University of New South Wales Faculty of Law Research Series
Last Updated: 9 August 2011
AFTER A REFERRAL: THE AMENDMENT AND TERMINATION OF COMMONWEALTH LAWS RELYING ON S 51(XXXVII)
Andrew Lynch, University of New South Wales[*]
This paper was published in Sydney Law Review (2010), Vol.32, no.3, pp.363-87. This paper may also be referenced as  UNSWLRS 24.
The importance of State referrals under section 51(xxxvii) as a basis for Commonwealth legislation on topics of national significance continues unabated. Yet, uncertainty still lingers over several aspects of the power. This article briefly revisits some of those perennial questions before embarking on a deeper discussion of the specific issues surrounding interpretation, amendment and termination of referrals, with recourse to recent judicial opinion on the power in Thomas v Mowbray (2007). In particular, the great challenge in the area is how the States may constrain the Commonwealth’s power of amendment so as to preserve the limited nature of their initial reference. The article argues that an appreciation of the distinctive nature of s 51(xxxvii) as a means of federal co-operation must underpin the Court’s approach to its interpretation and the questions to which it gives rise.
The enigmatic qualities of the Commonwealth’s power to legislate with respect to matters referred to it by the States have long dominated discussion of the relevant constitutional provision. For, while the terms of s 51(xxxvii) appear to convey a straightforward idea with succinct elegance, in truth the power has been laden with uncertainty from its inception. Having received only limited judicial attention since Federation, questions remain about the precise operation and effect of the power in many respects. This is highly undesirable given the contemporary importance of the power as demonstrated by its role in the creation of laws of great national significance in the last decade addressing the regulation of corporate entities, the prevention of terrorism and, as of late 2009, the industrial conditions of almost all Australians working in the private sector. Unsurprisingly, prominent commentators have sought to reinvigorate discussion about the power in light of its rather unexpected renaissance.  The purpose of this article is not simply to revisit many of the familiar mysteries of s 51(xxxvii), although some brief consolidation of its known parameters is a necessary overture to what follows. Instead, and as the title suggests, the particular focus of this contribution is to examine the flexibility of Commonwealth power with respect to laws which owe their original enactment, at least in part, to a referral from another legislature.
Although a law made pursuant to s 51(xxxvii) is of course a Commonwealth law, its subsequent amendment may present unique challenges to the Commonwealth’s legislative capacity. Most commonly nowadays, referrals take the form of legislative text, which the Commonwealth is then empowered to enact in the same or in ‘substantially’ the same terms as it appears in the referring Acts of the relevant States. The validity of subsequent Commonwealth amendment of legislation created in this way clearly gives rise to issues which are distinctively complex compared to those involved in the usual process of characterising Commonwealth statutes with respect to express constitutional grants of power.
Even if a referral from the State takes the form of a general statement of subject matter, it is perhaps still too much of a simplification to equate this with the express placita of s 51. For when legislating on a subject referred by the States, the Commonwealth, while not dependent on State power as such, must ensure its law has a character within the ambit set down by the legislatures of the constitutionally subordinate jurisdiction, instead of the typical requirement that it be sufficiently connected to that topic simply as a grant of constitutional power. This may potentially create practical differences where disputes as to the scope of such a referral arise, including those triggered by a later attempt by the Commonwealth to amend the legislation in question. In particular we might ask to what degree, and precisely how, might the application of the principles of constitutional interpretation bear upon attempts to interpret the relevant provision in State laws stipulating the ‘matter’?
