Abstract
This article explores the effect of treaty withdrawal on domestic legislation implementing a treaty in the Australian constitutional context. In R (Miller) v Secretary of State for Exiting the European Union (‘Miller’), the Supreme Court of the United Kingdom held that the executive cannot exercise its prerogative power to withdraw from a treaty where that withdrawal would frustrate or invalidate domestic law. This article contends that treaty withdrawal would be unlikely to have this effect on a law implementing a treaty in the Australian context. The article ultimately draws two conclusions. First, a law implementing a treaty would likely survive treaty withdrawal in most cases due to the law’s enduring nexus with Australia’s foreign relations, enabling its continued characterisation as a law ‘with respect to’ s 51(xxix) of the Constitution. Secondly, in the event that withdrawal does lead to a loss of constitutional support, the law would likely become prospectively invalid from the date of effective withdrawal (an outcome identical to legislative repeal in its effect). The article contends that this outcome would not, however, engage the constraint on executive power so emphatically reasserted in Miller. This is because the law’s invalidity is consistent with the implied will of the legislature and thus reinforces, rather than contravenes, the fundamental principle of parliamentary sovereignty which the constraint on executive power protects.
I Introduction
The decision of the Supreme Court of the United Kingdom in R (Miller) v Secretary of State for Exiting the European Union (‘Miller’) 1 has been characterised as ‘a victory for Parliament over the executive: of the prioritisation of considerations of democracy and accountability and…a reaffirmation of parliamentary sovereignty itself’. 2 In holding that the executive could not trigger the process of withdrawal from the European Union without parliamentary authorisation, the Court in Miller reaffirmed a constitutional principle of significance for dualist legal systems beyond the United Kingdom: the executive cannot exercise its prerogative power to withdraw from a treaty where that withdrawal would alter or destroy domestic law. 3 Similar scenarios have arisen in other jurisdictions: last year, South Africa revoked its notification of withdrawal from the Rome Statute of the International Criminal Court after the executive’s purported withdrawal was held by the High Court to be unconstitutional and invalid. 4 In so holding, the High Court emphasised that to allow the executive to terminate existing rights and obligations without first obtaining parliamentary approval would constitute a conferral of legislative power on the executive, thus contravening the separation of powers and the rule of law. 5
Fundamental to the decisions of both the Supreme Court of the United Kingdom and the High Court of South Africa was that the act of treaty withdrawal in each case would have had the effect of invalidating national implementing legislation, thereby terminating existing rights and obligations under domestic law. This effect is what was said to render the executive act beyond power, as it is contrary to the well-established principle that the executive cannot dispense with domestic law. 6 The thesis of this article is that withdrawal from a treaty would not necessarily have this effect on domestic legislation enacted to implement that treaty under Australian law. While one prominent public law scholar has suggested that Miller may have implications for the scope of the executive power to withdraw from treaties in the Australian context, 7 this would only be so if the effect of withdrawal would be to render the relevant implementing statute invalid. This article critically considers this assumption and the constitutional principles that both support and undermine it, finding that two fundamental questions arise in relation to the distinctive features of the Australian Constitution that have not yet been seriously analysed by the courts or the academy.
The first is whether a law initially enacted to implement a treaty could nonetheless sustain its constitutional validity if that treaty was subsequently altered or terminated — a question that has been contemplated by one member of the High Court of Australia, but never considered by a majority of the Court. 8 The scope of Commonwealth legislative power in Australia is constrained in a way that legislative power in the United Kingdom is not, by virtue of having a written constitution with specifically enumerated heads of legislative power. Parliament may enact a law implementing a treaty domestically only because such a law falls within the ambit of the legislative grant of power conferred by s 51(xxix): the external affairs power. Though case law indicates that the validity of legislation giving effect to a treaty is at least initially tied to the question of whether that legislation adequately conforms to the agreement it is said to implement, 9 it is less clear whether that legislation could remain constitutionally intact, supported by the external affairs power, in circumstances where the treaty is later terminated. This article contends that this depends on two factors: first, the relationship between the various ‘limbs’ of the head of power and possible doctrinal limits that may attach to these limbs; and secondly, the precise statutory method by which the treaty has been incorporated into domestic law. 10
The second question is whether a law that was validly supported by a constitutional head of power at the time of its enactment could subsequently become invalid due to a changed state of affairs. In other words, if withdrawal from a treaty does destroy the connection between the implementing legislation and any supporting head of power, does it necessarily follow that that law becomes invalid from that date? Members of the High Court have suggested that such notions of ‘creeping unconstitutionality’ are unorthodox, and that the rule of law tenets of stability and legal certainty require validity to be assessed at, and only at, the date of enactment. 11 Yet this position is difficult to reconcile with cases challenging the continuing validity of legislation made under s 51(vi) (the defence power) following the First and Second World Wars. Judicial statements in these cases suggest that where a grant of legislative power supports the making of a law based on a particular set of conditions, the relevant law purporting to exercise the power has no valid operation once those conditions cease to exist. 12 This result has been explained in the cases both as an inevitable consequence of the proportionality reasoning which attaches to the exercise of purposive legislative powers, and more persuasively, by reference to an implied legislative intent that such a law should only operate to the extent that it gives effect to its original purpose. 13 The implication that flows from the latter rationale is that legislative invalidity following treaty withdrawal may in fact affirm, rather than contravene, the will of Parliament — and thus may not engage the constraint on executive power identified above, as a rule which at its core ‘reflect[s] and support[s] the sovereignty of Parliament’. 14
The analysis proceeds as follows. Part II situates the law and politics of treaty withdrawal in the Australian constitutional context. Part III provides an overview of the operation of the external affairs power, focusing on the judicially prescribed limits on the ambit of the power and the underlying constitutional doctrine precipitating the imposition of those limits. Part IV then considers the first question identified above: whether the external affairs power might continue to support a law implementing a treaty notwithstanding that that treaty has been denounced at international law. Part V addresses the question that arises should such a law be found to no longer be supported by any head of federal legislative power: what becomes of that law, and how the possible consequences could be explained and defended as a matter of legal principle.
The purpose of this article must be made clear at the outset. In Miller, the Supreme Court reaffirmed a historic constraint on executive power that applies with equal force in Australian law: the executive cannot alter the law of the land. 15 It is not proposed here to examine the history and ambit of this constraint. Rather, the article seeks to examine and explain the probable effect of treaty withdrawal on domestic implementing legislation, on the basis of existing authority, so as to determine whether this constraint on power is likely to be engaged. Two conclusions are ultimately reached. First, a law implementing a treaty would likely survive treaty withdrawal in most cases due to the law’s enduring nexus with Australia’s foreign relations. Secondly, in the event that withdrawal does lead to a loss of constitutional support, the law would likely become prospectively invalid from the date of effective withdrawal (an outcome identical to legislative repeal in its effect). The article contends that this outcome would not, however, engage the constraint on executive power so emphatically reasserted in Miller. This is because the law’s consequential invalidity is consistent with the implied will of the legislature and thus reinforces, rather than contravenes, the fundamental principle of parliamentary sovereignty which the constraint on executive power protects.
