methodology will therefore depend on the strength of the presumption of invalidity of laws authorising non-criminal imprisonment. By this stage, it should be apparent that the presumption is very strong, one of those ‘edifices of extraordinary strength and durability which would take a powerful array of arguments to overcome.’[185] Constitutional doctrine governing non-criminal detention is ‘best understood as reflecting a strong presumption that the criminal process is the preferred means for addressing socially dangerous behavior’.[186] The history of non-criminal imprisonment, tracing the development of the Magna Carta, the Petition of Right and the Habeas Corpus Act 1679, as well as the separation of judicial power embedded in Australia’s Constitution, all point to freedom from arbitrary imprisonment as a core constitutional assumption that can be overcome only in very limited circumstances.[187]
Moreover, the diversity of the subject matters addressed in s 51 bespeaks a difficulty with a proportionality approach. In advocating a process of ‘weighing’ the burden of imprisonment against a legitimate governmental objective, a proportionality approach assumes that the burden and the objective are resolvable along a common axis, that is, that there is ‘a common metric in the weighing process’.[188] This requires adherence to a conception of value commensurability, so that the interests share some common measure.[189] This is unwise, not only because it makes constitutional adjudication dependent on the truth of a controversial theory in moral philosophy, but also because it ‘strip[s] the balancing approach of much of its theoretical motivation.’[190] The proportionality approach is suited when a plurality of various conflicting values are involved, but ‘[i]f all values are reducible to a common metric, the problem that gave rise to the need for a balancing method dissolves.’[191] Thus, it is difficult to weigh the burden of imprisonment against the multitude of purposes that is suggested by the variety of the heads of legislative power in s 51, and the difficulty and uncertainty surrounding such a task is a formidable prospect for day-to-day judicial decision-making. The importance of the prohibition against arbitrary imprisonment dictates that the relevant balancing be done in advance by the High Court, to ensure that its structural and historical priority is duly maintained.
In reducing various interests to a common scale, the proportionality approach appears surprisingly rudimentary.[192] The interests posed are the individual’s against the state’s, but ‘[n]o system of identification, evaluation, and comparison of interests has been developed’.[193] The attendant uncertainty is dangerous because it shows that the proportionality approach is capable of routine manipulation, and that a person may be detained so long as it is good social policy to do so. Leslie Zines, despite expressing support for a proportionality approach, conceded that it was used in Al-Kateb to undermine the ch III immunity itself:
In weighing the object of the power with the object of Chapter III it is clear that the majority gave weight to the former. Rather than the invalidity of executive detention being a prima facie assumption as suggested above, the reverse seems the case.[194]
This lends some support to an argument that a proportionality approach threatens the idea of constitutional supremacy.[195] Under a proportionality approach, a court’s validation of a law ‘might only show that the Court and the legislature used the same calculator’.[196] Under a categorical approach, the judiciary holds the political branches to their constitutional duty to refrain from non-criminal imprisonment.
Accordingly, categorisation is the appropriate methodology to rebut the presumption. The suspicion of arbitrary imprisonment is so deeply ingrained in Australia’s constitutional narrative that it should not be susceptible to being routinely overridden by governmental interests; rather, it should be categorically protected by the High Court. It is not a fluid constitutional norm, but an assumption upon which our very constitutional architecture is built. Chapter III’s prohibition of non-criminal detention cannot evaporate when the government shows that imprisoning a person or class of persons survives an ad hoc balancing of relative social costs and benefits. Indeed, ch III reflects the judgement that the benefits of restrictions on the executive outweigh the costs.[197] It is simply too important to leave to the variations and fluctuations that are sure to follow from a proportionality approach.[198]
C What is the Typology of the So-Called ‘Exceptional Cases’?
In Kruger, Gaudron J argued that the recognised exceptions do not fall within precise and confined categories.[199] Gleeson CJ agreed in Vasiljkovic v Commonwealth.[200] This section is devoted to demonstrating the contrary: the recognised exceptions to the ch III immunity are sufficiently precise and confined so as to justify the ‘categorical’ epithet. The exceptions identified by Brennan, Deane and Dawson JJ in Chu Kheng Lim will be outlined, commencing with the aliens exception itself.
1 Aliens
The existence of this exception was justified in Chu Kheng Lim. To summarise, Brennan, Deane and Dawson JJ proceeded in two steps. First, they appealed to judgments of the Privy Council, House of Lords and High Court, as well as international law and prominent text-writers, to conclude that the power to detain an alien for the purposes of deportation is executive in nature.[201] This established the category. Second, they held that the legislation under consideration was valid because the detention it required and authorised was ‘limited to what is reasonably capable of being seen as necessary for the purposes of deportation or necessary to enable an application for an entry permit to be made and considered’.[202] This step placed the particular legislation within the established category. Vital to the law’s validity was that the detention was seen as ultimately voluntary (a detainee could request removal from Australia, so a failure to request removal was an election to remain imprisoned in Australia).[203] There were also other limiting provisions recognised by Brennan, Deane and Dawson JJ: a statutory upper limit on the length of detention (273 days), a two-month deadline to make an entry application (with removal from Australia upon expiration of the deadline), and removal as soon as practicable if an entry application had been refused (and all avenues of appeal exhausted).[204] These elements sufficiently circumscribed the detention so that it could be reasonably described as necessary for the limited purposes of entry or removal.
2 Quarantine
Section 51(ix) of the Constitution confers legislative power on the Commonwealth Parliament with respect to quarantine, and the exercise of that power resulted in the Quarantine Act 1908 (Cth) (‘Quarantine Act’). The Quarantine Act grants broad powers to the executive to take quarantine measures in respect of persons. Section 4(1)(a) sets out the scope of quarantine measures, which includes examination, exclusion, detention, segregation, isolation, treatment and regulation of human beings; s 4(1)(b) restricts the object of such measures to the prevention or control of the introduction, establishment or spread of significantly damaging diseases or pests. Where the Governor-General makes a proclamation in the event of an epidemic under s 2B(1), the Minister may ‘give such directions and take such action as he or she thinks necessary to control and eradicate the epidemic, or to remove the danger of the epidemic, by quarantine measures or measures incidental to quarantine’.[205] Alternatively, where the Minister is of the opinion that ‘an emergency has arisen that requires the taking of action not otherwise authorised under this Act’, he or she
may take such quarantine measures, or measures incidental to quarantine, and give such directions, as he or she thinks necessary or desirable for the diagnosis, for the prevention or control of the introduction, establishment or spread, for the eradication, or for the treatment, of any disease or pest.[206]
Section 12A(2) makes it an offence to fail to comply with any direction given under s 12A(1).
Thus, in the case of an epidemic or emergency — the existence of which depends on the opinion of the executive — the Quarantine Act’s posture towards detention is very assertive. Under ordinary circumstances, s 18(1) lists the persons who are ‘subject to quarantine’, including: every person who is infected with a quarantinable disease or pest; every person who a quarantine officer reasonably suspects is infected with a quarantinable disease or pest; every person who has been in contact with or exposed to infection from any person subject to quarantine; and every person who is ordered into quarantine. Also subject to quarantine is every person who enters Australia unlawfully.[207] A person who is ‘subject to quarantine’ may be apprehended without warrant by any police officer.[208] Section 35(1) provides that a person may be ordered into quarantine ‘whether subject to quarantine or not’, so long as the quarantine officer believes that the person is or is likely to be infected with a quarantinable disease or pest. The consequences that follow are set out in s 45(1):
(1) All persons ordered into quarantine shall perform quarantine, and for that purpose may:
(a) be detained on board the vessel or installation;
(b) be detained upon the premises upon which they are found;
(c) be removed to and detained in a quarantine station; or
(d) be removed to and detained in any suitable place or building approved by a quarantine officer (which place or building shall, for the purposes of this Act, be deemed to be a quarantine station);
until released in accordance with this Act or the regulations; and while so detained shall be subject to the regulations regulating the performance of quarantine and the government of quarantine stations.
The first legislative exercise of a quarantine power occurred in the 18th century: Quarantine Act 1710.[209] It prevented persons coming to shore who were on board ships or vessels that originated in places infected with the plague (s I); where a person left a quarantined vessel and came to shore, it authorised ‘Persons appointed to see the Quarantine duly performed to compel and in case of Resistance by Force and Violence to compel such Person or Persons to return on board such Ships and there to remain during the Time of Quarantine’ (s III); and, once the quarantine had been duly performed, every person and persons on the vessel ‘shall be liable to no further Restraint or Detention’ (s VII). Blackstone classified offences associated with quarantine as offences ‘against the public health of the nation; a concern of the highest importance’[210] and documented the revision of the quarantine legislation during the 18th century.[211]
Even prior to the Quarantine Act 1710, there existed an independent quarantine power in the executive. So much is acknowledged in the statute of 1710, the object of which was ‘to oblige Ships coming from Places infected, more effectually to perform their Quarentine [sic]’.[212] The statute was required to strengthen a prior exercise of the royal prerogative, which consisted in ‘several Orders’ made by the Queen in Council, as well as a ‘Royal Proclamation’ requiring ‘a Quarentine [sic] to be performed by all Ships and Persons coming from Places infected’.[213] In 1684, the prerogative of the Crown to order quarantine was recognised by Lord Jeffreys CJ:
both by the law of nations, and by the common law of England, the regulation, restraint and government of foreign trade and commerce, is reckoned ‘inter Jura Regalia,’ ie is in the power of the king: and it is his undoubted prerogative ...[214]
The Lord Chief Justice accepted that ‘the king might prohibit his subjects to go or trade beyond the seas in cases of wars or plagues’.[215] As Cox noted, ‘[w]ith respect to the latter prerogative, it may be observed that the laws of Quarantine are now regulated by statute.’[216]
Acting in concert with the recognition of the prerogative of the Crown to order quarantine is the consistent acknowledgment by courts that quarantine was a legitimate form of detention precluding the issue of habeas corpus. In 1758, Wilmot J (answering questions posed by the House of Lords) said that were the writ to be issued ‘of course’ (that is, without the need of showing cause), then:
it would have been suffering this great remedial mandatory writ to have been used as an instrument of vexation and oppression; it would have become a weapon in the hands of madmen, and of dissolute, profligate and licentious people, to harass and disturb persons acting under the powers which the law had given them. — One most frightful instance occurs: the case of a crew performing quarantine. — If this writ were to issue of course, it might bring back pestilence and death along with it.[217]
In 1818, Lord Eldon LC cited this with approval, noting that if it were otherwise, ‘the most serious mischiefs might ensue; supposing, for example, that they were confined on board of ship subject to quarantine.’[218]
Therefore, historically the power to order detention in the performance of quarantine was seen as a prerogative of the Crown, until legislative enactment displaced exercise of the prerogative in 1710; and so it falls within the legislative power to authorise quarantine detention. However, history also shows abuses of the quarantine power,[219] and so detention authorised in the performance of quarantine must be strictly circumscribed. By the beginning of the 19th century, the general English practice of quarantining every inbound vessel had been discarded in favour of a policy to detain only where there was communicable disease aboard during the voyage, or where such existed on arrival.[220] Thus, quarantine legislation will be valid if the detention it requires and authorises is reasonably capable of being seen as necessary for preventing and containing the spread of a specific and currently threatening infectious disease.