In addition, through what means can the States most effectively constrain the Commonwealth’s powers of amendment of referred legislative text? This article argues that this issue is of central importance in the continued use of s 51(xxxvii) as a major facilitator of ‘co-operative federalism’. The States will only be willing to hand over areas to Commonwealth control if they can be confident that sufficient safeguards are in place to prevent over-reaching or misuse of those powers by the national legislature. In all three of the legislative schemes referred to above, the drafters have attempted to contain the scope of possible Commonwealth exploitation of the power to amend the initial text reference through largely similar, though not identical, semantic formulations. While this is doubtless a challenging task, it seems fair to say that this well-founded concern has introduced increasing complexity and ambiguity to the underpinning legislation. In the case of the anti-terrorism referrals supporting Part 5.3 of the Commonwealth Criminal Code, this problem was compounded by a dubious attempt to give statutory force in Commonwealth law to an approval process more typically confined to the terms of an underlying intergovernmental agreement. Some light, none of it terribly positive, was shed on these developments by the High Court in Thomas v Mowbray and this provides a useful point of departure for exploring how the issue of ongoing State control may be addressed by future cooperative schemes.
Finally, there is the matter of terminating a Commonwealth law made in reliance on a State referral via s 51(xxxvii). Of course, what the Commonwealth Parliament enacts it may repeal, but what role exists for the States after the referral? This question has chiefly focused on the spectre of a referring State withdrawing the ‘matter’ from the Commonwealth, but the possibility of the State seeking to narrow the scope of the referral through amendment of its own enactment should not be discounted. The issue of what effect follows from a State withdrawal of the referred matter has long been pondered but is yet to be determined. In the context of recent referrals from the States, this issue has particular relevance for provisions that expressly purport to maintain the Commonwealth legislation as amended even if the States withdraw their ‘amending reference’ (an ‘amending reference’ accompanies the primary or initial reference and, as its name suggests, empowers the Commonwealth Parliament to subsequently amend the statute which it has enacted using the initial reference). Whether this is valid or whether instead the Commonwealth law reverts to the original text as initially referred (the ground upon which any later amendments rested having now been dissolved), is an unknown. In the past, several commentators have attempted to address the problem of termination by making an analogy of the referral power with those other few powers the operation of which also hinges upon external factors, namely the power to legislate with respect to defence in s 51(vi). However, Johnson highlights a crucial distinction when he acknowledges that ‘the set of facts in regard to s 51(37) [sic] is dependent on the State legislatures and can be changed by enactment’ rather than being ‘beyond the control of any legislature’. As a result, very unusual considerations clearly apply in respect of Commonwealth laws made using this power rather than any other.
That these are all issues calling for better understanding is beyond doubt. Not only has the referral power been used in the last decade to enable the Commonwealth to introduce regulatory schemes in areas of major importance, but the power is also central to extant proposals for either a Commonwealth ‘takeover’ of other policy areas such as health or further instances of Commonwealth-State co-operation. In this vein, greater use of the referral facility is seen by many as a means through which real reform of the Australian federal system might be achieved more generally without resort to the referendum mechanism in s 128. In short, while traditionally scholarship on the referrals power tended to treat it as something of an academic curiosity (reflecting its marginal role in 20th century Australian federalism), that situation no longer pertains. As the growth in literature on s 51(xxxvii) since the enactment of the Corporations Act demonstrates, there is now a very strong need to resolve the uncertainties surrounding this unique source of Commonwealth power.
II DEBATE AND CONSENSUS ON KEY ASPECTS OF S 51(XXXVII)
Before turning to the particular areas of inquiry identified in the introduction, some attempt to map the general topography of s 51(xxxvii) is necessary. The grant of power is made in the following terms:
(xxxvii.) Matters referred to the Parliament of the Commonwealth by the Parliament or Parliaments of any State or States, but so that the law shall extend only to States by whose Parliaments the matter is referred, or which afterwards adopt the law.
A History and Purpose
The placitum has clear precursors in proposals for colonial co-operation dating back to the mid 19th century. Its most direct antecedent was s 15(h) and (i) of the Federal Council of Australasia Act 1885 (Imp). The latter clause introduced to the referral concept two of the more striking features present in the text of s 51(xxxvii): the possibility of Commonwealth laws limited in application to only referring states, and also a facility for subsequent adoption of the law by others. Earlier versions of the referral power did not so explicitly envisage a selective operation for the law which might result from a referral, even when they required that the referring legislatures have ‘an interest in the question so submitted’ to the federal Assembly. In contrast, the ‘spectacle of a kind of Swiss cheese Commonwealth law...is plainly open’ under s 51(xxxvii).