II Treaty Withdrawal in Context
A. Political Climate
Treaty withdrawal is said to sit at a ‘critical intersection of law and power’. 16 Nothing better illustrates the truth of this statement than the United Kingdom’s intended (as at the date of writing) withdrawal from the European Union. The key legal question triggered by the results of the June 2016 referendum — whether statutory authority was required to initiate the process of Britain’s withdrawal from the Treaty on European Union — culminated in the decision of the Supreme Court in Miller. Though the constitutional implications of the decision may in many respects be inapplicable outside the ‘esoteric’ or sui generis regime of European law, 17 academic writers immediately began to speculate on the extent to which the reasoning of the Supreme Court might be harnessed to preclude the government from exiting other treaties which have been given effect domestically and may be susceptible to frustration upon withdrawal. Gavin Phillipson and Alison Young, for instance, have considered whether the United Kingdom’s withdrawal from the European Convention on Human Rights would nullify certain provisions of the Human Rights Act 1998 (‘HRA’) in contravention of constitutional principle. 18
These questions are not unique to the United Kingdom. The declaration by the High Court of South Africa that the government’s purported withdrawal from the Rome Statute of the International Criminal Court was unconstitutional is evidence of this. 19 The United States’ withdrawal from the Paris Agreement is a further high-profile example of treaty withdrawal. 20 What these examples all illustrate is that an attempt, or any suggestion of an attempt, by a government to withdraw from a treaty can be politically fraught. In each case, shifts in domestic politics have created an apparent disconnect between the political interests of the governing party and its international commitments under the relevant treaty. More broadly, these case studies seem to reflect the increasing ‘rhetoric of scepticism’ directed towards international law and the growing backlash against globalisation seen across liberal democracies in recent years. 21 In this context, public denunciation of treaty obligations is increasingly being employed by States as a means of placating domestic interest groups and achieving internal political objectives. 22
In dualist legal systems such as Australia and the United Kingdom, 23 the political consequences of attempted treaty exit only translate into legal consequences if, first, the relevant treaty has been incorporated into domestic law, and secondly, the government is unable to secure repeal of that domestic legislation due to a lack of sufficient support in one of the Houses of Parliament. If the government enjoys sufficient support, securing the repeal or amendment of implementing legislation is likely to occur notwithstanding objection in some corners — of course, this was the eventual outcome of the Miller litigation. 24 It is not inconceivable that an Australian government could lack the domestic support needed to secure the repeal of implementing legislation: to the contrary, it is a salient feature of the Australian constitutional landscape that the political party in power does not necessarily hold a majority in the Senate. 25 Moreover, the current political landscape is highly combative, replete with examples of disagreements between the government and the Senate with respect to the subject matter of international law. In March 2017, for example, the federal government announced it would not proceed with the ratification of an extradition treaty with China following strong indications that the Senate would disallow the regulations needed to implement the agreement domestically. 26 In a similar vein, an attempt by the government to amend the prohibition on ‘offending’ a person on the ground of race or ethnicity contained within the Racial Discrimination Act 1975 (Cth) (‘RDA’) — a law giving effect to the Convention on the Elimination of All Forms of Racial Discrimination (‘CERD’) — failed due to insufficient support in the Senate. 27
In this context, it is highly conceivable that a current or future Australian government might seek to withdraw from a treaty that has been implemented domestically without first securing parliamentary authorisation. Should this scenario eventuate, what would be the ramifications for the law giving effect to the treaty domestically? Would the law continue in force, or would the abrogation of the treaty at international law have consequences for the constitutional validity of the domestic law?
B. Constitutional Aspects of Treaty Withdrawal
Central to the issues considered in this article are two fundamental and interrelated principles of Australian constitutional law. The first is that an obligation arising under a treaty does not create enforceable rights and duties at municipal law unless and until implementing legislation is passed. The second is that the executive cannot dispense with, alter or create domestic law absent legislative authorisation. Both of these principles derive from the British legal tradition and are necessary corollaries of Australia’s dualist approach to international law.
Australia has adopted an approach to treaty implementation that mirrors almost exactly the process in the United Kingdom, which is best described as one of strict dualism: in both countries, the domestic implementation and termination of treaties falls within the prerogative of the Crown. 28 A treaty that has not been expressly incorporated into domestic law does not create rights and obligations under that law in the absence of legislation transforming treaty obligations into the municipal legal order. This has been referred to as the principle of no direct effect. 29
This itself cannot be separated from another underlying constitutional principle of the Anglo-Australian legal tradition: namely, that the executive cannot alter, dispense with or create domestic law without legislative authorisation. 30 First embodied in the Case of Proclamations, 31 the principle was reinforced by the Bill of Rights 1688, which provided that ‘the pretended power of suspending of laws or the execution of laws by regall authority without the consent of Parlyament is ilegall’. 32 This ‘subordination of the Crown (ie the executive government) to law is the foundation of the rule of law’ in England. 33 It is this constitutional principle that makes it necessary to ascertain the effect of treaty withdrawal on implementing legislation: if withdrawal does not result in the dispensation or modification of statute law, then this constraint on power is not engaged.
In the leading modern case applying these principles prior to Miller, 34 the House of Lords affirmed that treaties under English law are not self-executing, and moreover, that the power of the executive does not extend to ‘altering the law or conferring rights upon individuals or depriving individuals of rights which they enjoy in domestic law without the intervention of Parliament’. 35 This principle was expressly reaffirmed by a majority of the Supreme Court in Miller. Though recognising that treaties may have ‘indirect interpretive effects’ in relation to domestic law — a reference to the presumption in favour of interpreting a statute in accordance with the United Kingdom’s international obligations 36 — the majority made clear that such interpretive effects do not affect the proposition that an international treaty obligation is not part of English law unless and until it has been incorporated through an Act of Parliament. 37 Additionally, the executive cannot ‘frustrate the purpose of a statute or statutory provision’. 38 Finally, the exercise of a prerogative power may change the status of a person or thing, or change the facts to which the law applies, in a way that is legally significant: for example, an executive declaration of war may mean that activities previously lawful have become treasonable. 39 However, the exercise of this power does not create or change domestic law, ‘merely the extent of its application’. 40
These principles have been expressly adopted into Australian law by the High Court of Australia. Stephen J in Koowarta v Bjelke-Petersen (‘Koowarta’) referred to the ‘settled common law doctrine’ that the exercise of the prerogative treaty making power leaves unaffected the state of Australian municipal law.
41
In Minister of State for Immigration and Ethnic Affairs v Teoh, Mason CJ and Deane J observed that [I]t is well established that the provisions of an international treaty to which Australia is a party do not form part of Australian law unless those provisions have been validly incorporated into our municipal law by statute. This principle has the foundation in the proposition that in our constitutional system the making and ratification of treaties fall within the province of the Executive in the exercise of its prerogative power whereas the making and the alteration of the law fall within the province of Parliament, not the Executive.
42
III The External Affairs Power
The Commonwealth Parliament’s power to implement treaties into Australian domestic law is granted, and limited, by s 51(xxix) of the Constitution. Section 51(xxix) confers power on the Commonwealth to legislate ‘with respect to…external affairs’. It is an elementary principle of Australian constitutional law that in order for any Commonwealth law to be valid, there must be a sufficient connection between the law and the constitutional head of power under which the legislation is said to have been enacted. 45 The process of determining whether the law is one ‘with respect to’ a subject of power granted to the Commonwealth is referred to as ‘characterisation’.
Whether a law enacted to carry a treaty into effect domestically can maintain its constitutional character as one with respect to external affairs absent the treaty therefore depends on the ambit of the power itself. Though the most prominent aspect of the power is the power to implement treaties, a growing body of jurisprudence has established that the scope of the power is not confined to this. The gradual expansion of the power has led to the development of four ‘limbs’ which may be relied upon to uphold the validity of Commonwealth legislation: (a) a power to implement treaties to which Australia is a party, and possibly, to implement non-binding international agreements and recommendations of international agencies; (b) a power to legislate on matters affecting Australia’s foreign relations; (c) a power to legislate on matters geographically external to Australia; and (d) possibly, a power to legislate on matters of international concern.