This is not the place to undertake a sustained analysis of the validity of the Quarantine Act 1908 (Cth), but a few observations are in order. Diseases are designated by proclamation as ‘quarantinable diseases’, and are relatively few in number,[221] suggesting that the legislation is directed to those specific diseases only. The width of the powers conferred by the Quarantine Act, however, is cause for concern. In particular, the only formal requirement of an order into quarantine is that it must be in writing,[222] although it should be noted that a person ordered into quarantine may seek an independent medical assessment[223] and must be informed of this right by the quarantine officer.[224] In order to strictly adhere to the required purpose of preventing and containing the spread of a specific and currently threatening infectious disease, the Quarantine Act should require the order to specify the quarantine location, the date and time when the quarantine will begin and end, and a statement setting out scientific principles and evidence of exposure or infection that form the basis of the order.[225] Finally, the Quarantine Act contains no temporal hint in respect of release: s 46 provides that ‘[w]hen quarantine has been performed’ a person ‘shall forthwith be released from quarantine’; and s 35B provides that a person ordered into quarantine by reason of an infectious disease ‘must be released from quarantine once the person receives a certificate of release from a quarantine officer’. No objective statutory guidance is given as to when quarantine has been performed or when an officer must issue a certificate of release, and there is no entitlement to periodic review. A person’s continued detention is entirely at the discretion of the executive.[226]
3 Mental Illness
The civil commitment of the mentally ill is governed by the states.[227] In the states, where the separation of judicial power is not constitutionally entrenched, the limitation on state legislative power that derives from ch III of the federal Constitution is correspondingly much weaker. Accordingly, the implementation of a regime of involuntary detention and treatment in mental health facilities is largely left to the state political process. That is not to say that the federal government could not, conformably with ch III, institute a regime of civil commitment of the mentally ill (assuming, of course, the existence of the necessary legislative power).[228] The historical legitimacy of the power to imprison in such circumstances lies in the government’s capacity as parens patriae, as well as the consistent historical recognition that mental illness precludes the issue of habeas corpus.
4 Contempt of Parliament
Section 49 of the Constitution provides:
The powers, privileges, and immunities of the Senate and of the House of Representatives, and of the members and the committees of each House, shall be such as are declared by the Parliament, and until declared shall be those of the Commons House of Parliament of the United Kingdom, and of its members and committees, at the establishment of the Commonwealth.
The legislative power conferred by s 49 was not exercised until the Parliamentary Privileges Act 1987 (Cth). Section 5 continues the powers, privileges and immunities of each House ‘as in force under section 49 of the Constitution immediately before the commencement of this Act’. Section 7(1) declares a power of each House to ‘impose upon a person a penalty of imprisonment for a period not exceeding 6 months for an offence against that House determined by that House to have been committed by that person’. Section 4 defines ‘an offence against a House’ as ‘an improper interference with the free exercise by a House or committee of its authority or functions, or with the free performance by a member of the member’s duties as a member’.
In 1955, the Speaker of the House of Representatives issued warrants ordering the imprisonment of Fitzpatrick and Browne for three months on account of them ‘being guilty of a serious breach of privilege’.[229] The full High Court refused the applications of Fitzpatrick and Browne for writs of habeas corpus. Dixon CJ delivered the judgment of the court, holding that in the UK ‘it is for the courts to judge of the existence ... of a privilege, but ... for the House to judge of the occasion and manner of its exercise’;[230] and one of the most important privileges of the House of Commons is the privilege of committing for contempt by warrant stating generally that a contempt had taken place.[231] As a warrant ‘would clearly be sufficient if it had been issued by the Speaker of the House of Commons’, so it was sufficient by the language of s 49.[232]
The warrants were challenged on a separation of powers ground, namely that
the power exercised by resolving upon the imprisonment of two men and issuing a warrant to carry it into effect belonged to the judicial power and ought therefore not to be conceded under the words of s 49 to either House of the Parliament.[233]
To counter this, Dixon CJ merely observed that ‘in unequivocal terms the powers of the House of Commons have been bestowed upon the House of Representatives’, and ‘a general view of the Constitution and the separation of powers is not a sufficient reason for giving to these words, which appear to us to be so clear, a restrictive or secondary meaning which they do not properly bear’.[234] Dixon CJ also noted the historical and analytical approaches to judicial power, and the potential of those conceptions to diverge:
throughout the course of English history there has been a tendency to regard [the House of Commons’] powers as not strictly judicial but as belonging to the legislature, rather as something essential or, at any rate, proper for its protection. This is not the occasion to discuss the historical grounds upon which these powers and privileges attached to the House of Commons. It is sufficient to say that they were regarded by many authorities as proper incidents of the legislative function, notwithstanding the fact that considered more theore-tically — perhaps one might even say, scientifically — they belong to the judicial sphere.[235]
Had s 49 been absent from the Constitution, the challenge to the House’s power to commit for contempt should have succeeded. The power to adjudicate upon a complaint of contempt is not necessary for the exercise of Parliament’s functions and can be performed with the assistance of the judiciary.[236] The only reason that the House of Commons enjoys the power to punish for contempt is ‘by virtue of ancient usage and prescription; the lex et consuetudo Parliamenti’.[237] It follows that, without s 49, the Commonwealth Parliament may not have had any historical analogy because ‘the Legislative Assemblies in the British Colonies have, in the absence of express grant, no power to adjudicate upon, or punish for, contempts committed beyond their walls’.[238] With no relevant custom to fall back on, and with unanimous analytical recognition that the power is judicial, ch III would require punishment for contempt to be meted out by the courts alone.[239]
5 Courts-Martial and Service Tribunals
Courts-martial or service tribunals routinely punish members of the defence forces for breaches of military discipline, issuing declarations of guilt and sentences of imprisonment. Standing outside the requirements of ch III and yet purporting to convict individuals of criminal guilt, these tribunals have vexed the High Court on several occasions because they appear to fly in the face of Chu Kheng Lim’s classification of the adjudgment and punishment of criminal guilt as exclusively judicial. All of the challenges brought to the High Court have a common posture: a member of the defence forces has been charged with or convicted of an offence (or offences) before a court-martial or service tribunal and seeks an order in the nature of habeas corpus or prohibition and a declaration that the legislation conferring jurisdiction on the court-martial or service tribunal is invalid.
Two themes have emerged from the High Court’s rejection of the separation of powers objection. The first is that a military and naval justice system is necessary and essential to the defence forces. In R v Bevan; Ex parte Elias and Gordon (‘R v Bevan’), Starke J held that s 51(vi) of the Constitution together with ss 69 and 51(xxxix) are ‘legislative provisions special and peculiar to’ the armed forces ‘in the way of discipline and otherwise, and indeed the Court should incline towards a construction that is necessary, not only from a practical, but also from an administrative point of view’.[240] In R v Cox; Ex parte Smith, Dixon J held ‘[t]o ensure that discipline is just, tribunals acting judicially are essential to the organization of an army or navy or air force’.[241] However, such tribunals ‘do not form part of the judicial system administering the law of the land.’[242] In Re Tracey; Ex parte Ryan (‘Re Tracey’), Mason CJ, Wilson and Dawson JJ noted that even though s 51(vi) does not expressly provide for enacting rules to regulate the armed forces, ‘so much is necessarily comprehended by the first part of s 51(vi) for the reason that the naval and military defence of the Commonwealth demands the provision of a disciplined force or forces’.[243] Notwithstanding that s 51(vi) is expressly subject to ch III, ‘the proper organization of a defence force requires a system of discipline which is administered judicially, not as part of the judicature erected under Ch III, but as part of the organization of the force itself’.[244] Similarly, Brennan and Toohey JJ held that the creation of a military jurisdiction to discipline army personnel is an ‘essential concomitant’ of the power to establish the permanent armed forces of the Commonwealth.[245] In White v Director of Military Prosecutions (‘White’), Gummow, Hayne and Crennan JJ recognised that military and naval justice systems are directed to the maintenance of ‘the defining characteristic of armed forces as disciplined forces organised hierarchically.’[246]
The second theme is the High Court’s heavy reliance on history for the validation of military tribunals outside the requirements of ch III. The significance of history is that it ‘forms part of the context relevant to the construction of the Constitution and, in particular, to an understanding of the relationship between s 51(vi) and Ch III’.[247] It was not until Re Tracey that the historical ground was thoroughly traversed. Mason CJ, Wilson and Dawson JJ acknowledged that ‘the pre-1900 legislative history of the power of courts-martial to try members of the forces for civil offences is relevant to the consideration of the scope of s 51(vi).’[248] Brennan and Toohey JJ extensively reviewed the history of courts-martial[249] and considered that, by the time of federation, the statutory encapsulation of the ‘scope of naval and military law and of the special jurisdictions to enforce that law ... reflected the resolution of major constitutional controversies.’[250] They concluded that ‘the creation of a military jurisdiction to discipline army personnel has been regarded since the first Mutiny Act to be the essential concomitant of the raising and keeping of a standing army’.[251] Thus, history and established practice have placed the execution of military justice by military authorities outside ch III, and ‘[h]istory and necessity combine to show that courts-martial and other service tribunals, though judicial in nature and though erected in modern times by statute, stand outside the requirements of Ch III’.[252] In White, Gummow, Hayne and Crennan JJ held that
[t]o attribute to the presence in the Constitution of Ch III a rejection of service tribunals ... would be to prefer ‘abstract reasoning alone’ ... to an appreciation of the content of ‘the judicial power of the Commonwealth’ which must have been universally understood in 1900.[253]
These two themes of necessity and history have legitimised courts-martial despite the strong similarity between the power exercised by courts-martial and judicial power. In R v Bevan, Starke J expressed this similarity by saying that courts-martial exercise ‘judicial power’ but not ‘the judicial power of the Commonwealth’, the phrase used in ch III.[254] Mason CJ, Wilson and Dawson JJ adopted this distinction in Re Tracey and said that ‘[t]here has never been any real dispute’ over whether a court-martial exercises judicial power; rather, ‘[t]he question is whether it is exercising the judicial power of the Commonwealth under Ch III of the Constitution’.[255] They also thought that ‘no relevant distinction can, in our view, be drawn between the power exercised by a service tribunal and the judicial power exercised by a court.’[256] Brennan and Toohey JJ relied on English authority holding that courts-martial exercise judicial power, but ‘the imposition of punishments by service authorities as for the commission of criminal offences in order to maintain or enforce service discipline has never been regarded as an exercise of the judicial power of the Commonwealth.’[257]
The most recent case, Lane v Morrison,[258] both affirmed the High Court’s heavy reliance on history and cast doubt upon its previous assurance that the power exercised by a court-martial is indistinguishable from the judicial power exercised by a court. In 2006, in response to human rights concerns with the existing system of service tribunals, Parliament created the Australian Military Court (‘AMC’) with a view to strengthening the independence of the defence force’s disciplinary decision-making. The AMC was to perform the disciplinary functions previously entrusted to service tribunals, but, unlike service tribunals, the AMC’s decisions were not subject to confirmation or review within the defence force’s chain of command. This was to ensure that penalties were imposed upon members of the defence forces by an independent and impartial tribunal. It was this distinguishing feature that took the AMC out of the ‘historical stream’[259] to which courts-martial and service tribunals belonged, and upon which the High Court struck down the legislation that conferred jurisdiction.
French CJ and Gummow J took a historical approach, surveying the legislative history back to 1866 and holding that ‘the AMC was designed to make a break’ with the traditional location of courts-martial within the chain of command.[260] This break with tradition made the legislation ‘vulnerable to the attack now successfully made upon the validity of the AMC.’[261] On the other hand, Hayne, Heydon, Crennan, Kiefel and Bell JJ preferred to take a largely analytical approach, holding that the feature of courts-martial setting them apart from the exercise of the judicial power of the Commonwealth was that ‘[t]he decisions of courts-martial were not “definitive” of guilt; the punishments awarded by courts-martial were subject to confirmation or review.’[262] This robbed the power exercised by courts-martial of finality and authoritativeness, which are crucial to an analytical perspective of judicial power:
Courts-martial were convened only by order from within the chain of command; conclusions of guilt and determinations of punishment were subject to review or confirmation within that chain of command. A court-martial did not make a binding and authoritative decision of guilt or determination of punishment. A court-martial did not enforce its decisions. Enforcement of any decision, other than acquittal of the accused, depended upon the outcome of review of the decision within the chain of command.[263]
However,
a central purpose of the creation of the AMC was to have the new body make binding and authoritative decisions of guilt and determinations about punishment which, without further intervention from within the chain of command, would be enforced.[264]
It followed that the AMC was to exercise the judicial power of the Commonwealth and yet was not constituted in accordance with ch III.[265] This was sufficient to prove its invalidity.