The present Chief Justice has noted the possibility that a Commonwealth law made under s 51(xxxvii) may ‘have application to one or more, but not necessarily all, States of Australia...does not seem to have been prominent in the consideration of the power during the Convention debates’. In so remarking, his Honour presumably had in mind Commonwealth laws of limited application but concerned with generic topics, rather than those addressed to a localised situation. For, as the Convention Debates show, the utility of the power as one which would allow the States to overcome their lack of extraterritorial legislative capacity so as to address a shared problem across a region of the country (two examples given of such a scenario were a boundary dispute or the management of a river system), was central to support for its inclusion in the Constitution. Indeed, the possibility of enabling a federal law limited in such a way featured far more frequently in the debates than potential use of the referral power to create a national law of uniform application. Sir John Downer put it most clearly when he said:
It may be that questions may afterwards arise which concern one, two, or three states, but which are not sufficiently great to require a complete revision of the whole Constitution, with all the troublesome proceedings that have to be taken to bring about a reform. It would much facilitate matters if these questions could be referred to the Federal Parliament.
French J’s suggestion of the inadequacy of the framers’ contemplation of the possible ‘Balkinisation of Commonwealth laws’ underplays the significant appeal with which they viewed the possibility of statutes enacted pursuant to s 51(xxxvii) so as to address a localised situation of no relevance to non-referring States. Use of the power in precisely that way clearly shaped much of the relevant debate. At the same time, the anxieties of those with concerns about the provision were more strongly elicited by the prospect of a federal law of selective application than other uncertainties. In particular, Deakin viewed as problematic the inequities that would follow from the expenditure of Commonwealth funds raised uniformly through the taxation power in support of such legislation. This prompted Symon to suspect the inclusion of the provision in the draft Constitution was a ‘mistake’ carried over from the Federal Council of Australasia Act 1885 which made no provision for the Council to exercise fiscal powers. However, the prospect of some States referring to the Commonwealth a generic ‘matter’ over which their neighbours continued to enjoy exclusive legislative power seemed generally unrealistic to the framers.
That so much of the Convention debate focussed upon the issue of selective application of federal laws may explain the very inconclusive attention given to other aspects of the power which, while less explicit, are clearly central to any use of the provision. The limitations of the Debates has been compounded by the rarity of opportunities which the High Court has enjoyed to shed light on many of these pivotal questions. With no cause to discuss the power at all in the first half-century after Federation, and only sporadic attention since, the Court’s contribution to our understanding of s 51(xxxvii) has necessarily been significantly supplemented by secondary commentary. Between them, these various sources have enabled a dominant consensus to emerge on many, though far from all, parts of the referral power.
B What is Referred?
There is, for instance, no doubt that the ‘matter’ that is the subject of a State referral is, as indicated in the introduction above, not State legislative power per se. The resulting law is still one made in the exercise of the legislative power of the Commonwealth. Thus it is subject to all the restrictions imposed by the Constitution upon the exercise of that brand of power. At the same time, however, the Commonwealth is able to enact a law free of any limitations upon State legislative power – a point well appreciated as a rationale for the power by the Constitution’s framers.
Distinguishing the simple referral by a State of a ‘matter’ from an actual transfer or delegation of its legislative power is entirely consistent with the referral power being found amongst the list of powers shared by the Commonwealth and State Parliaments. In Graham v Paterson, McTiernan J stressed that s 51(xxxvii) ‘is a power concurrent with the power of the State to legislate with respect to the referred matters. It is not that power itself. Having regard to the terms of s 51 (xxxvii) and s 107 it could not be that power.’ Anne Twomey has noted that any attempt, however unlikely it must seem, by a State to cede exclusive power over a matter to the Commonwealth would be invalid.