A. Treaty Implementation
It has never been in dispute that s 51(xxix) empowers the Parliament to pass a law carrying into effect within Australia the provisions of a treaty to which Australia is a party. 46 Although this aspect of the external affairs power has been described as ‘perhaps the most obvious subject for the operation of the power’, 47 it has also been the most contentious, with significant divergence in judicial opinion as to the ambit of the power going unresolved for some time.
The primary issue that has occupied the attention of the High Court is the potentially destabilising effect of executive action on the international plane for the distribution of power throughout Australia’s federal system. As a number of scholars have recognised, federal constitutional structures pose a particular challenge for the implementation of international agreements.
48
In Australia, where legislative powers are distributed vertically between the Commonwealth and States, any State law that is incompatible with a Commonwealth law will be rendered invalid to the extent of the inconsistency.
49
This means that the power to implement a treaty conferred by the external affairs power may extend the ambit of Commonwealth power into fields which would otherwise fall within the legislative competence of the States, if the subject matter of the treaty otherwise falls outside the bounds of Commonwealth power prescribed by s 51. In a series of successive cases, this concern took shape as a debate over whether s 51(xxix) permits the Commonwealth to pass a law giving effect to any treaty, even one dealing with an entirely domestic subject matter (such as a prohibition on racial discrimination,
50
or regulation of environmental conservation
51
within Australia), or whether the subject matter of the treaty must itself contain some inherently international element. Those advocating for a narrow construction of the head of power expressed concern that an expansive view of its scope would undermine the federal compact predicated on the specific distribution of State and federal powers. In Koowarta, Gibbs CJ articulated the concern in the following terms: It is impossible to envisage any area of power which could not become the subject of Commonwealth legislation if the Commonwealth became a party to an appropriate international agreement. In other words, if s 51(xxix) empowers the Parliament to legislate to give effect to every international agreement which the executive may choose to make, the Commonwealth would be able to acquire unlimited legislative power.
52
Nevertheless, the power is not unlimited. The cases clearly establish that there are a number of conditions attaching to a law passed to implement a treaty. 62 First, as all constitutional heads of power under s 51 are ‘subject to the Constitution’, a law implementing a treaty cannot contravene any express or implied prohibitions in the Constitution. 63 Further, a law must adequately conform to the treaty it is said to implement, so that it is ‘reasonably capable of being considered appropriate and adapted to implementing the treaty’. 64 Implicit in this requirement is that there be a ‘reasonable proportionality’ between the obligations imposed by the treaty and the legislative measures adopted to achieve those obligations. 65 Further, in order for a law implementing a treaty to be valid, the treaty must be ‘bona fide’ in the sense that the executive did not enter into it merely as a circuitous device employed by the government to attract legislative power. 66 Finally, the treaty itself must prescribe a course of action to be taken by contracting parties with sufficient specificity such that it constitutes an ‘obligation’, which most likely (though not necessarily) refers to a binding obligation at international law. 67
Thus, the case law makes clear that the Commonwealth Parliament does not have plenary power to legislate on the general subject matter of a treaty merely on account of ratifying the treaty. The full ambit of legislative power under s 51(xxix) is, however, unsettled, with implications for the issues addressed in Part IV.
B. Foreign Relations
The external affairs power also supports laws with respect to matters affecting Australia’s relations with other countries. 68 In XYZ v Commonwealth (‘XYZ’), Kirby J referred to the ‘long-standing authority’ of the High Court that the power afforded by s 51(xxix) is ‘at its core’ concerned with Australia’s international relationships. 69 This limb of the head of power supports laws directed at the ‘advancement of comity with foreign governments’ and the ‘preservation of integrity of foreign states’. 70 Scholars have suggested that this would extend to laws which may negatively affect Australia’s relations with other countries, such as laws implementing sanctions and embargos against other countries. 71 As explained in Zines’s The High Court and the Constitution, ‘[w]hether the Commonwealth wishes to pursue friendly or unfriendly relations is a matter of policy’. 72
While there is no doubt that laws affecting foreign relations may be enacted under this head of power, this limb of the power has been invoked relatively infrequently; indeed, it has been relied upon as the sole basis for upholding a law on only one occasion. The case of R v Sharkey (‘Sharkey’) involved a challenge to a Commonwealth law prohibiting the publication of seditious material intended to ‘excite disaffection against the Government or Constitution of any of the King’s Dominions’. 73 In holding that the offence provision was supported by s 51(xxix), Latham CJ stated that ‘[t]he relations of the Commonwealth with all countries outside Australia…are matters which fall directly within the subject of external affairs’ and ‘[t]he preservation of friendly relations with other Dominions is an important part of the management of the external affairs of the Commonwealth’. 74
The High Court has considered this aspect of the power on several other occasions, though in every other case the challenged provisions could derive support from other heads of power or other limbs of the external affairs power. In Thomas v Mowbray, Gummow and Crennan JJ said that this aspect of s 51(xxix) could sustain the validity of criminal provisions defining a ‘terrorist act’ as one that was designed to coerce or intimidate a foreign government on the basis that terrorism is a ‘worldwide phenomenon’ affecting Australia’s relations with other countries, 75 but the majority ultimately upheld the law by reference to the defence power. Similarly, in XYZ, Kirby J was of the view that a Commonwealth offence prohibiting Australian citizens from engaging in sexual acts with minors overseas was valid because the regulated conduct was capable of affecting Australia’s relationship with other countries (relevantly in that case, Thailand). 76 The other members of the Court upheld the impugned provisions, however, on the basis that the law regulated conduct that was geographically external to Australia and thus enlivened s 51(xxix) on this basis. It was not necessary for their Honours to determine whether the law could be supported by reference to Australia’s foreign relations.
Because there are so few cases on the foreign relations aspect of s 51(xxix), it remains an ill-defined and protean concept, despite apparently sitting at the core of the head of power. One reason why this limb of the external affairs power receives so little attention is that it needs only to be invoked in fairly confined circumstances. As Gleeson CJ perceptively observed in XYZ, there is almost always an overlap between the treaty implementation and the foreign relations limb of the external affairs power; this is because any law implementing a treaty will necessarily also be a law affecting Australia’s relations with other States party to that treaty. 77 Indeed, there is some scholarly support for the view that the Commonwealth’s power to pass laws in pursuance of treaties derives wholly from the fact that the treaty itself concerns Australia’s relations with other countries. 78 The foreign relations limb is for this reason particularly important for the questions treated in this article. On one view, a law implementing a treaty would sustain its constitutional support even following executive withdrawal from that treaty, because the subject matter of the law may still be capable of affecting Australia’s relations with other States party to the treaty. This possibility and the possible doctrinal limits on relying upon this limb of the external affairs power are considered further in Part IV.