The central proposition of Lane v Morrison is that a service tribunal will not exercise the judicial power of the Commonwealth, and will therefore be supported by s 51(vi), only if it is a part of the chain of command, that is, if it is a part of the very entity whose individual members it purports to discipline. Since a place in the chain of command is necessary to the service tribunal’s validity, it follows that s 51(vi) does not support a ‘parallel’ court system located outside both the defence force and the federal judiciary contemplated by ch III. In the words of French CJ and Gummow J, s 51(vi) does not support the existence of a system of ‘legislative courts’ akin to the art I tribunals in the US.[266] Hayne, Heydon, Crennan, Kiefel and Bell JJ, in offering a distinguishing analytical feature of the power exercised by courts-martial, placed Lane v Morrison in tension with the earlier case Re Tracey, where a majority of the High Court failed to discern any theoretical difference between the respective powers of service tribunals and courts.[267]
A related aspect of the tension between Re Tracey and Lane v Morrison is the latter case’s disavowal of the distinction between ‘judicial power’ and ‘the judicial power of the Commonwealth’, first suggested by Starke J in R v Bevan and later adopted by a majority in Re Tracey. In Lane v Morrison, French CJ and Gummow J held that the distinction ought to be discarded because ‘the only judicial power which the Constitution recognises is that exercised by the branch of government identified in Ch III’;[268] and Hayne, Heydon, Crennan, Kiefel and Bell JJ stated that the distinction distracted from, and did not assist in answering, the relevant constitutional question.[269] Coupled with Kirby J’s strident criticism of the distinction in White,[270] it can now safely be regarded as officially discarded.
D Creation of New Categories
Chapter III permits the involuntary detention of a person in custody only as a consequential step in the adjudication of the criminal guilt of that person for past acts.[271] The fundamental purpose of the separation of judicial power is to prevent arbitrary imprisonment at the will of the political branches.[272] Scalia J recognised in Hamdi:
To be sure, certain types of permissible noncriminal detention — that is, those not dependent upon the contention that the citizen had committed a criminal act — did not require the protections of criminal procedure. However, these fell into a limited number of well-recognized exceptions — civil commitment of the mentally ill, for example, and temporary detention in quarantine of the infectious.[273]
If the legislature attempts to authorise, outside the ordinary procedures of the criminal law, the involuntary detention of a person, then a rebuttable presumption of unconstitutional detention arises. This presumption can be overcome by proving that the detention falls within one of the traditional categories: detention for the purposes of determining a person’s eligibility to enter Australia; detention for the purposes of preventing and containing the spread of a specific and currently threatening infectious disease; civil commitment of the mentally ill; detention for contempt of Parliament; and detention resulting from conviction by a court-martial or service tribunal.
Freedom from arbitrary detention is one of the most fundamental aspects of the system of justice in Australia, and for this reason the exceptions to the ch III prohibition are few and strictly defined. Courts must be very slow to recognise new categories of cases in which non-criminal detention is permitted. However, categories can rot with age. A category may no longer align with the underlying constitutional values that it was originally designed to protect.[274] Courts must be sensitive to the possibility of a novel category of non-criminal detention arising. This section will consolidate some themes of the existing categories and develop a response to the possibility that a new exceptional category may be identified. Broadly speaking, a new exceptional category will be admitted if history and necessity combine to show that a specific kind of detention may be authorised other than by the exercise of criminal jurisdiction.[275]
It may be that there are some categories of non-criminal detention that enjoy the imprimatur of history but have not yet been specifically identified or discussed in the case law. For a historical analysis to weigh in favour of a ‘new’ category, it must indicate that the power to order the particular species of detention has traditionally been viewed as legitimately non-judicial. This will be done if history discloses that detention has been perceived as ancillary or incidental to legislative or executive power. To take the logical inverse of Kitto J’s statement in R v Davison, where the action to be taken had come by 1900 to be so consistently regarded as peculiarly appropriate for legislative or executive performance, then the power to take that action, even though judicial in appearance, will be within the concept of executive or legislative power conferred by chs I and II of the Constitution.[276]
The aliens, quarantine and military justice categories all exhibit this historical or traditional association with executive or legislative power. In Chu Kheng Lim, Brennan, Deane and Dawson JJ were at pains to identify the consistent recognition that the power to detain an alien for the purposes of entry or deportation is incidental to, and takes its character from, the executive powers to exclude, admit and deport.[277] The variety and status of the sources that Brennan, Deane and Dawson JJ referred to are revealing. Thus, the judgment of ‘a strong Judicial Committee of the Privy Council’, which in turn referred to Vattel, establishes the traditional position of international law that the executive power to exclude or expel an alien is an incident of territorial sovereignty.[278] Brennan, Deane and Dawson JJ cite Blackstone, who said, ‘it is left in the power of all states, to take such measures about the admission of strangers, as they think convenient’;[279] and Chitty, who pointed out, ‘the Crown, even at common law, and by the law of nations ... possesses a right to order [aliens] out of the country, or prevent them from coming into it, whenever his Majesty thinks proper.’[280] Similarly, as noted above, the power of quarantine before 1710 was seen as exclusively executive in character, but was subsequently regulated by statute. Finally, the ability of the legislature to institute military tribunals to discipline members of the armed forces dates to 1689 (the first Mutiny Act passed in the aftermath of the Glorious Revolution).[281]
Two points should be made about the required historical analysis. First, the historical practice must be continuous, or at least, it must not have fallen into disuse or been consciously discarded. Thus, it is not possible to resurrect imprisonment for the non-payment of a contractual debt, because in 1869 Parliament provided that ‘no person shall ... be arrested or imprisoned for making default in payment of a sum of money’.[282] Second, it is not sufficient to prove the mere historical existence of the species of non-criminal imprisonment. The historical material must disclose not only the bare fact of exercise of the power of detention, but also the historical recognition of the necessity of the imprisonment for the exercise of the executive or legislative power. For example, there is ample historical evidence for the traditional view that a system of military justice is necessary for the maintenance of order and discipline in the armed forces. Lord Loughborough said in 1792:
The army being established by the authority of the Legislature, it is an indispensable requisite of that establishment that there should be order and discipline kept up in it, and that the persons who compose the army, for all offences in their military capacity, should be subject to a trial by their officers. That has induced the absolute necessity of a mutiny act accompanying the army ... It is one object of that act to provide for the army; but there is a much greater cause for the existence of a mutiny act, and that is, the preservation of the peace and safety of the kingdom: for there is nothing so dangerous to the civil establishment of a state, as a licentious and undisciplined army; and every country which has a standing army in it, is guarded and protected by a mutiny act.[283]
As noted above, the High Court has consistently cited the necessity of a military justice system existing outside the ch III judiciary.[284]
In some instances there may be historical precedent acknowledging the fact that a traditional power of non-criminal imprisonment is not necessary for the executive or legislative function. In Kielley v Carson,[285] notwithstanding the House of Commons’ power to punish for contempt, Parke B denied the same power to the House of Assembly of Newfoundland. The Commons’ power to punish for contempt is a peculiar privilege that exists only ‘by virtue of ancient usage and prescription; the lex et consuetudo Parliamenti’[286] and is ‘by no means essentially necessary for the exercise of its functions by a local Legislature’.[287] Of course, explicit constitutional authorisation trumps any contrary tradition.[288]
This leaves the situation where a proposed category of detention is completely novel in that it enjoys no substantive historical support. In such a case the new category can be admitted in one of two ways.[289] First, it will be admitted if it is strictly necessary to achieve a vital and urgent governmental objective. The question is whether that objective can be achieved in any way without resorting to non-criminal imprisonment. Where, for example, a state of total war imperils the Constitution itself, it is conceivable that s 51(vi) can expand to contemplate involuntary detention. In such a case, ch III would not vanish, but its protections may weaken. Otherwise, where it is possible that the power to imprison can be conveniently achieved via the criminal law, then that is what ch III requires. An analogous point was made by Parke B in Kielley v Carson, saying that the functions of a legislature can be ‘well performed’ without the ‘extraordinary power’[290] of ‘punishing any one for past misconduct as a contempt of its authority’.[291] That power is best left to the ‘ordinary tribunals’.[292]
Second, a new category might be admitted if it is very closely analogous to an existing category and shares the same rationale. If this process of reasoning is to succeed, the analogy must be very close indeed. Consider, for example, the claim that the continued indefinite detention, after expiration of sentence, of persons convicted with sexual offences is analogous to the civil commitment of the mentally ill. Despite an apparent coincidence of rationales (protection of the community), the analogy must be tightened if it is to succeed. The recognised exception of civil commitment of the mentally ill exists because the criminal law is unintelligible when applied to those who are dangerous and yet lack the capacity to form the requisite mental states. In the case of a dangerous sex offender whose release is imminent, if the notion of a ‘guilty mind’ can be given content, then it is impermissible to detain beyond sentence. A conviction for a criminal offence acknowledges legal culpability and assumes that the offender is not mentally ill. In order to invent novel categories by analogical reasoning, the analogy must be strong. The strin-gency of the methods for admitting novel categories of detention is consistent with the fundamental status of the freedom from arbitrary detention.
It is hoped that the above offers a principled approach to the circum-stances of permissible non-criminal detention. Christos Mantziaris has argued:
the interplay of historical and analytical approaches defeats the expectation of consistency. New forms of decision-making might qualify as analogous or ‘ancillary’ to historical examples of judicial power, but fail the analytical test. Older forms of decision-making might fall within the historical concept of judicial power, but be excluded from the judicial power of the Commonwealth because they fall into a recognised exception (for example, military justice) ...[293]
He contends that this invites an ‘inconsistent use of the historical and analytical approaches to the concept of “judicial power”’.[294] No doubt the interplay of the historical and analytical approaches to ‘judicial power’ is complicated, but it is necessarily so. For one thing, the historical and analytical approaches are not mutually exclusive. The concept of ‘judicial power’ is a human invention, and so any ‘purely’ analytical approach must have a historical ancestor. But further, the expectation of consistency is preserved because the basic thesis of this article can be succinctly stated[295] and is closely moulded to Australia’s core constitutional commitments.
III CONTEMPORARY CHALLENGES: INTERIM CONTROL ORDERS
The prevailing political climate inevitably strains existing categories of permissible non-criminal detention. The government’s response to a person or class of persons it perceives as threatening will invariably include an assertion of power to detain outside the criminal process. Terrorism is an exemplary challenge for modern constitutional democracies committed to the separation of powers. In the 19th century, Dicey posed a hypothetical to illustrate that the authority of judges ‘cuts down the discretionary powers of the Crown’ and ‘prevents the English government from meeting public danger by measures of precaution which would as a matter of course be taken by the executive of any continental country’:
Suppose, for example, that a body of foreign anarchists come to England and are thought by the police on strong grounds of suspicion to be engaged in a plot, say for blowing up the Houses of Parliament. Suppose also that the existence of the conspiracy does not admit of absolute proof. An English Minister, if he is not prepared to put the conspirators on their trial, has no means of arresting them, or of expelling them from the country. In case of arrest or imprisonment they would at once be brought before the High Court on a writ of habeas corpus, and unless some specific legal ground for their detention could be shown they would be forthwith set at liberty.[296]
Things have changed since Dicey wrote those words in 1885. Few would now regard that as a permissible constitutional position. Almost every academic article dealing with non-criminal detention in the last decade commences with a narrative of the terrible events of September 11, 2001, and the gathering consensus is that nations like the UK, US and Australia possess the constitutional tools necessary to thwart nascent terrorist plots where the available evidence will not and may never secure a criminal conviction. Adhering to Dicey’s absolute position, according to one scholar, ‘is not a realistic response’.[297]
In Thomas v Mowbray,[298] the High Court upheld the validity of div 104 of the Criminal Code Act 1995 (Cth) sch (‘Criminal Code’), which authorised the Federal Court of Australia, the Family Court of Australia and the Federal Magistrates Court to make ‘interim control orders’ in relation to a person if the court was satisfied on the balance of probabilities that an order would ‘substantially assist in preventing a terrorist act’[299] and was ‘reasonably necessary, and reasonably appropriate and adapted, for the purpose of protecting the public from a terrorist act’.[300] A ‘terrorist act’ was defined as an action or threat of action of specified kinds which is done with the intention of ‘advancing a political, religious or ideological cause’ and with the intention of ‘coercing, or influencing by intimidation, the government of the Commonwealth or a State, Territory or foreign country’ or of ‘intimidating the public or a section of the public’.[301]
Section 104.5(3) catalogued the various ‘obligations, prohibitions and restrictions’ that the court may impose by a control order, including: being at specified areas or places; leaving Australia; communicating or associating with specified individuals; accessing or using telecommunication or other technology; possessing or using specified articles or substances; and carrying out specified activities (including in respect of the person’s work or occupation). Section 104.5(3) also permitted the imposition of requirements that the person: remain at specified premises between specified times each day or on specified days; wear a tracking device; report to specified persons at specified times and places; allow himself or herself to be photographed; allow fingerprints to be taken; and participate in specified counselling or education.
Although placed in the Criminal Code, control orders are preventive, and may be imposed without contemplation of a criminal trial at all (indeed, the standard of proof adopted in div 104 is the civil standard).[302] Among the various ch III issues raised by div 104, a majority of the High Court summarily rejected the submission that the power to impose restrictions on liberty by a control order exists only as an incident of the declaration and punishment of criminal guilt. That is because control orders merely impose a restraint on liberty rather than a wholesale deprivation. Gleeson CJ held that ‘[i]t may be accepted that control orders may involve substantial deprivation of liberty, but we are not here concerned with detention in custody’.[303] The restraints that div 104 enabled were more appositely compared to apprehended violence orders. The Chief Justice thought that it is ‘too broad’ to say that restraints on liberty (as distinct from full-blown detention in custody) exist only as an incident of adjudging and punishing criminal guilt.[304] Similarly, Gummow and Crennan JJ refused to transmute the ch III prohibition into the control order context: ‘[d]etention in the custody of the State differs significantly in degree and quality from what may be entailed by observance of an interim control order’.[305] Callinan J agreed generally with Gummow and Crennan JJ,[306] adding that div 104 ‘makes and implies the usual indicia of the exercise of judicial power’.[307] Heydon J agreed with Gleeson CJ, Gummow and Crennan JJ, and Callinan J.[308]
Only Kirby J (in dissent) was of the opinion that div 104 had the potential to effect ‘virtual house arrest’.[309] Christos Mantziaris observed that it is possible to ‘imagine a set of control orders with prohibitions and restrictions so onerous as to amount to constructive detention’.[310] It is hard to disagree with this view, especially given that s 104.5(3)(c) expressly authorises ‘a requirement that the person remain at specified premises ... on specified days’. The Constitution is concerned with substance and not mere form,[311] and Blackstone counts house arrest as imprisonment;[312] the phrase ‘detention in custody by the State’ aptly includes a command from the government to remain in one’s house for some days. The phrase ‘in custody’ has been given a liberal interpretation in the US, where persons not within the physical control of the government have been held to be ‘in custody’ for the purposes of habeas corpus.[313]
Assuming, then, that div 104 authorises imprisonment outside the criminal process, a rebuttable presumption of invalidity arises. The majority in Thomas v Mowbray offered some historical analogies to suggest that control orders are not entirely novel to the judicial function: binding over orders, the writ of supplicavit, apprehended violence orders, bail and sentencing.[314] However, the majority also conceded the limitations of the analogies that they tendered. Gleeson CJ remarked that they were ‘not exact’,[315] and Gummow and Crennan JJ said that the historical considerations did not ‘furnish any immediate analogy to the modern legislative regime ... now under challenge’.[316] Kirby J’s analysis is most persuasive:
Each of the propounded analogies is distinguishable from the orders for which div 104 provides. Each is decided on the basis of the past conduct of the person to be subject to the order and each is directed against what that particular person might do in the future. They are not directed, as orders under div 104 may be, at what third parties not subject to the order might do.
In the case of bail proceedings, the court may consider the protection and welfare of the community. However, it will only do so having regard to the nature and seriousness of the offence with which the accused is charged with having committed and any other offences that may be taken into account. The court may only consider possible future offences in defined circumstances. The protection of the community is only one of a great number of otherwise strict and ascertainable criteria to be considered in bail proceedings. It is not the only factor.[317]
It may be added that the analogy to bail and sentencing is plainly inadequate because bail is imposed as a consequential step in the adjudication of criminal guilt, and ‘at the point of sentencing, liability has already been determined as a result of a judicial process which has involved the adjudication of criminal guilt.’[318]
Since the control orders envisaged by div 104 enjoy no substantive historical support, they will be constitutionally valid only if they are strictly necessary or if they are very closely analogous to an existing category of permitted non-criminal detention. As regards strict necessity, the importance of protecting the public from a terrorist act is undeniable. But given the stringency of the obligations, prohibitions and restrictions envisioned by
s 104.5(3), it is not self-evident that house arrest is really necessary to prevent a terrorist act. Presumably, detention is truly necessary to prevent a terrorist attack only where an attack is imminent or near certain, that is, where a person is already criminally liable for inchoate offences such as attempt or as an accomplice, conspirator or principal. In such a case, the criminal law offers a sufficient mechanism by which a suspect ought to be detained. This lends support to the argument that div 104 was inserted in pt 5.3 of the Criminal Code, under the heading ‘Terrorism’, in order to circumvent the criminal process altogether. As Lynch and Reilly argue, if the court is satisfied on the balance of probabilities that making a control order would substantially assist in preventing a terrorist act, then surely the same evidence would ground criminal liability under the ‘very wide preparatory offences’ in divs 101–3.[319] An argument of strict necessity therefore fails.
The final option is to find an analogy in the pre-existing categories. The most promising candidate for analogy is the civil commitment of the mentally ill, because it exists for the protection of the community from potentially dangerous individuals. This analogy, however, is unsound. Civil commitment of the mentally ill is permitted because the mentally ill will never be criminally responsible. They lack the requisite intent and are incapable of being deterred from future criminal conduct. Accordingly, not being amenable to the criminal justice system, and posing a danger to society, they are held to receive proper treatment and rehabilitation. On the other hand, presumably a person the subject of a control order is capable of being held criminally guilty and is capable of being deterred. For this reason, the analogy is unreliable.
It follows that, insofar as div 104 purports to authorise constructive detention or house arrest, it is invalid. It is certainly arguable that s 104.5(3)(c) can be read down to preclude the possibility of constructive detention, or that it is severable. If severed, it is similarly arguable that as a matter of statutory construction div 104 does not authorise imprisonment. These questions are beyond the scope of this article. It is also beyond the scope of this article to determine whether ch III is engaged by restrictions on liberty which amount to something less than constructive imprisonment.[320] At the very least,
s 104.5(3)(c) is invalid because it invests courts with the jurisdiction to make these curious control orders, which purport to authorise non-criminal imprisonment and are foreign to Australia’s common law heritage.
IV CONCLUSION
Imprisonment is a civil death reflecting society’s judgement that a person is not entitled to liberty.[321] The decision to imprison must be cautious and considered, providing the full panoply of protections of the criminal law. There is a limited number of narrow exceptions to this rule, and courts should be slow to create new exceptional categories. That is because the government must treat those under its jurisdiction ‘with the respect and dignity that adult members of the community claim from each other.’[322] Necessarily, even the narrow exceptional categories run afoul of this principle. That is why those in immigration detention often protest that ‘[w]e have not committed a crime’ and that ‘[w]e are not criminals’.[323]
Imprisonment almost always accompanies a judgment of criminal guilt, and as the latter forms a core part of judicial power, so any attempt by the legislature or executive to detain by circumventing the criminal process should be prima facie invalid. The fact of non-criminal imprisonment raises a rebuttable presumption against the validity of that confinement, which may be overcome by proving that a person’s detention falls within a categorical exception. Underlying objections to the categorical approach is a conviction
that the criminal law is unwieldy, unrealistic, and not appropriate to meet the security challenges of the 21st century. But, as Gummow J in Al-Kateb[324] and Scalia J in Hamdi[325] amply demonstrated, a categorical approach is not only workable but also required under our shared constitutional traditions.
[*] BSc (Adv) (Hons), LLB (Hons) (Syd), LLM (Columbia). Sincerest thanks to Peter Gerangelos for reading a draft, for generous advice, and for sparking my interest during his fascinating course on Advanced Constitutional Law at the University of Sydney. I am grateful to the two anonymous referees and the Review’s Editorial Board for their comments and suggestions. All errors are mine. This article is dedicated to the memory of the Hon Justice D H Hodgson AO.
[1] [2004] HCA 37; (2004) 219 CLR 562.
[2] Ibid 648 [258].
[3] (2004) 225 CLR 1, 24 [57].
[4] Aboriginals Ordinance 1918 (NT) s 6.
[5] [1997] HCA 27; (1997) 190 CLR 1.
[6] Ibid 110.
[7] George Winterton, ‘The Separation of Judicial Power as an Implied Bill of Rights’ in Geoffrey Lindell (ed), Future Directions in Australian Constitutional Law (Federation Press, 1994) 185, 192–3.
[8] G J Lindell, ‘Recent Developments in the Judicial Interpretation of the Australian Constitution’ in Geoffrey Lindell (ed), Future Directions in Australian Constitutional Law (Federation Press, 1994) 1, 35 n 136.
[9] (1992) 176 CLR 1, 27–9.
[10] Ibid 27.
[11] Katherine Nesbitt, ‘Preventative Detention of Terrorist Suspects in Australia and the United States: A Comparative Constitutional Analysis’ (2007) 17 Boston University Public Interest Law Journal 39, 51.
[12] Al-Kateb [2004] HCA 37; (2004) 219 CLR 562, 648–9 [258] (Hayne J); Re Woolley (2004) 225 CLR 1, 24 [57] (McHugh J).
[13] Kruger [1997] HCA 27; (1997) 190 CLR 1, 110–11 (Gaudron J).
[14] Chu Kheng Lim (1992) 176 CLR 1, 27–9 (Brennan, Deane and Dawson JJ); Fardon v A-G (Qld) [2004] HCA 46; (2004) 223 CLR 575, 611–12 [77]–[80] (Gummow J) (‘Fardon’).
[15] Chu Kheng Lim (1992) 176 CLR 1, 55 (Gaudron J); Kruger [1997] HCA 27; (1997) 190 CLR 1, 110
(Gaudron J); Al-Kateb [2004] HCA 37; (2004) 219 CLR 562, 648–9 [258] (Hayne J).
[16] Chu Kheng Lim (1992) 176 CLR 1, 65–7 (McHugh J); Re Woolley (2004) 225 CLR 1, 24–6 [57]–[61] (McHugh J).
[17] Al-Kateb [2004] HCA 37; (2004) 219 CLR 562, 649–50 [263]–[265] (Hayne J). Cf Re Woolley (2004) 225 CLR 1, 25–6 [60]–[61] (McHugh J).
[18] Al-Kateb [2004] HCA 37; (2004) 219 CLR 562, 612 [136] (Gummow J); Fardon [2004] HCA 46; (2004) 223 CLR 575, 612–13 [81] (Gummow J).
[19] Chu Kheng Lim (1992) 176 CLR 1, 57 (Gaudron J), 65–6, 71 (McHugh J).
[20] Ibid 28–9 (Brennan, Deane and Dawson JJ).
[21] Christos Mantziaris, ‘Commonwealth Judicial Power for Interim Control Orders — The Chapter III Questions Not Answered’ (2008) 10 Constitutional Law and Policy Review 65, 68, 73.
[22] 233 F Supp 2d 564, 591 (SD NY, 2002).
[23] [2004] USSC 2730; 542 US 507 (2004).
[24] See, eg, Monica Hakimi, ‘International Standards for Detaining Terrorism Suspects: Moving Beyond the Armed Conflict–Criminal Divide’ (2008) 33 Yale Journal of International Law 369, 383–6; Stella Burch Elias, ‘Rethinking “Preventive Detention” from a Comparative Perspective: Three Frameworks for Detaining Terrorist Suspects’ (2009) 41 Columbia Human Rights Law Review 99, 156–9; John Ip, ‘Comparative Perspectives on the Detention of Terrorist Suspects’ (2007) 16 Transnational Law & Contemporary Problems 773, 808–9.