C. Geographic Externality
Section 51(xxix) confers a power to legislate with respect to places, persons, matters and things that are geographically external to Australia. 79 In reliance on this limb of the power, the High Court has upheld laws retrospectively criminalising war crimes committed in Europe during the Second World War, 80 prohibiting the commission of sexual acts against minors outside Australia, 81 and authorising the exploration and exploitation of petroleum resources in the Timor Gap. 82 A more recent decision of the New South Wales Court of Appeal indicates that this limb of the external affairs power extends to supporting a law prohibiting preparatory terrorism offences occurring within Australia. 83 The fact that the law prohibits conduct undertaken with the intention of supporting or promoting a militant’s entry into a foreign State was said to provide a sufficient nexus with a matter geographically external to Australia to enliven the power. 84 Gummow and Crennan JJ in Thomas v Mowbray indicated that control order provisions for suspected terrorists would be supported by this aspect of the power for the same reason. 85
D. International Concern
In Koowarta, Stephen J stated that legislative power under s 51(xxix) would be enlivened where the subject matter of the law was a matter of ‘international concern’. 86 Several statements by members of the High Court in the Tasmanian Dam Case similarly indicated that a nexus with matters of such concern may be sufficient to attract legislative power. 87 However, ‘grave doubt’ was cast over the existence of this limb of the power in XYZ. 88 Responding to an argument advanced by the Commonwealth that the prevention of child sex tourism constituted a shared concern of the international community sufficient to enliven the power, Callinan and Heydon JJ expressly rejected the existence of any such doctrine as a standalone basis for legislative power. 89 Their Honours noted that many of the comments made in obiter dicta appearing to support the doctrine were made in the context of considering whether a requirement of ‘international concern’ narrows the treaty implementation limb of the power, and should not be given wider application beyond this context. 90 Going on to cite concerns regarding the vagueness and volatility of ‘international concern’ as a concept (a point returned to in Part V), their Honours concluded that ‘immense difficulties’ face any court wishing to recognise the doctrine, and declined to do so. 91
Their Honours’ analysis is persuasive, and the existence of the doctrine is uncertain at best. No majority of the High Court has ever relied upon it to uphold a Commonwealth law, and several members of the Court have expressly refused to recognise its existence as a freestanding factor to which legislative power may attach. It seems more likely to be the case, as has been suggested in the literature, that evidence of a matter being of ‘international concern’ has been referred to in judgments only because this is of direct relevance to Australia’s relations with other countries. 92
IV Character of a Law Implementing a Treaty No Longer in Force
It is axiomatic that insofar as s 51(xxix) empowers the Commonwealth to pass laws giving effect to treaties, the exercise of the executive power to enter into treaties plays a critical role in determining the ambit of legislative power, at least initially. But how and to what extent the exercise of the power to withdraw from a treaty may alter the constitutional status of a law implementing that treaty remains to be seen. Absent the relevant treaty, could implementing legislation sustain its constitutional character as a law ‘with respect to’ external affairs?
A. The Relationship Between Treaty Implementation and Foreign Relations
At the heart of this inquiry is the relationship between two of the four limbs of the power identified in Part III: the treaty implementation limb and the foreign relations limb. The geographic externality limb calls for no separate consideration, because it is clear that a law with respect to a matter geographically external to Australia will be supported by s 51(xxix) whether or not there is a corresponding treaty. 93 Moreover, the existence of the fourth limb — that of ‘matters of international concern’ — can be put to one side because its very existence is dubious at best. Thus the ‘type’ of law with which this article is concerned is one that regulates a purely domestic subject matter (such as a prohibition on racial discrimination within Australia).
Where the validity of a law depends on the existence of a treaty, the law must conform to that treaty in the sense of being ‘reasonably capable of being considered appropriate and adapted to implementing the treaty’. 94 Once Australia is no longer party to the relevant treaty, it is difficult to see how a law implementing the treaty would continue to meet this criterion of validity. This is especially so given that it now seems clear that a binding obligation at international law is required to enliven the treaty implementation limb of the power, and the effect of treaty withdrawal on the international plane is to prospectively terminate a State’s obligations at international law. Although statutory provisions giving effect to recommendations of the International Labour Organisation were upheld as valid in Victoria v Commonwealth (‘Industrial Relations Act Case’), the plurality emphasised that such recommendations can only support a law under the external affairs power if the recommendations themselves can reasonably be regarded as appropriate and adapted to giving effect to the terms of a treaty to which they relate. 95 It was for this reason that Heydon J in obiter dicta in Pape v Federal Commissioner of Taxation stated that mere recommendations of various international agencies could not support economic stimulus legislation in the wake of the global financial crisis by reference to the external affairs power. 96 Moreover, although early cases provide some support for the proposition that s 51(xxix) may confer power to pass a law implementing a draft treaty or a treaty not yet in force, 97 the High Court has subsequently stated that the Commonwealth can only enact legislation implementing a draft treaty if the commencement date of the Act is postponed until the treaty becomes binding at international law. 98 If the Commonwealth is not able to rely on a treaty not yet in force for constitutional support, it seems clearly to follow by analogy that it is also unable to rely on a treaty that has been terminated as a source of legislative power. 99
But the legislative implementation of a treaty also has a direct connection with Australia’s ‘external affairs’ in the sense of affecting Australia’s relations with other countries party to that treaty. 100 For instance, the RDA and its prohibition on racial discrimination would undoubtedly continue to be a law that fosters positive relations with other countries around the world, even if the government denounced the CERD at international law. It is for this reason that Gleeson CJ in XYZ seemed to contemplate that these two strands of the head of power would invariably overlap, noting that the foreign relations limb of the power ‘includes matters the subject of treaties entered into by Australia’. 101 Similarly, Brennan J in the Tasmanian Dam Case observed that ‘[i]t is difficult to imagine a case where a failure by Australia to fulfil an express obligation owed to other countries to deal with the subject matter of a treaty would not be…a matter capable of affecting Australia’s external relations’. 102 These statements suggest these two limbs of the head of power are not mutually exclusive, a conclusion consistent with the High Court’s approach to all heads of legislative power under s 51 (‘[a] law with respect to corporations is within the power of Parliament notwithstanding that it is also a law with respect to trade, notwithstanding the limited power in relation to trade conferred upon Parliament by s 51(i)’). 103
For this reason, one view taken in the academic literature is that a law enacted to implement a treaty would maintain its validity following withdrawal from that treaty by reference to the foreign relations limb of the external affairs power. 104 On this view, the executive’s withdrawal from a treaty may not pose any difficulties at all for the continuation of the implementing law. If the subject matter of the law continues to affect Australia’s relations with other countries, it will continue to do so notwithstanding any executive act that takes place on the international plane. This result would arguably be uncontroversial: the implementing legislation would remain in force, supported by s 51(xxix), unless and until it is repealed by Parliament. The withdrawal would be effective to terminate Australia’s treaty obligations at international law, whilst the implementing legislation would remain on foot. In other words, the process for unwinding the treaty would mirror the process by which it was made and incorporated into municipal law: the separation between the international and domestic planes would be upheld, and the dualist legal system would continue to operate as intended.
Yet there are suggestions in the case law that a law affecting Australia’s foreign relations must be subject to at least the same limitations that attach to a law implementing a treaty. 105 If this were not so, it is said that the ‘strictures of compliance’ 106 attaching to legislation implementing a treaty would be redundant, because the subject matter of the law at all times would have pertained to Australia’s foreign relations. If this view is accepted more widely by the High Court, it is foreseeable that the Court may import a requirement of reasonable proportionality to a law which relies on the foreign relations limb of the power for its validity — an approach that could have consequences for the validity of a law enacted to implement a treaty once that treaty is no longer in force.