[25] Tung Yin, ‘Ending the War on Terrorism One Terrorist at a Time: A Noncriminal Detention Model for Holding and Releasing Guantanamo Bay Detainees’ (2005) 29 Harvard Journal of Law and Public Policy 149, 182–3.
[26] David Cole, ‘Out of the Shadows: Preventive Detention, Suspected Terrorists, and War’ (2009) 97 California Law Review 693, 695.
[27] Chu Kheng Lim (1993) 176 CLR 1, 3.
[28] Ibid 1. The definition of ‘designated person’ is now contained in Migration Act s 177.
[29] This section has since been repealed and substituted: see Migration Act s 178(1), which states that ‘a designated person must be kept in immigration detention’.
[30] Chu Kheng Lim (1992) 176 CLR 1, 10 (Mason CJ), 14–38 (Brennan, Deane and Dawson JJ).
[31] Ibid 26 (Brennan, Deane and Dawson JJ).
[32] Ibid 32.
[33] Ibid 28.
[34] Ibid 27 (citations omitted).
[35] Ibid 27.
[36] Ibid 28.
[37] Ibid 32.
[38] Ibid 29.
[39] Ibid 30 (citations omitted).
[40] Ibid 30–1.
[41] Ibid 29.
[42] Ibid 32.
[43] Ibid, citing Re Yates; Ex parte Walsh [1925] HCA 53; (1925) 37 CLR 36, 60–1 (Knox CJ), 96 (Isaacs J);
O’Keefe v Calwell [1949] HCA 6; (1949) 77 CLR 261, 278 (Latham CJ); Koon Wing Lau v Calwell [1949] HCA 65; (1949) 80 CLR 533, 555 (Latham CJ); Chu Shao Hung v The Queen [1953] HCA 33; (1953) 87 CLR 575, 589 (Kitto J) (‘Chu Shao Hung’). In Re Yates; Ex parte Walsh [1925] HCA 53; (1925) 37 CLR 36, 60, Knox CJ was of the opinion that deportation is ‘clearly preventive and not punitive in its nature’. Isaacs J acknowledged the existence of a parliamentary discretion to impose deportation as a punitive measure, noting that where deportation ‘is enacted as punishment for a crime, it necessarily falls to the judicial department’, but where it is enacted ‘as a political precaution, it must be exercised by the political department — the Executive’: at 96. Chu Shao Hung [1953] HCA 33; (1953) 87 CLR 575 concerned s 5(6) of the Immigration Act 1901 (Cth), which provided that any person deemed a prohibited immigrant was guilty of an offence and imposed a penalty of ‘[i]mprisonment for six months, and, in addition to or substitution for such imprisonment, deportation from the Commonwealth pursuant to an order made in that behalf by the Minister’: quoted at 588 (Kitto J). Kitto J said that ‘deportation is a matter for executive decision by the Minister, and, if ordered, it does not in any sense partake of the character of a punishment’: at 589. As regards detention pending deportation under s 5(6), Kitto J noted that ‘there may be no purpose to be served by the imprisonment except that of keeping the “offender” available for immediate deportation in the event of the Minister’s deciding upon that course, and it is quite right, therefore, to say that the provision for imprisonment is ancillary to the provision with respect to deportation’.
[44] Chu Kheng Lim (1992) 176 CLR 1, 33.
[45] Migration Act s 54P(1), cited in Chu Kheng Lim (1992) 176 CLR 1, 33–4. This section has since been repealed and substituted: see Migration Act s 181(1).
[46] Chu Kheng Lim (1992) 176 CLR 1, 34. Of course, it is questionable whether s 54P(1) rendered the detention actually voluntary: see Mary E Crock, ‘Climbing Jacob’s Ladder: The High Court and the Administrative Detention of Asylum Seekers in Australia’ [1993] SydLawRw 27; (1993) 15 Sydney Law Review 338, 347–8; Tania Penovic, ‘The Separation of Powers: Lim and the “Voluntary” Immigration Detention of Children’ (2004) 29 Alternative Law Journal 222.
[47] Chu Kheng Lim (1992) 176 CLR 1, 34.
[48] Ibid 10 (Mason CJ).
[49] Ibid 47 (citations omitted).
[50] Ibid 55.
[51] Ibid 57.
[52] Ibid 65–6, 71.
[53] Ibid 67, 71.
[54] Ibid 65–6, 71.
[55] Ibid 26.
[56] Ibid.
[57] Ibid 57 (Gaudron J).
[58] [1997] HCA 27; (1997) 190 CLR 1, 111. Hayne J and McHugh J agree with Gaudron J. In Al-Kateb [2004] HCA 37; (2004) 219 CLR 562, 649 [259], Hayne J said that ‘[a]t least in many cases it will be right to say that a law authorising detention divorced from any breach of the law is not a law with respect to a head of power and for that reason is invalid’. In Re Woolley (2004) 225 CLR 1, 27 [63], McHugh J said ‘[t]he federal Parliament has no general power to make laws with respect to imprisonment or detention ... [M]ost heads of federal legislative power do not seem expansive enough to justify a law that authorises or requires detention divorced from a breach of law’.
[59] [1948] HCA 7; (1948) 76 CLR 1, 333.
[60] Ibid.
[61] Sir Owen Dixon, ‘Two Constitutions Compared’ (1942) 28 American Bar Association Journal 733, 734.
[62] Chu Kheng Lim (1992) 176 CLR 1, 26 (Brennan, Deane and Dawson JJ).
[63] Al-Kateb [2004] HCA 37; (2004) 219 CLR 562, 611 [133]. Gummow J’s example with respect to s 51(xvii) is especially salient, as the United Kingdom (‘UK’) only abolished imprisonment for civil debt in 1869: Debtors Act 1869, 32 & 33 Vict, c 62, s 4.
[64] Stephen McDonald, ‘Involuntary Detention and the Separation of Judicial Power’ [2007] FedLawRw 2; (2007) 35 Federal Law Review 25, 36.
[65] Chu Kheng Lim (1992) 176 CLR 1, 55.
[66] Amalgamated Society of Engineers v Adelaide Steamship Co Ltd [1920] HCA 54; (1920) 28 CLR 129, 162 (Higgins J) (‘Engineers’ Case’), speaking of s 51(xxxv).
[67] Kruger [1997] HCA 27; (1997) 190 CLR 1, 110.
[68] Ibid 111; Re Woolley (2004) 225 CLR 1, 27 [63] (McHugh J).
[69] That is, the aliens exception was established by Chu Kheng Lim (1992) 176 CLR 1, and in no other High Court decision has the establishment of an exception constituted ratio decidendi.
[70] ‘The power and jurisdiction of parliament, says [S]ir Edward Coke, is so transcendent and absolute, that it cannot be confined, either for causes or persons, within any bounds’: William Blackstone, Commentaries on the Laws of England (Clarendon Press, 1765–69) vol 1, 156.
[71] See generally Northern Pipeline Construction Co v Marathon Pipe Line Co, [1982] USSC 187; 458 US 50 (1982).
[72] Re Woolley (2004) 225 CLR 1, 27 [63] (McHugh J).
[73] Melbourne Corporation v Commonwealth [1947] HCA 26; (1947) 74 CLR 31, 79 (Dixon J).
[74] Chu Kheng Lim (1992) 176 CLR 1, 27.
[75] Ibid.
[76] There is a third ‘functional’ approach, in which each arguable breach of the separation of judicial power is examined on an ad hoc basis to determine whether the judiciary’s exercise of its functions has been impaired. An abstract discussion of functionalism is unnecessary here. Most functionalists agree that there is ‘a nucleus of activities that uniquely belongs to each of the three branches’ and that functionalism is inapt for those core activities: Thomas W Merrill, ‘The Constitutional Principle of Separation of Powers’ [1991] Supreme Court Review 225, 232–3. This article contends that imprisonment and the criminal process are core aspects of judicial power (that is, part of the nucleus of uniquely judicial activity), and so a functional approach should not be adopted. See also Peter Gerangelos, ‘Interpretational Methodology in Separation of Powers Jurisprudence: the Formalist/Functionalist Debate’ (2005) 8 Constitutional Law and Policy Review 1; M J C Vile, Constitutionalism and the Separation of Powers (Liberty Fund, 2nd ed, 1998) 402; William B Gwyn, ‘The Indeterminacy of the Separation of Powers and the Federal Courts’ (1989) 57 George Washington Law Review 474.
[77] ‘Many attempts have been made to define judicial power, but it has never been found possible to frame a definition that is at once exclusive and exhaustive’: R v Davison [1954] HCA 46; (1954) 90 CLR 353, 366 (Dixon CJ and McTiernan J). See also Precision Data Holdings Ltd v Wills [1991] HCA 58; (1991) 173 CLR 167, 188–9 (Mason CJ, Brennan, Deane, Dawson, Toohey, Gaudron and McHugh JJ); Brandy v Human Rights and Equal Opportunity Commission [1995] HCA 10; (1995) 183 CLR 245, 267 (Deane, Dawson, Gaudron and McHugh JJ); Yanner v Minister for Aboriginal and Torres Strait Islander Affairs [2001] FCA 36; (2001) 108 FCR 543, 553 [37] (Sackville J); Australian Communications Authority v Viper Communications Pty Ltd [2001] FCA 637; (2001) 110 FCR 380, 403 [96] (Sackville J); Merrill, above n 76, 233, noting that ‘[f]ormalism is often attacked on the ground that the definitions of the legislative, executive, and judicial powers are elusive and lead to a question-begging analysis’; Vile, above n 76, 402, arguing that ‘[i]t was clear many years ago that attempts to allocate particular functions precisely to particular branches of government must fail’; Richard H Fallon Jr et al, Hart and Wechsler’s The Federal Courts and the Federal System (Foundation Press, 6th ed, 2009) 49, observing that ‘there is no intrinsically correct or universally accepted idea of appropriate judicial power’.
[78] See P H Lane, A Manual of Australian Constitutional Law (Law Book, 6th ed, 1995) 189–203; Huddart, Parker & Co Pty Ltd v Moorehead [1909] HCA 36; (1909) 8 CLR 330, 357 (Griffith CJ); Waterside Workers’ Federation of Australia v J W Alexander Ltd [1918] HCA 56; (1918) 25 CLR 434, 463 (Isaacs and Rich JJ).
[79] In R v Quinn; Ex parte Consolidated Food Corporation [1977] HCA 62; (1977) 138 CLR 1, 12, Jacobs J argued that where legislation confers a discretionary power to affect traditionally basic legal rights, ‘the judicial nature of the power springs from the effect which the exercise of the decision-making function under the legislation will have upon the legal rights’.
[80] Ibid 11.
[81] Ibid.
[82] Ibid. See also Nathan S Chapman and Michael W McConnell, ‘Due Process as Separation of Powers’ (2012) 121 Yale Law Journal 1672, 1682–4, 1688, 1692.
[83] Re Nolan; Ex parte Young [1991] HCA 29; (1991) 172 CLR 460, 497. Similarly, in Polyukhovich v Commonwealth [1991] HCA 32; (1991) 172 CLR 501, 705, Gaudron J said that the ascertainment of guilt or innocence ‘is the function of criminal proceedings and the exclusive function of the courts’.
[84] See Chu Kheng Lim (1992) 176 CLR 1, 27.
[85] Ibid 55 (Gaudron J).
[86] Kruger [1997] HCA 27; (1997) 190 CLR 1, 110.
[87] Ibid 111.
[88] Ibid 110.
[89] Gaudron J argues that ‘it is difficult to assert exclusivity except within a defined area and, if the area is to be defined by reference to exceptions, the exceptions should be clear or should fall within precise and confined categories’: ibid.
[90] Ibid.
[91] Ibid.
[92] See R v Porter [1933] HCA 1; (1933) 55 CLR 182, 186–7 (Dixon J); Michael Louis Corrado, ‘Sex Offenders, Unlawful Combatants, and Preventive Detention’ (2005) 84 North Carolina Law Review 77, 100–10.
[93] [1933] HCA 1; (1933) 55 CLR 182, 186.
[94] R v Quinn; Ex parte Consolidated Food Corporation [1977] HCA 62; (1977) 138 CLR 1, 11 (Jacobs J); Re Yates; Ex parte Walsh [1925] HCA 53; (1925) 37 CLR 36, 79, where Isaacs J deduced two corollaries of the Magna Carta: first, ‘that there is always an initial presumption in favour of liberty, so that whoever claims to imprison or deport another has cast upon him the obligation of justifying his claim by reference to the law’, and second, ‘that the Courts themselves see that this obligation is strictly and completely fulfilled before they hold that liberty is lawfully restrained’. See also Alexander Kaufman, ‘Incompletely Theorized Agreement: A Plausible Ideal for Legal Reasoning?’ (1996) 85 Georgetown Law Journal 395, 398–400.
[95] Wilson v Minister for Aboriginal and Torres Strait Islander Affairs [1996] HCA 18; (1996) 189 CLR 1, 11–12 (Brennan CJ, Dawson, Toohey, McHugh and Gummow JJ); Plaut v Spendthrift Farm Inc, [1995] USSC 32; 514 US 211, 239–40 (Scalia J, for Rehnquist CJ, O’Connor, Scalia, Kennedy, Souter and Thomas JJ) (1995). See also Chapman and McConnell, above n 82, 1682–94.
[96] (1992) 176 CLR 1, 27.
[97] Ibid 32.
[98] (2004) 225 CLR 1, 24 [57] (emphasis in original).
[99] Ibid 24–6 [58]–[60].
[100] Ibid 26 [60].
[101] Ibid 27 [62].
[102] Leslie Zines, The High Court and the Constitution (Federation Press, 5th ed, 2008) 289.
[103] Al-Kateb [2004] HCA 37; (2004) 219 CLR 562, 648 [258] (Hayne J).
[104] Hayne J finds that ‘at its root, the answer made to the contention that the laws now in question contravene Ch III is that they are not punitive’ (emphasis in original): ibid 649–50 [263].
[105] Al-Kateb [2004] HCA 37; (2004) 219 CLR 562, 650 [265], quoting H L A Hart, Punishment and Responsibility (Clarendon Press, 1968) 4–5.
[106] Al-Kateb [2004] HCA 37; (2004) 219 CLR 562, 650 [265].
[107] Ibid 650 [266].
[108] Ibid.
[109] Re Woolley (2004) 225 CLR 1, 26 [60] (McHugh J).
[110] [1963] USSC 21; 372 US 144 (1963) (‘Mendoza-Martinez’).
[111] Ibid 168–9 (Goldberg J) (citations omitted).
[112] Ibid 169.
[113] Wayne A Logan, ‘The Ex Post Facto Clause and the Jurisprudence of Punishment’ (1998) 35 American Criminal Law Review 1261, 1282.
[114] Christopher Moseng, ‘Iowa’s Sex Offender Residency Restrictions: How the Judicial Definition of Punishment Leads Policy Makers Astray’ (2007) 11 Journal of Gender, Race and Justice 125, 134.
[115] Ibid 147.
[116] George Fletcher, ‘The Concept of Punitive Legislation and the Sixth Amendment: A New Look at Kennedy v Mendoza-Martinez’ (1965) 32 University of Chicago Law Review 290, 300.
[117] Logan, above n 113, quoting Bell v Wolfish, [1979] USSC 82; 441 US 520, 565 (Marshall J) (1979).
[118] Daniel L Feldman, ‘The “Scarlet Letter Laws” of the 1990s: A Response to Critics’ (1997) 60 Albany Law Review 1081, 1088.
[119] Moseng, above n 114, 137.
[120] Ibid 147.
[121] Note, ‘Punishment: Its Meaning in Relation to Separation of Power and Substantive Constitutional Restrictions and its Use in the Lovett, Trop, Perez, and Speiser Cases’ (1959) 34 Indiana Law Journal 231, 287, quoted in Note, ‘Prevention Versus Punishment: Toward a Principled Distinction in the Restraint of Released Sex Offenders’ (1996) 109 Harvard Law Review 1711, 1724.
[122] See Maria Foscarinis, ‘Toward a Constitutional Definition of Punishment’ (1980) 80 Columbia Law Review 1667, 1672–3.
[123] [1951] HCA 5; (1951) 83 CLR 1.
[124] See ibid 258 (Fullagar J): ‘A power to make laws with respect to lighthouses does not authorize the making of a law with respect to anything which is, in the opinion of the law-maker, a lighthouse’.
[125] See, eg, Trop v Dulles, [1958] USSC 57; 356 US 86, 95 (Warren CJ, for Warren CJ, Black, Douglas and Whittaker JJ) (1958).
[126] It could be argued that the force of this criticism is limited, because in other contexts the High Court has approved a greater level of parliamentary control over constitutional requirements. For example, the consistent interpretation of s 80 of the Constitution’s guarantee of a jury for the ‘trial on indictment of any offence’ has been restricted to those offences that Parliament deems indictable: R v Archdall; Ex parte Carrigan [1928] HCA 18; (1928) 41 CLR 128, 136 (Knox CJ, Isaacs, Gavan Duffy and Powers JJ), 139–40 (Higgins J). This literalist approach has been forcefully criticised: see, eg, Zines, above n 102, 571–3. The emasculation of s 80 is not a paragon of constitutional interpretation.
[127] Al-Kateb [2004] HCA 37; (2004) 219 CLR 562, 650 [265]–[266].
[128] Hayne J’s insistence on ‘an offence against legal rules’ is similar to Frankfurter J’s insistence on a declaration of guilt for a law to constitute a bill of attainder in United States v Lovett, [1946] USSC 104; 328 US 303, 318, 323 (1940) (see United States Constitution art I § 9). John Hart Ely labelled this a literalist approach, noting that it ‘would of course permit a legislature, merely by omitting its ground of condemnation, to avoid having invalidated as a bill of attainder a statute imprisoning named parties’: Comment, ‘The Bounds of Legislative Specification: A Suggested Approach to the Bill of Attainder Clause’ (1962) 72 Yale Law Journal 330, 336–7. (Although unattributed, the comment in the Yale Law Journal was authored by John Hart Ely: see Alan M Dershowitz, ‘In Memoriam: John Hart Ely’ (2004) 117 Harvard Law Review 1743, 1743–4.)
[129] Al-Kateb [2004] HCA 37; (2004) 219 CLR 562, 649–50 [260]–[263].
[130] See R v Federal Court of Bankruptcy; Ex parte Lowenstein [1938] HCA 10; (1938) 59 CLR 556, 582 (Dixon and Evatt JJ): ‘There is high authority for the proposition that “the Constitution is not to be mocked”’.
[131] See Al-Kateb [2004] HCA 37; (2004) 219 CLR 562, 650 [265]. Although Hayne J cites H L A Hart, Punishment and Responsibility (Clarendon Press, 1968), ch 1 of Punishment and Responsibility first appeared as H L A Hart, ‘Prolegomenon to the Principles of Punishment’ (1959) 60 Proceedings of the Aristotelian Society 1, a paper read before the meeting of the Aristotelian Society on 19 October 1959.
[133] Al-Kateb [2004] HCA 37; (2004) 219 CLR 562, 650 [265].
[134] Ibid 650 [266].
[136] Hayne J’s second reason — that preventing a non-citizen making landfall is non-punitive, so segregating those who unlawfully make landfall is similarly non-punitive — does not absolve his reasoning from the definitional stop. That is because it is unclear the extent to which Hayne J’s second reason is related to the standard or central case of punishment set out by Hart. Insofar as Hayne J employed Hart’s definition, he committed the definitional stop.
[137] Zines, above n 102, 289–90.
[139] Ibid 6.
[140] Zines, above n 102, 288. See also Leslie Zines, ‘A Judicially Created Bill of Rights?’ [1994] SydLawRw 14; (1994) 16 Sydney Law Review 166, 174, where he argues that ‘one should, at least, begin with a suspicion that incarceration by legislative decree is, in effect, a legislative punishment, placing the onus on the Commonwealth to show that (outside the accepted categories) it is not.’
[141] The invocation of effects and impacts is not a reference to the constitutional significance of the harshness of detention conditions. The High Court has held that inhumane conditions do not of themselves render imprisonment unlawful: Behrooz v Secretary, Department of Immigration and Multicultural and Indigenous Affairs [2004] HCA 36; (2004) 219 CLR 486.
[142] See Doe v Pataki, 919 F Supp 691, 693 (Chin J) (SD NY, 1996): ‘Nonetheless, no matter how compelling the reasons, no matter how pure the motive, constitutional protections for individuals — even unsympathetic ones — cannot be cast aside in the name of the greater good.’
[143] Victor S Navasky, ‘Deportation as Punishment’ (1959) 27 University of Kansas City Law Review 213, 217–18 (citations omitted).
[144] Al-Kateb [2004] HCA 37; (2004) 219 CLR 562, 612 [137] (Gummow J), citing Blackstone, above n 70, vol 1, 132.
[145] Al-Kateb [2004] HCA 37; (2004) 219 CLR 562, 612 [137], quoting Hamdi[2004] USSC 2730; , 542 US 507, 554–5 (Scalia J, for Scalia and Stevens JJ) (2004).
[146] Al-Kateb [2004] HCA 37; (2004) 219 CLR 562, 612–13 [138], quoting Witham v Holloway (1995) 183 CLR 525, 534.
[147] [2004] HCA 46; (2004) 223 CLR 575, 612 [79], quoting Chief Executive Offıcer of Customs v Labrador Liquor Wholesale Pty Ltd (2003) 216 CLR 161, 178–9 [56].
[148] [2007] HCA 33; (2007) 233 CLR 307, 429–30 [352]–[353].
[149] Comment, ‘The Bounds of Legislative Specification: A Suggested Approach to the Bill of Attainder Clause’ (1962) 72 Yale Law Journal 330, 355. (This comment was written by John Hart Ely: see above n 128.)
[150] Norval Morris, ‘Introduction’ (1967) 13 McGill Law Journal 534, 538, quoted in Fardon [2004] HCA 46; (2004) 223 CLR 575, 613 [82] (Gummow J).
[151] Al-Kateb [2004] HCA 37; (2004) 219 CLR 562, 612 [136] (Gummow J).
[152] Blackstone, above n 70, vol 1, 132.
[153] Ibid.
[154] On the difference between detention in custody and ‘preventive restraints on liberty by judicial order’, see Thomas v Mowbray [2007] HCA 33; (2007) 233 CLR 307, 330 [18] (Gleeson CJ). See also at 356 [116] (Gummow and Crennan JJ), 459–60 [444] (Hayne J). It is beyond the scope of this article to offer a principled schema for determining whether a person has been imprisoned. See, eg, Sonia R Farber, ‘Forgotten at Guantánamo: The Boumediene Decision and Its Implications for Refugees at the Base under the Obama Administration’ (2010) 98 California Law Review 989, 1010.
[155] (1992) 176 CLR 1, 28–9 (citations omitted).
[156] Kruger [1997] HCA 27; (1997) 190 CLR 1, 110–11.
[157] Wesley Newcomb Hohfeld, Fundamental Legal Conceptions as Applied in Judicial Reasoning (Yale University Press, 1919) 36, 58–63.
[158] Hohfeld argues that the best synonym for ‘immunity’ is ‘exemption’: ibid 62.
[159] Winterton, above n 7, 192–3.
[160] Al-Kateb [2004] HCA 37; (2004) 219 CLR 562, 648 [258] (emphasis in original).
[161] Ibid 648–9 [258].
[162] Re Woolley (2004) 225 CLR 1, 24 [57] (emphasis in original).