The logic supporting this approach is relatively straightforward. As described above, the High Court has imposed a number of conditions which attach to a law implementing a treaty. These conditions require, among other things, that the measures the law authorises to carry the treaty obligations into effect be reasonably proportionate to that object. This strict requirement of conformity with the treaty would be rendered otiose if the law was in any event valid at all times because its subject matter affects Australia’s foreign relations. Grappling with this, one academic writer has suggested that the existence of a treaty may have a restrictive effect on the head of power, such that legislative power ‘contracts’ if a matter broadly affecting Australia’s foreign relations is subsequently covered by a more limited treaty. 107 But this view is not compatible with the approach taken by the High Court in cases such as Horta, which make clear that the various limbs of the head of power are not mutually exclusive, 108 nor with the constitutional doctrine relating to legislative heads of power more broadly, as described above.
The better view, finding some support in the case law, is that a requirement of reasonable proportionality attaches to a law which relies for its validity on the foreign relations limb of the power. Murphy J in the Tasmanian Dam Case stated that a requirement of reasonable proportionality attaches to all aspects of the external affairs power, such that ‘if the subject of external affairs is some other circumstance [other than treaty implementation], the legislative power will extend to laws which could reasonably be regarded as appropriate for dealing with that circumstance’.
109
This argument was also accepted by Kirby J in obiter dicta in XYZ. The plaintiff had argued that the impugned law in that case, which criminalised the commission of sexual acts with persons under the age of 16 outside Australia, went far beyond what was necessary to preserve Australia’s relations with other countries or international organisations. This was because the law imposed criminal sanctions upon the conduct of Australians overseas in countries with considerably lower ages of consent for lawful sexual relations.
110
In finding that the law was proportionate, Kirby J proceeded on the assumption that proportionality was a valid criterion falling for consideration under the foreign relations aspect of the power. He stated: In so far as it is relevant to determining the constitutional question of proportionality…and thus the constitutional character of the law, the Australian legislation is neither unusual nor impermissibly overreaching. It appears generally consistent with laws passed by many other nation states.
111
In summary, a law giving effect to a treaty is likely not able to continue to be supported by the treaty implementation limb of the power following termination of that treaty at international law. However, subject to the possibility of a requirement of reasonable proportionality attaching to the foreign relations limb of the power, it is possible that implementing legislation could maintain a sufficient nexus with Australia’s external affairs following treaty withdrawal so as to continue to attract support under the foreign relations s 51(xxix). This is especially so in light of the primacy of this limb of the head of power and the recurrent statements in the case law recognising the inevitable overlap between the treaty implementation and foreign relations limbs of the power.
B. Relevance of the ‘Form’ of Legislative Incorporation
So far, this discussion has been couched in fairly abstract terms. In addition to the doctrinal considerations canvassed above, whether a law can remain a law ‘with respect to’ external affairs would depend upon the text and structure of the relevant implementing Act, and in particular, the precise statutory form in which the treaty has been incorporated into domestic law. The following discussion demonstrates that it is likely only possible for a law to sustain its constitutional character with respect to s 51(xxix) where that law operates in a way that is, at least to some extent, independent of the relevant treaty.
The reasoning of the Supreme Court in Miller is illustrative of this point. The Court’s finding in Miller ultimately turned upon the majority’s interpretation of the critical provision of the European Communities Act 1972 (‘ECA’), which provided for European Union law to be introduced into British domestic law as a primary source of law whereby all ‘rights, powers, liabilities, obligations and restrictions’ arising under the European Union Treaties from ‘time to time’ were to be incorporated without further enactment and given legal effect. 115 Because s 2(1) of the ECA was expressed in this way, the Court held that effective withdrawal from the treaties would frustrate the purpose of the provision, in the sense of ‘emptying it of content or preventing its effectual operation’. 116 At risk of oversimplification, and putting aside the Court’s broader (and more controversial) 117 concerns regarding the use of the prerogative to bring about a ‘major change’ to the constitutional arrangements of the United Kingdom, it was said that frustrating the Act in this way would offend the fundamental constitutional principle espoused in the Case of Proclamations. 118
Yet the legislative implementation of treaties in Australian domestic law rarely operates in this way. It is far more common, as Donald Rothwell explains, for an Act to give partial effect to a treaty such that the treaty obligations are modified and adapted into Australian law, either with or without an express declaration of this purpose by the implementing legislation.
119
In such situations, the rights and duties imposed under the treaty may be ‘adapted and adopted for the purposes of Australian law’, the terms of the treaty might be modified to reflect Australian circumstances or a ‘distinctive statutory scheme’ may develop around a particular treaty.
120
Moreover, even in cases where a piece of legislation annexes a treaty, or expressly states that treaty obligations are to have the force of law under Australian municipal law, the relevant obligations under the treaty are usually ‘translated’ into traditional legislative language so as to avoid the uncertainty inherent in the language of many international instruments.
121
For instance, the Migration and Maritime Powers Legislation Amendment (Resolving the Asylum Legacy Caseload) Act 2014 (Cth) deliberately removed most references to the Refugees Convention from the Migration Act 1958 (Cth). The Explanatory Memorandum explained the reason for this in the following terms: The Bill…removes most references to the Refugees Convention from the Migration Act and instead creates a new, independent and self-contained framework which articulates Australia’s interpretation of its protection obligations under the Refugees Convention. It is not the intention of the Government to resile from Australia’s protection obligations under the Refugees Convention but rather to codify Australia’s interpretation of these obligations within certain sections of the Migration Act.
122
In the event, however, that a law is expressed in such terms that the law merely transposes the full text of the treaty into Australian domestic law, or relies on the continued existence of the treaty to give the statute meaning and content, it is difficult to conceive of that law sustaining its constitutional character as one with respect to external affairs absent the treaty. The Federal Court of Australia has recognised that, at least from an interpretive perspective, the difference between a law that implements a treaty using its own expressions and one that gives effect to the ‘exact text’ of a treaty is ‘crucial’. 128 One example is the Carriage of Goods by Sea Act 1991 (Cth), which implements the Hague Rules in their entirety by providing simply that they will ‘have the force of law’. 129 The same may be said of s 8(1) of the RDA, which provides that the Act does not apply in relation to the application of ‘special measures’ as defined in art 1(4) of the CERD. 130 Such an Act or statutory provision is vulnerable to an ambulatory interpretation, because it does no more than give direct effect to the terms of the treaty as they exist on the international plane. In such a case, in the event of Australia’s denunciation of, for example, the Hague Rules at international law, the Act would be effectively ‘emptied’ of its content, and would have no practical operation once the treaty ceases to be effective at international law. It would thus lose its constitutional character as a law with respect to s 51(xxix), because the character of a law is determined by its practical effect and the rights and duties which it creates or regulates. 131
V Constitutional Validity According to Changed Circumstances
It has been said that ‘law must be stable and yet it cannot stand still’. 132 If treaty withdrawal results in the implementing legislation being deprived of its character as a law with respect to external affairs, the question that follows is what becomes of that law. Though a statute incapable of characterisation as a law with respect to a head of power under s 51 is constitutionally invalid, what is the status of a law which was valid when made but which subsequently loses its constitutional support? As noted in this article’s introduction, Anne Twomey, a prominent Australian constitutional law scholar, has assumed that in the case of treaty withdrawal, the implementing legislation would become invalid from the date of the executive’s withdrawal. 133 Twomey states that the Commonwealth’s act of withdrawal would have the effect of terminating the law’s validity — and that ‘[t]his would be the case not only when the statute gave effect to treaty obligations as they existed from time to time, but also when the statute gave substantive and enduring rights that were not ambulatory in nature’. 134 Twomey goes on to conclude that if Miller were followed in Australia, it would have the significant consequence that the executive would not have the capacity to withdraw from a treaty without parliamentary authorisation if the statute’s sole source of legislative power was found in s 51(xxix). 135 But, in her analysis of Miller, Twomey has not considered statements by members of the High Court that a law’s validity is to be assessed as at the date of enactment.
The authorities on the constitutional status of a law validly enacted but subsequently unsupported by a head of power lack clarity. The orthodox view is that the time to determine validity is the time of enactment, and subsequent judicial declarations of invalidity merely reflect the state of the law as it has always been. On this approach, a law that was supported by a head of power at the time of its passage would remain valid notwithstanding ‘changes in the milieu in which [the law] operates’. 136 Thus, it is at least arguable that as long as the implementing legislation was supported by s 51(xxix) at the time it was made, it would remain valid following termination of the corresponding treaty, notwithstanding its changed operation.
Yet the High Court has recognised that changes of fact and circumstance may be of constitutional significance — most conspicuously in the context of the defence power, the ambit of which depends upon the exigencies of the nation’s defence. 137 The parallels between the defence power and the treaty implementation limb of the external affairs power — as heads of power enlivened by international conditions prevailing at a material moment in time — seem strongly to support the proposition that a law validly passed to implement a treaty would become prospectively invalid following the treaty’s termination. For this reason, it seems that the ‘orthodox’ view appears to be only superficially embedded in Australian constitutional law, if indeed at all.
A. The ‘Orthodox’ View
The orthodox view is that the time to determine legislative validity is the time of a law’s enactment. This manifests as the ‘void ab initio’ doctrine, which holds that ‘[a] pretended law made in excess of power is not and never has been a law at all…If it is beyond power it is invalid ab initio’. 138 Under this doctrine, where a statute is found to be constitutionally invalid, such invalidity operates from the moment that the statute was enacted. 139 Though controversies surrounding the application of the doctrine have primarily tended to arise in the context of retrospective invalidation of governmental acts done in reliance on an invalid statute, 140 members of the High Court have on several occasions grappled with the question of when a law’s validity is to be assessed in circumstances analogous to the scenario envisaged in this article.
In XYZ, Callinan and Heydon JJ rejected the existence of the ‘international concern’ doctrine as a basis for Commonwealth legislative power under s 51(xxix), ostensibly because the volatility that would result if a source of legislative power was susceptible to change over time was said to be incompatible with the rule of law. In the course of doing so, their Honours stated that [a]t different times a matter may not be of international concern, may then become of international concern, and may then cease to be of international concern again. But if validity is to depend on the position not only at the time of enactment but also at the time of contravention, the outcome will be that legislation which was once invalid can later become valid, and legislation which was valid when enacted can become invalid. This volatility, and the elusiveness connected with attempts to define ‘international concern’, strongly suggest that the international concern doctrine does not exist; for if it did, it would operate antithetically to the rule of law.
141
The notion of ‘creeping unconstitutionality’, as it was described by Keane J, fell directly for scrutiny by the High Court in the context of laws said to be incompatible with ss 7 and 24 of the Constitution in Murphy v Electoral Commissioner. 144 The plaintiffs sought a declaration that certain provisions of the Commonwealth Electoral Act 1918 (Cth) were invalid on the basis that the ‘suspension period’ following the issue of writs for an election, during which no new enrolments can be processed, had the potential to disenfranchise. The plaintiffs argued that technological advancements and the availability of new resources meant there was no longer a substantial reason for suspending enrolments from seven days after the issue of writs, and that these advancements would support the closure of the roll at a date closer to the election. In other words, though the impugned provisions were accepted to have been previously valid (a concession the High Court suggested had been made to avoid the ‘destabilising implications’ of arguing otherwise), 145 they had become invalid in their current operation in light of the availability of new technologies.
French CJ and Bell J observed that such an argument ‘would allow a court to pull the constitutional rug from under a valid legislative scheme’.
146
Keane J was the only judge to explicitly engage with the above argument, in obiter dicta, and he rejected it. His Honour stated that ‘the dearth of authority supporting the plaintiffs’ contention that a law valid when made may become invalid by changes in the milieu in which it operates suggests that the plaintiffs’ argument is unorthodox at a fundamental level’.
147
Keane J reasoned that: It is the function of Parliament to make laws in order to change the world. To assert that changes in the world may unmake laws made by Parliament is to assert the existence of an exception to this understanding of the role of Parliament. In this regard, the defence power in s 51(vi) of the Constitution is indeed exceptional.
148
However, this conclusion is difficult to sustain when regard is had to judicial statements indicating that where Commonwealth power depends on a particular set of facts, legislation addressed to those facts will be ‘spent’ once the conditions to which it was directed have passed. 149 As the following section demonstrates, it is this view that finds stronger support in the case law.
Further, on closer reading, Callinan and Heydon JJ’s criticisms of the doctrine of international concern in XYZ seem to be primarily focused on the difficulties of identifying and measuring what is a matter of global concern at a given moment in time. Indeed, their Honours contrast the vague notion of ‘international concern’ with ‘public and solemn acts like treaties’, which are seen to be readily identifiable as a criterion of legislative power. 150 This indicates that their Honours’ concerns regarding legislative instability should be understood as confined to a situation where prospective invalidity is triggered by an amorphous criterion that eludes objective assessment, such as ‘international concern’.
Keane J’s proposition that laws cannot become valid due to a change in external circumstances should therefore be regarded as potentially problematic, and requiring closer examination by the High Court. Furthermore, the legislative provisions in Murphy were challenged on the basis that they infringed constitutional prohibitions, not because they lacked the character of a law with respect to a legislative head of power under s 51 — and thus, Murphy can be distinguished.
B. The Prospective Invalidity of Previously Valid Laws
In the United States Supreme Court decision of Nashville, Chattanooga and St Louis Railway v Walters, Brandeis J stated that ‘[a] statute valid as to one set of facts may be invalid as to another. A statute valid when enacted may become invalid by change in the conditions to which it is applied’. 151 A decade later this doctrine was adopted into Australian constitutional law by Dixon J in Australian Textiles Pty Ltd, who observed that a power applying to authorise measures only to meet particular facts cannot outlast those facts as an operative law. 152 This principle has been deployed by the High Court numerous times in the context of the defence power, yet has also found support in obiter dicta of the Court and in scholarly commentary relating to laws made under the external affairs power, 153 the race power, 154 and laws made in reliance on a State referral of power under s 51(xxxvii). 155 A review of these authorities reveals that in each case the grant of legislative power is conditioned upon a state of affairs which no longer exists.