[163] See Winterton, above n 7, 192–3 n 55, who placed emphasis on the words ‘or by the law of the land’ in ch 39 of the Magna Carta. Presumably, the emphasis was to point out that ‘lawful judgement of his peers’ (that is, conviction by a jury) was not the only manner in which a person could be incarcerated. But, interestingly, ‘or’ was probably mistranslated into English from the original Latin. The Latin phrase is ‘per legale judicium parium suorum vel per legem terre’. In this context, the Latin word ‘vel’ was most likely equivalent to ‘et’, meaning ‘and’ rather than ‘or’: see William Sharp McKechnie, Magna Carta: A Commentary on the Great Charter of King John (James Maclehose and Sons, 1905) 436, 442–3; Rodney L Mott, Due Process of Law (Bobbs-Merrill, 1926) 3–5 n 8. Moreover, the phrase ‘by the law of the land’ in the Magna Carta soon evolved into a due process requirement, which, in turn, came to be epitomised by the criminal trial. The first known use of the phrase ‘due process’ occurred in 1354: Statute of Westminster of the Liberties of London 1354, 28 Edw 3, c 1 (providing that the Great Charter ‘be kept and maintained in all Points’), c 3 (providing ‘That no Man of what Estate or Condition that he be, shall be put out of Land or Tenement, nor taken nor imprisoned, nor disinherited, nor put to Death, without being brought in Answer by due Process of the Law’). ‘A long course of development produced a conception of the judicial process which placed the court in the position of a detached tribunal entertaining and determining civil and criminal pleas brought before it’: R v Federal Court of Bankruptcy; Ex parte Lowenstein [1938] HCA 10; (1938) 59 CLR 556, 588 (Dixon and Evatt JJ). Thus, where a person’s liberty is at stake, the requirement of due process is a requirement of criminal process. See also Chapman and McConnell, above n 82, 1682–94; Amanda L Tyler, ‘The Forgotten Core Meaning of the Suspension Clause’ (2012) 125 Harvard Law Review 901, 923–9, 932–3.
[164] Chu Kheng Lim (1992) 176 CLR 1, 27 (Brennan, Deane and Dawson JJ) (emphasis added).
[165] Fletcher, above n 116, 290.
[166] See Fardon [2004] HCA 46; (2004) 223 CLR 575, 612 [80].
[167] Chu Kheng Lim (1992) 176 CLR 1, 55 (Gaudron J); Kruger [1997] HCA 27; (1997) 190 CLR 1, 110 (Gaudron J), 162 (Gummow J); Veen v The Queen [No 2] (1988) 164 CLR 465, 495 (Deane J); Behrooz v Secretary, Department of Immigration and Multicultural and Indigenous Affairs [2004] HCA 36; (2004) 219 CLR 486, 527–8 [121] (Kirby J); Al-Kateb [2004] HCA 37; (2004) 219 CLR 562, 646 [251] (Hayne J); Fardon [2004] HCA 46; (2004) 223 CLR 575, 613 [83] (Gummow J), 633–4 [154] (Kirby J), 653–4 [214]–[215] (Callinan and Heydon JJ); Re Woolley (2004) 225 CLR 1, 30 [69] (McHugh J), 65–6 [182] (Kirby J).
[168] Judge J Harvie Wilkinson III, ‘Towards a Jurisprudence of Presumptions’ (1992) 67 New York University Law Review 907, 908.
[169] Chu Kheng Lim (1992) 176 CLR 1, 65, 71 (McHugh J).
[170] Ibid 33 (Brennan, Deane and Dawson JJ). There is probably no difference between the ‘reasonably necessary’ and ‘reasonably capable of being seen as necessary’ formulations. In Al-Kateb [2004] HCA 37; (2004) 219 CLR 562, 610–11 [126]–[132], Gummow J approved both.
[171] Chu Kheng Lim (1992) 176 CLR 1, 57. See also Kruger [1997] HCA 27; (1997) 190 CLR 1, 109–11.
[172] Zines, above n 102, 289.
[173] McDonald, above n 64, 53.
[174] Stavros Tsakyrakis, ‘Proportionality: An Assault on Human Rights?’ (2009) 7 International Journal of Constitutional Law 468, 474.
[175] Joseph Blocher, ‘Categoricalism and Balancing in First and Second Amendment Analysis’ (2009) 84 New York University Law Review 375, 381.
[176] Kathleen M Sullivan, ‘The Supreme Court, 1991 Term — Foreword: The Justices of Rules and Standards’ (1992) 106 Harvard Law Review 22, 58, cited in ibid 381.
[177] Blocher, above n 175, 428.
[178] Chu Kheng Lim (1992) 176 CLR 1, 28.
[179] The categories are: the committal to custody, pursuant to executive warrant, of a suspect awaiting trial; involuntary detention in cases of mental illness or infectious disease; the traditional powers of the Parliament to punish for contempt; and the powers of military tribunals to punish for breach of military discipline: ibid.
[180] Ibid 29.
[181] Blocher, above n 175, 381–2.
[182] Ibid 381.
[183] Kathleen M Sullivan, ‘Post-Liberal Judging: The Roles of Categorization and Balancing’ (1992) 63 University of Colorado Law Review 293, 293, cited in ibid 382.
[184] Blocher, above n 175, 393.
[185] Wilkinson, above n 168, 907. The strength of a presumption depends on ‘its rootedness in the rule of law’, and so ‘[t]hose presumptions that are deeply rooted in constitutional or statutory text are by far the most difficult to overcome’: at 908.
[187] See generally Tyler, above n 163, 923–44, 952–4.
[188] Tsakyrakis, above n 174, 471.
[189] A general discussion of value incommensurability is beyond the scope of this article. See, eg, Joseph Raz, ‘Value Incommensurability: Some Preliminaries’ (1986) 86 Proceedings of the Aristotelian Society 117. Perhaps the central thesis of the present article can be characterised as an exercise in weak, reasoned incommensurability, that is, an argument in favour of the priority of ch III’s prohibition of non-criminal detention over the conferral of legislative power — covering a variety of subject matter — in s 51. See Jeremy Waldron, ‘Fake Incommensurability: A Response to Professor Schauer’ (1994) 45 Hastings Law Journal 813, 818. The implications of this characterisation cannot be explored here.
[190] Tsakyrakis, above n 174, 471.
[191] Ibid 472.
[192] T Alexander Aleinikoff, ‘Constitutional Law in the Age of Balancing’ (1987) 96 Yale Law Journal 943, 982.
[193] Ibid.
[194] Zines, above n 102, 289.
[195] Aleinikoff, above n 192, 991.
[196] Ibid.
[197] See United States v Stevens, 130 S Ct 1577, 1585 (Roberts CJ, for Roberts CJ, Stevens, Scalia, Kennedy, Thomas, Ginsburg, Breyer and Sotomayor JJ) (2010).
[198] Contrary to Russell Blackford’s view, it is not sufficient for the High Court to ‘simply ask whether there are good historical grounds to identify an exceptional case, or convincing practical reasons for custody that negate an appearance of punishment’: Russell Blackford, ‘Judicial Power, Political Liberty and the Post-Industrial State’ (1997) 71 Australian Law Journal 267, 280.
[199] [1997] HCA 27; (1997) 190 CLR 1, 109–10.
[200] [2006] HCA 40; (2006) 227 CLR 614, 631 [37], saying that ‘those exceptions do not fall within precise and confined categories’. Cf Thomas v Mowbray [2007] HCA 33; (2007) 233 CLR 307, 330 [18] (Gleeson CJ): ‘It is true that the circumstances under which restraints on liberty may be imposed by judicial order other than as an incident of adjudging and punishing criminal guilt are carefully confined, both by the Parliament and by the courts’.
[201] Chu Kheng Lim (1992) 176 CLR 1, 29–32.
[202] Ibid 33.
[203] Ibid 34.
[204] Ibid.
[205] Quarantine Act s 2B(2).
[206] Ibid s 12A(1).
[207] Ibid s 18(1)(e).
[208] Ibid s 31.
[209] 9 Anne, c 2.
[210] Blackstone, above n 70, vol 4, 161 (emphasis in original).
[211] Ibid 161–2.
[212] Quarantine Act 1710, 9 Anne, c 2, preamble (emphasis added).
[213] Ibid s I.
[214] The East-India Co v Sandys (1684) 10 St Tr 371, 530.
[215] Ibid 532. See also W S Holdsworth, ‘A Neglected Aspect of the Relations between Economic and Legal History’ (1927) 1 Economic History Review 114, 117–18.
[216] Homersham Cox, The Institutions of the English Government (H Sweet, 1863) 611 n (b).
[217] Opinion on the Writ of Habeas Corpus (1758) Wilm 77, 92 (Wilmot J).
[218] Crowley’s Case [1818] EngR 614; (1818) 2 Swans 1, 61–2; [1818] EngR 614; 36 ER 514, 531 (Lord Eldon LC).
[219] For example, between 1900 and 1920, some American states enacted laws subjecting prostitutes to quarantine, detention and internment. ‘The crackdown on prostitutes constituted the most concerted attack on civil liberties in the name of public health in American history ... Although this story is not well known, the parallels to the internment of Japanese Americans during World War II are unavoidable.’ Allan M Brandt, ‘AIDS: From Social History to Social Policy’ in Elizabeth Fee and Daniel M Fox (eds), AIDS: The Burdens of History (University of California Press, 1988) 147, 151–2.
[220] J M Eager, ‘The Early History of Quarantine: Origin of Sanitary Measures Directed against Yellow Fever’ (Bulletin No 12, Yellow Fever Institute, 1903) 24.
[221] ‘Quarantinable Diseases’ under the Proclamation are: cholera, highly pathogenic avian influenza in humans, human swine influenza with pandemic potential, the plague, rabies, severe acute respiratory syndrome, smallpox, viral haemorrhagic fevers and yellow fever: Quarantine Proclamation 1998 (Cth) s 21.
[222] Quarantine Regulations 2000 (Cth) reg 36.
[223] Quarantine Act s 35C(1).
[224] Ibid s 35(1C).
[225] See Michelle A Daubert, ‘Pandemic Fears and Contemporary Quarantine: Protecting Liberty Through a Continuum of Due Process Rights’ (2007) 54 Buffalo Law Review 1299, 1347–8.
[226] See generally Scott Guy and Barbara Ann Hocking, ‘Times of Pestilence: Would a Bill of Rights Assist Australian Citizens Who Are Quarantined in the Event of an Avian Influenza (Bird Flu) Pandemic?’ (2006) 17 Current Issues in Criminal Justice 451; Christopher Reynolds, ‘Quarantine in Times of Emergency: The Scope of s 51(ix) of the Constitution’ (2004) 12 Journal of Law and Medicine 166.
[227] See, eg, Mental Health Act 2007 (NSW) ch 3.
[228] So much was acknowledged by the inclusion of civil commitment of the mentally ill as a qualification to the ch III immunity in Chu Kheng Lim (1992) 176 CLR 1, 28 (Brennan, Deane and Dawson JJ).
[229] R v Richards; Ex parte Fitzpatrick and Browne [1955] HCA 36; (1955) 92 CLR 157, 158 (‘Fitzpatrick and Browne’).
[230] Ibid 162.
[231] Ibid 163.
[232] Ibid 164.
[233] Ibid 166.
[234] Ibid 167.
[235] Ibid.
[236] Kielley v Carson [1842] EngR 593; (1842) 4 Moo PC 63, 88–9; [1842] EngR 593; 13 ER 225, 235 (Parke B for Lords Lyndhurst LC, Brougham, Denman, Abinger, Cottenham, Campbell, Shadwell V-C, Tindal CJ, Parke B, Erskine J and Dr Lushington).
[237] Kielley v Carson [1842] EngR 593; (1842) 4 Moo PC 63, 89; [1842] EngR 593; 13 ER 225, 235.
[238] Doyle v Falconer [1866] EngR 184; (1866) 4 Moo PC NS 203, 217; [1866] EngR 184; 16 ER 293, 298 (Sir James W Colvile for Lord Westbury, Sir James W Colvile and Sir Edward Vaughan Williams).
[239] In the US, the power of Congress to commit for contempt is considered incidental to the legislative power if limited to self-preservation, that is, preventing acts which obstruct the discharge of legislative duty: Marshall v Gordon, [1917] USSC 82; 243 US 521 (1917). In Australia, Anne Twomey has challenged the reasoning of Fitzpatrick and Browne [1955] HCA 36; (1955) 92 CLR 157, and has argued that s 49 should be read down so that the powers of Parliament only apply to the extent that they are necessary to support the functions of Parliament: Anne Twomey, ‘Reconciling Parliament’s Contempt Powers with the Constitutional Separation of Powers’ (1997) 8 Public Law Review 88.