1. The Defence Power
The judicial treatment of the defence power (s 51(vi)) provides the strongest support for the proposition that a law made pursuant to a treaty subsequently terminated is likely to be rendered invalid following the treaty’s conclusion. The High Court has recognised on numerous instances that the defence power is ‘elastic’ 156 in its application: it expands during times of war to meet the exigencies of the Commonwealth, operating as the ‘ultima ratio of the nation’, 157 and contracts following the abatement of threat. 158 Its strength is ‘commensurate with the exigency or danger which calls for its exercise’. 159 Thus it is the case that a state of war may justify legislative measures which would be unwarranted during peacetime. 160 In short, the application of the power depends on facts — and ‘as those facts change so may its actual operation as a power enabling the legislature to make a particular law’. 161
This principle has been applied by the High Court to declare wartime measures made under the defence power invalid following the transition from war to peace — even where the same measures have been found to be valid at an earlier point in time. Foster involved a challenge to a law purporting to extend regulations made under the Women’s Employment Act 1942 (Cth). 162 In finding that the defence power did not support the extension of these measures, the Court accepted that although the original Act did not specify a date of termination, the special purpose for which it was enacted had been exhausted. A law purporting to extend regulations made under the Act thus fell beyond the scope of s 51(vi) because the regulations themselves no longer had any valid operation. 163 The attempt to extend the regulations was said to be ‘nothing but a postponement of the time when a statutory regulation no longer appropriate to existing circumstances terminates’ and for this reason had no bearing upon defence. 164 Thus although the practical effect of the judgment was to invalidate only the statute purporting to extend the original regulations, the decision was based upon a finding that the original Act was itself no longer valid — thereby making clear that wartime legislation can cease to operate notwithstanding previous findings of constitutional validity. 165 Separately, the Court emphasised that the defence power could continue to support wartime legislative measures during the ‘winding down’ of a war: the cessation of hostilities does not have a guillotine-like effect on laws enacted under s 51(vi). In each case it is necessary to determine the nature and application of the particular measure to determine whether a law may continue in force following the establishment of peace. 166 But for present purposes the significance of the decision lies in its affirmation that a validly enacted law may expire as a response to a changed factual situation. This was confirmed by Gaudron J in McGinty v Western Australia, who cited Foster in support of the proposition that laws validly enacted under the defence power can become ‘void’ following the cessation of hostilities. 167
2. Possible Application Beyond the Defence Power
External Affairs Power
Several statements by members of the High Court indicate that laws made under the external affairs power may operate in a similar way to the defence power in response to changes in the ‘extra-statutory’ environment. This appeared to be the view of Brennan J, observing in Polyukhovich that the scope of the external affairs power must be affected by the international conditions prevailing at a given moment in time: If it be right to say that the external affairs power which may enliven legislative power under s 51(xxix) are affairs with which Australia has some connexion, the scope of the power may alter from time to time as an Australian connexion with particular relationships, sets of circumstances or fields of activity comes into existence or disappears.
168
It is there that an analogy may be drawn between the defence power and the external affairs power. In cases on the defence power this Court has determined the validity of legislative measures by reference to their capacity to assist the purpose of defence…It will be open to the Court, in the case of a challenged exercise of the external affairs power, to adopt an analogous approach, testing the validity of the challenged law by reference to its connexion with international subject-matter and with the external affairs of the nation.
169
(ii) Race Power
In Kartinyeri, Gaudron J observed that the scope of the race power under s 51(xxvi) ‘necessarily varies according to circumstances as they exist from time to time’. 171 According to Gaudron J, because the power conferred by s 51(xxvi) is ‘premised on there being some matter or circumstance pertaining to the people of a particular race…necessitating the making of a special law’, the scope of the power varies from time to time. As with the defence power, ‘a law that is authorised by reference to circumstances existing at one time may lose its constitutional support if circumstances change’. 172
(iii) State Referral Power
Similar arguments have been advanced in academic analysis of Commonwealth laws enacted under the State referral power in s 51(xxxvii), which allows States to confer on the Commonwealth the power to legislate with respect to matters not falling within its remit. Several scholars have analogised this power and the defence power in seeking to determine the likely effect of a State’s revocation of its referral upon a Commonwealth law enacted in reliance on that referral. 173 In an early analysis of this question, Graeme Johnson argued that the effect of revocation would be the invalidation of the Commonwealth law, because the subject matter with respect to which legislative power was conferred had been removed by the revocation. 174 According to Johnson, in such a case the extent of the Commonwealth power is defined by the terms of the reference, such that if the reference is revoked the effect on a Commonwealth law would be its invalidation. 175
C. Explaining Prospective Invalidity
1. Purposivity
As we have seen, laws made under the defence power may become invalid in the aftermath of war or an abatement of a threat to national security. The operation of the defence power in this way has been explained, in some cases, as an emanation of its ‘purposive’ nature. 176 Section 51(vi) is one of the few instances where legislative power is conferred not by reference to subject matter but by reference to purpose. 177
The characterisation, and thus constitutional validity, of a law made under a purposive head of power is determined by a proportionality analysis which relates ‘ends to means’. 178 A draconian law may be considered proportionate during a time of war (or during peacetime in order to respond to internal forces threatening the security of the nation, such as terrorism) 179 but not once the threat has abated. 180 Importantly, it is a critical feature of the proportionality doctrine that the impugned law’s validity must be determined ‘continuously’, rather than only at the time of enactment. According to Aharon Barak, where proportionality reasoning is the test used to assess a law’s validity, ‘there is no determining point in time…[t]he issue of constitutionality accompanies the law throughout its existence’. 181 Where a law meets the requirements of proportionality testing on enactment, but later does not, the result is that the law does not satisfy the test for validity ‘from here on in’. 182 Thus, unlike subject matter heads of power, it seems that the time at which legislative validity is to be tested against a purposive head of power is the time at which the law’s validity is challenged, and not the time of enactment.
2. A More Compelling Rationale: Implied Legislative Intent
However, this rationale does not adequately explain why laws made under subject matter heads of power may become invalid following a change in the extrinsic circumstances pursuant to which they were enacted. Nor does it acknowledge that the existence of constitutional facts establishing the necessary connection with the head of power is a condition precedent to the exercise of any power under s 51. 183
Australian law has for some time recognised that prospective legislative invalidity may be occasioned by an effluxion of time accompanied by a change in facts. In addition to the defence power cases, the very early case of Ex parte Dunne determined that the effect of ‘expiry’ of an Act of Parliament due to the effluxion of time is identical to the effect of repeal. 184 This has since been codified in statute. Under the Acts Interpretation Act 1901 (Cth), the repeal of an Act is taken to include — in addition to and aside from formal legislative repeal — its ‘expiry, lapsing or cessation of effect’. 185 This has a prospective operation: it does not affect the previous operation of the Act nor ‘any right, privilege, obligation or liability acquired’ under the Act. 186 P H Lane has similarly observed the ‘unusual phenomenon’ of changed facts resulting in ‘a lapse or exhaustion of the operation of a law’, 187 while Zines states that ‘to accept that rules of law can depend on facts is to invite a degree of instability. It must follow that a law declared valid can cease to be so when the material facts change’. 188 These statements contemplate that a law may become prospectively invalid in a much wider range of circumstances than would be brought about by the application of proportionality reasoning to a purposive head of power.
The better view, in the author’s opinion, is that such a law’s invalidity is attributable to an implied legislative intent that the law should not outlive the factual circumstances to which it was directed. This appears to have been the approach adopted by the High Court in some of the defence power cases. In Foster, for instance, finding that the impugned regulations facilitating employment measures following the conclusion of the Second World War could no longer be held constitutionally valid by reference to the defence power, the Court stated: It is open to much doubt whether the legislature has expressed sufficiently an actual intention to continue them in operation after 31st December 1946…we think it is clear that the Women's Employment Regulations must have come to an end on 31st December 1946, unless some fresh provision were enacted expressing an intention to continue them further.
189
Probably the difficulty should be met by a free application of the principles of statutory interpretation to the regulation or other enactment which was adopted in the exercise of the defence power. These principles enable the court to imply in a statutory provision obviously addressed to a particular state of facts a restriction upon its operation confining it to those facts. When the conditions to which it was directed have passed the statutory provision will then be spent.
192
Similar reasoning was advanced in academic commentary prior to the Supreme Court’s decision in Miller. John Finnis, as a prominent example, distinguished between ‘statutory rights’ and ‘treaty-based’ rights. Professor Finnis argued that treaty-based rights — that is, rights acquired by statute but given content by treaties — could be removed by executive treaty termination and that such termination of rights was not contrary to any established constitutional principle. 197 This is because the original source of the ‘rights’ in such cases is not the domestic law but the treaty itself. According to Professor Finnis, it follows that those rights are from the outset ‘inherently conditional’ on the State’s continued membership of the relevant treaty. This was the view preferred by Lord Reed in his leading dissenting judgment in Miller — that is, that treaty-based rights can be extinguished by executive action without offending constitutional principle because the existence of such rights has always been contingent upon executive action. 198 According to Lord Reed, the ECA merely constituted ‘a scheme under which the effect given to EU law in domestic law reflects the UK’s international obligations…whatever they may be’. 199 The effect that Parliament has given to European Union law is ‘inherently conditional’ on the United Kingdom’s continued membership of the European Union. 200 On this view, no existing domestic rights are removed as a result of the United Kingdom’s withdrawal from the European Union, and nor is the purpose of the ECA frustrated — rather, the Act continues to serve its purpose of ensuring that the United Kingdom adheres to its international obligations. 201
Although the majority rejected this construction of the ECA, the ‘foundering’ of this argument can in many respects be attributed to the scale of the constitutional change entailed by the executive act in those circumstances. 202 The key determinative factor for the majority in Miller appeared neither to be the scope of the executive power to withdraw from a treaty, nor the question of whether the prerogative can be employed to alter or diminish legal rights, but whether a ‘unilateral action’ by the executive should be allowed to effect a fundamental change in the constitutional arrangements of the United Kingdom. 203 Moreover, as previously explained, there has been no recognition under English law of the doctrine that the constitutionality of a law predicated on a particular state of affairs may be challenged once those circumstances have ceased to exist. By way of contrast, this doctrine has been accepted in Australian constitutional law — thereby, in the author’s view, justifying its application to laws made under the treaty implementation limb of the external affairs power.
A statement made by Dixon J in Higgins may offer some insights into why, perhaps, some judges of the High Court, such as Keane J in Murphy, have approached the question of prospective legislative invalidity with some trepidation. Noting that the principles of English law do not ‘supply a parallel to the lapse or exhaustion of an enactment because of a change of events’, his Honour suggested that to Australian judges the notion of a law ‘going automatically out of operation’ was somewhat ‘anomalous’. 204 As previously noted, the position is different in the United States, where the Supreme Court has long recognised that a law that is dependent upon a state of emergency may cease to operate when that emergency passes, notwithstanding that that law was valid when passed. 205 As the case law considered above illustrates, nor is this an entirely novel doctrine in Australian constitutional law. In the case of a law implementing a treaty, this reasoning may well apply to invalidate the law from the date of the executive’s effective withdrawal on the basis of an implied legislative intent that the law should expire at the same time as the treaty.
D. Implications for Legislative Displacement of the Prerogative
The important doctrinal implication flowing from the preceding paragraph is that if a law’s invalidity following treaty withdrawal can be understood as giving effect to the will of Parliament, then there is no scope to argue that the Act has ousted the executive power to withdraw from the treaty (as was held in Miller). 206 Nor is there scope to argue that the executive prohibition on dispensation and execution of domestic law has been engaged.
It is a settled principle of Australian constitutional law, as it is in the United Kingdom, that the prerogative can be displaced by statute where the statute evinces an express intention to do so, or where that intention can be inferred by necessary implication. 207 The foundational case, De Keyser’s, has been interpreted as standing for the proposition that where there is a conflict between a statute and a prerogative power, the statute prevails and the executive power to act contrary to the statute is displaced. 208 In CPCF v Minister for Immigration and Border Protection, Kiefel J (as her Honour then was) confirmed that executive powers may be superseded by legislation ‘when a matter is directly regulated by statute’. 209 In Miller, the majority construed s 2(1) of the ECA as precluding the executive from triggering art 50 without statutory authority, in part because doing so would frustrate the will of Parliament, rendering the Act in substance a ‘dead letter’. 210
The principle allowing for the legislative displacement of executive power protects the sovereignty of Parliament. The constraint on executive power prohibiting the executive from altering domestic law serves the same values, protecting democratic legitimacy and the primacy of the legislature over the executive. 211 Though the doctrine of parliamentary sovereignty exists in a more unbridled form in the United Kingdom than it does in Australia, where Parliament’s authority is limited by the terms of the Constitution, it is nonetheless an accepted aspect of Australian constitutional law that legislative authority is unconstrained, except where questions of constitutional invalidity arise. 212 The principle of responsible government — a fundamental part of Australia’s constitutional system 213 — means that the executive is responsible to, subject to control by, and by extension, subordinate to, Parliament. It is for this reason that it is now a general constitutional principle in Australian law that the executive cannot unilaterally dispense with statute law. 214
Paul Craig explains the values served by these constraints in the following terms: If the executive could change the law of its own volition, it could thereby bypass legislation without amendment or repeal, hence the principle in Proclamations. If the executive could use the prerogative where Parliament had already addressed the issue in a statute it could then avoid the legislation crafted by Parliament, hence the principle in De Keyser, and its extension to cases where the prerogative would frustrate the legislation…The former protects parliamentary sovereignty directly, by preventing recourse to the prerogative where it would change the law; the latter protects sovereignty indirectly, by precluding use of the prerogative where the formal law is left intact, but the executive seeks to circumvent it by use of the prerogative.
215
VI Conclusions
This article has explored the ramifications of treaty withdrawal for legislation that has been enacted to carry a treaty into force domestically in the Australian constitutional setting. It has attempted to consolidate, and clarify, the contested and complex constitutional principles that would arise should this scenario come to pass in Australia. Throughout the article, an effort has been made to identify, critique and balance the competing constitutional considerations that arise.
It would be to misrepresent the current state of the authorities to suggest that any of the questions presented in this article yield clear or unambiguous answers. Moreover, the article has not exhausted all the possible ways that this scenario could be analysed in doctrinal terms. In Australia, just as in other jurisdictions, the issue of treaty withdrawal raises fundamental and difficult questions of constitutional principle that are nuanced and multifaceted in their application. It may be the case that how a court ultimately chooses to deal with the contemplated scenario would depend on the constitutional context in which it arises, and the underlying political ‘fault lines’ at stake. 216 Yet these questions are complicated further by the distinctive features of Australian constitutional law, and in particular, its specific constitutional text and accompanying jurisprudence. Having considered these aspects, the article proffers two conclusions.
First, subject to further clarification from the High Court as to any doctrinal limits that may narrow the scope of the foreign relations limb of the external affairs power, a law implementing a treaty could in principle maintain its constitutional character as a law with respect to s 51(xxix) following treaty withdrawal. This outcome would depend on the precise statutory form of incorporation: whether the law is capable of freestanding operation absent the treaty, or whether the law is a mere ‘conduit’ giving effect to Australia’s obligations as they exist in the relevant international agreement.
Secondly, in the event that withdrawal does lead to a loss of constitutional support, the law would likely become prospectively invalid from the date of effective withdrawal. This ‘expiration’ or ‘lapse’ of the law — its effects identical to that of repeal — would not, however, engage the constraint on executive power established in the Case of Proclamations and reaffirmed in Miller. This is because the law’s consequential invalidity is best understood as a realisation of the implied will of the legislature and thus affirms, rather than violates, the fundamental principle of parliamentary sovereignty which the constraint on executive power protects.