[240] [1942] HCA 12; (1942) 66 CLR 452, 467–8 (citations omitted).
[241] [1945] HCA 18; (1945) 71 CLR 1, 23.
[242] Ibid.
[243] [1989] HCA 12; (1989) 166 CLR 518, 540.
[244] Ibid 541.
[245] Ibid 563–4.
[246] [2007] HCA 29; (2007) 231 CLR 570, 599 [61].
[247] Ibid 583 [9] (Gleeson CJ).
[248] Re Tracey [1989] HCA 12; (1989) 166 CLR 518, 541.
[249] Ibid 554–62.
[250] Ibid 562.
[251] Ibid 563–4.
[252] Ibid 573.
[253] White [2007] HCA 29; (2007) 231 CLR 570, 598 [58] (citations omitted).
[254] [1942] HCA 12; (1942) 66 CLR 452, 466–7.
[255] [1989] HCA 12; (1989) 166 CLR 518, 540.
[256] Ibid 537.
[257] Ibid 572.
[258] [2009] HCA 29; (2009) 239 CLR 230.
[259] Ibid 250 [61] (French CJ and Gummow J).
[260] Ibid 250–1 [62].
[261] Ibid.
[262] Ibid 257 [86].
[263] Ibid 261 [97] (citations omitted).
[264] Ibid.
[265] See ibid 237 [9] (French CJ and Gummow J), 251 [65] (Hayne, Heydon, Crennan, Kiefel and Bell JJ), noting that the Military Judges appointed to the AMC did not have the tenure and security of remuneration required of federal judges by s 72 of the Constitution.
[266] Ibid 243 [30] (French CJ and Gummow J).
[267] Lane v Morrison [2009] HCA 29; (2009) 239 CLR 230, 261 [97]. Cf Re Tracey [1989] HCA 12; (1989) 166 CLR 518, 537, 540 (Mason CJ, Wilson and Dawson JJ), 572 (Brennan and Toohey JJ).
[268] [2009] HCA 29; (2009) 239 CLR 230, 247–8 [48].
[269] Ibid 260–1 [96].
[270] [2007] HCA 29; (2007) 231 CLR 570, 616–19 [123]–[133].
[271] Fardon [2004] HCA 46; (2004) 223 CLR 575, 612 [80] (Gummow J).
[272] ‘The very core of liberty secured by our Anglo-Saxon system of separated powers has been freedom from indefinite imprisonment at the will of the Executive’: Hamdi[2004] USSC 2730; , 542 US 507, 554–5 (Scalia J, for Scalia and Stevens JJ) (2004), quoted in Al-Kateb [2004] HCA 37; (2004) 219 CLR 562, 612 [137] (Gummow J).
[273] Hamdi, [2004] USSC 2730; 542 US 507, 556 (Scalia J, for Scalia and Stevens JJ) (2004) (emphasis in original).
[274] ‘Rules get interesting ... when some member of the category is within the category as stated but not within the background justification’: Frederick Schauer, ‘Rules, the Rule of Law, and the Constitution’ (1989) 6 Constitutional Commentary 69, 75, quoted in Blocher, above n 175, 427.
[275] See Re Tracey; Ex parte Ryan [1989] HCA 12; (1989) 166 CLR 518, 573, where Brennan and Toohey JJ said ‘[h]istory and necessity combine to show that courts-martial and other service tribunals, though judicial in nature and though erected in modern times by statute, stand outside the requirements of Ch III of the Constitution’.
[276] In R v Davison [1954] HCA 46; (1954) 90 CLR 353, 382, Kitto J classified a power as judicial if by 1900 it had come to be ‘so consistently regarded as peculiarly appropriate for judicial performance that it then occupied an acknowledged place in the structure of the judicial system’.
[277] (1992) 176 CLR 1, 29–31.
[278] Ibid 29–30 (Brennan, Deane and Dawson JJ), quoting A-G (Canada) v Cain [1906] UKLawRpAC 37; [1906] AC 542, 546, citing in turn M de Vattel, Le Droit des Gens ou Principes de la Loi Naturelle (1758) book I § 231, book II § 125.
[279] Chu Kheng Lim (1992) 176 CLR 1, 30 n 71, citing Blackstone, above n 70, 251.
[280] Chu Kheng Lim (1992) 176 CLR 1, 30 n 71, citing Joseph Chitty, A Treatise on the Law of the Prerogatives of the Crown and the Relative Duties and Rights of the Subject (J Butterworth, 1820) 49.
[281] Mutiny Act 1689, 1 Wm & M, c 5.
[282] Debtors Act 1869, 32 & 33 Vict, c 62, s 4.
[283] Grant v Gould [1792] EngR 3085; (1792) 2 H Bl 69, 99; [1792] EngR 3085; 126 ER 434, 450.
[284] See especially Re Tracey [1989] HCA 12; (1989) 166 CLR 518, 557–8 (Brennan and Toohey JJ).
[285] [1842] EngR 593; (1842) 4 Moo PC 63; 13 ER 225.
[286] [1842] EngR 593; (1842) 4 Moo PC 63, 89; [1842] EngR 593; 13 ER 225, 235 (Parke B for Lords Lyndhurst LC, Brougham, Denman, Abinger, Cottenham, Campbell, Shadwell V-C, Tindal CJ, Parke B, Erskine J and Dr Lushington).
[287] [1842] EngR 593; (1842) 4 Moo PC 63, 88; [1842] EngR 593; 13 ER 225, 235.
[288] See, eg, Constitution s 49.
[289] The following is adapted from R v Kwok [2005] NSWCCA 245; (2005) 64 NSWLR 335, 341–2 [14]–[21] (Hodgson JA) and A-G (NSW) v Nationwide News Pty Ltd [2007] NSWCCA 307; (2007) 73 NSWLR 635, 639–42 [28]–[41] (Hodgson JA).
[290] [1842] EngR 593; (1842) 4 Moo PC 63, 89; [1842] EngR 593; 13 ER 225, 235.
[291] [1842] EngR 593; (1842) 4 Moo PC 63, 88; [1842] EngR 593; 13 ER 225, 235.
[292] [1842] EngR 593; (1842) 4 Moo PC 63, 89; [1842] EngR 593; 13 ER 225, 235.
[293] Mantziaris, above n 21, 68.
[294] Ibid 73.
[295] See above Part IIA(3)(d).
[296] A V Dicey, Lectures Introductory to the Study of the Law of the Constitution (Macmillan, 1885) 239.
[297] Cole, above n 26, 695. The commentary highlights three features of the criminal process to demonstrate its inaptness to combat terrorism. First, it is said that the criminal law is retrospective in focus, whereas fighting terrorism requires the ability to prevent future attacks. Second, evidence obtained to justify detention of terrorist suspects is often inadmissible in a criminal trial, and the open trial format can compromise intelligence sources and methods. Finally, the criminal process is incapable of dealing with the transnational aspects of terrorism, with nations often unwilling to share evidence with one another. See, eg, Monica Hakimi, ‘International Standards for Detaining Terrorism Suspects: Moving beyond the Armed Conflict–Criminal Divide’ (2008) 33 Yale Journal of International Law 369, 383–6; Stella Burch Elias, ‘Rethinking “Preventive Detention” from a Comparative Perspective: Three Frameworks for Detaining Terrorist Suspects’ (2009) 41 Columbia Human Rights Law Review 99, 156–7; John Ip, ‘Comparative Perspectives on the Detention of Terrorist Suspects’ (2007) 16 Transnational Law and Contemporary Problems 773, 808–9. These perceived inadequacies have generated a kind of judicial anxiety in the US. For example, in Esmail v Obama, 639 F 3d 1075, 1077–8 (DC Cir, 2011), Judge Silberman said:
In the typical criminal case, a good judge will vote to overturn a conviction if the prosecutor lacked sufficient evidence, even when the judge is virtually certain that the defendant committed the crime. That can mean that a thoroughly bad person is released onto our streets, but I need not explain why our criminal justice system treats that risk as one we all believe, or should believe, is justified.
When we are dealing with detainees, candor obliges me to admit that one can not help but be conscious of the infinitely greater downside risk to our country, and its people, of an order releasing a detainee who is likely to return to terrorism. One does not have to be a ‘Posnerian’ — a believer that virtually all law and regulation should be judged in accordance with a cost/benefit analysis — to recognize this uncomfortable fact.
[298] [2007] HCA 33; (2007) 233 CLR 307.
[299] Criminal Code s 104.4(1)(c).
[300] Ibid s 104.4(1)(d).
[301] Ibid s 100.1.
[302] This also distinguishes the control order regime of div 104 from the state legislative scheme upheld in Fardon [2004] HCA 46; (2004) 223 CLR 575. The Dangerous Prisoners (Sexual Offenders) Act 2003 (Qld) div 3 authorised the Supreme Court of Queensland to order the continued detention in custody of a prisoner serving imprisonment for a ‘serious sexual offence’. Division 104 contains no requirement that the subject of a control order be serving a sentence of imprisonment for a previously committed offence. See 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, 126.
[303] Thomas v Mowbray [2007] HCA 33; (2007) 233 CLR 307, 330 [18].
[304] Ibid.
[305] Ibid 356 [116].
[306] Ibid 509 [600].
[307] Ibid 508 [599].
[308] Ibid 526 [651].
[309] Ibid 430 [354].
[310] Mantziaris, above n 21, 67 (emphasis in original).
[311] Chu Kheng Lim (1992) 176 CLR 1, 27 (Brennan, Deane and Dawson JJ).
[312] Blackstone, above n 70, vol 1, 132: ‘The confinement of the person, in any wise, is an imprisonment. So that the keeping of a man against his will in a private house ... is an imprisonment’.
[313] For a discussion of ‘constructive custody’ in the US, see Stephen I Vladeck, ‘Deconstructing Hirota: Habeas Corpus, Citizenship, and Article III’ (2007) 95 Georgetown Law Journal 1497, 1539–40.
[314] See, eg[2007] HCA 33; , (2007) 233 CLR 307, 328–9 [16] (Gleeson CJ), 355 [109], 356–7 [116]–[118] (Gummow and Crennan JJ), 507 [595] (Callinan J).
[315] Ibid 329 [17].
[316] Ibid 357 [120].
[317] Ibid 425 [338] (emphasis in original) (citations omitted).
[318] Mantziaris, above n 21, 71 (emphasis in original).
[319] Lynch and Reilly, above n 302, 127. Offences in divs 101–3 include offences for: providing or receiving training connected with terrorist acts; possessing things connected with terrorist acts; collecting or making documents likely to facilitate terrorist acts; preparing for or planning terrorist acts; directing the activities of, being a member of, recruiting for, training or receiving training from, funding or receiving funding from, providing support to, or associating with, a terrorist organisation; and financing terrorism.
[320] See Lynch and Reilly, above n 302, 121: ‘The imposition of conditions which fall short of a total deprivation of an individual’s liberty cannot be assumed to be immune from difficulty’. Cf Thomas v Mowbray [2007] HCA 33; (2007) 233 CLR 307, 330 [18] (Gleeson CJ): ‘It is not correct to say, as an absolute proposition, that, under our system of government, restraints on liberty, whether or not involving detention in custody, exist only as an incident of adjudging and punishing criminal guilt’.
[321] ‘[T]he power of loco-motion, of changing situation, or removing one’s person to whatsoever place one’s own inclination may direct’: Blackstone, above n 70, vol 1, 130.
[322] Ronald Dworkin, Taking Rights Seriously (Duckworth, 1977) 11.
[323] Australian Human Rights Commission, 2011 Immigration Detention at Villawood (2011) <http://www.hreoc.gov.au/human_rights/immigration/idc2011_villawood.pdf> 6.
[324] [2004] HCA 37; (2004) 219 CLR 562, 611–14 [135]–[140].
[325] [2004] USSC 2730; 542 US 507, 554–8, 563–9, 572 n 3, 573 (Scalia J for Scalia and Stevens JJ) (2004).