As a consequence of this interpretation, the States retain their legislative capacity with respect to any referred ‘matter’, with the potential for any existing or future laws they enact to be rendered invalid to the extent of any inconsistency with a Commonwealth statute passed pursuant to a State referral. It is possible, of course, to devise strategies which avoid or at least limit the likelihood of this occurring and which might be agreed upon by the respective legislatures in order to facilitate the referral. A clear statement of an intention within the Commonwealth law that it is not intended to ‘cover the field’ of the subject-matter with which it is concerned will not dispose of the risk of a direct inconsistency being found, and something more specific may be necessary to hem in Commonwealth power.
There remains the issue, highlighted in the introduction, of whether the ‘matter’ that is referred is subject to any special rules of interpretation should the dependent legislation be challenged. This will be considered in the following Part.
C How is Referral Achieved?
As to how the ‘matter’ itself is referred, in R v Public Vehicles Licensing Tribunal (Tas); Ex parte Australian National Airways Pty Ltd, the High Court said it ‘seems absurd to suppose that the only matter that could be referred was the conversion of a specific bill for a law into a law’. In that case, the challenged State law simply provided that, subject to certain temporal conditions, ‘the matter of air transport is referred to the parliament of the Commonwealth’. Although such generally expressed referrals are permitted, Craven has pointed out that substantial policy and drafting issues in expressing the precise limits of any reference remain – whether text-based (ie a complete bill) or one of subject-matter. In one sense his pessimism about the success of such a ‘daunting exercise’ seems misplaced given the reinvigorated use of the power since he wrote on the topic. But generally, the scope potentially afforded the Commonwealth by recent major referrals to enact additional legislative initiatives bears out his scepticism that a State can effectively contain Commonwealth power in respect of the ‘matter’ referred. This topic is discussed at greater length below in respect of the later amendment of federal laws reliant upon s 51(xxxvii), but it may need to be recognised that it is actually meaningless to talk of ‘exact limits of the matter or matters to be referred’ in most cases.
In addition to confirming the open-ended form in which a referral may be made, Public Vehicles Licensing also made it clear that the States could impose conditions upon the referral – including those which would determine the referral after the elapse of a specified period or upon the happening of an event. This had been a contentious issue in the negotiations over a war-time referral from the States, with no clear opinion emerging on whether s 51 (xxxvii) authorised a ‘gift’ or merely a ‘loan’ of a subject-matter of State legislative competence to the Commonwealth. The Court’s unanimous view was that:
It is plain enough that the Parliament of the State must express its will and it must express its will by enactment. How long the enactment is to remain in force as a reference may be expressed in the enactment. It none the less refers the matter. Indeed the matter itself may involve some limitation of time or be defined in terms which involve a limitation of time...There is no reason to suppose that the words ‘matters referred’ cannot cover matters referred for a time which is specified or which may depend on a future event even if that event involves the will of the State Governor-in-Council and consists in the fixing of a date by proclamation.
However, while the Court confirmed that a referral was so ‘determinable’, it declined to address what one might have thought was the interrelated question of whether the State might simply repeal a referring statute, strangely describing this as ‘only a subsidiary matter which if decided might throw light on the whole ambit or operation of the paragraph’. The issue of revocation will be discussed below, but the effect of the Court’s reluctance to clarify whether this is possible, has understandably led to States seeking to retain some level of control over the referrals granted through insertion of clauses providing for their determination after a certain time or through executive action (in accordance with the decision in Public Vehicles Licensing).
D Authority to Refer a Matter
The text of s 51(xxxvii) requires the referral of the matter to the Commonwealth Parliament to come from its State counterparts, with the High Court having confirmed that ‘the Parliament of the State must express its will and it must express its will by enactment’. At the Convention Debates, Barton stressed that the need for parliamentary referral was ‘in the spirit of democracy’ and accordingly the referral would not amount to an ‘evasion of responsibility’, which might well be the case were the power left to the discretion of the executive governments of the States.
The requirement of parliamentary referral is more likely to arise as an issue when State governments agree, as parties to the underlying intergovernmental agreement, to later amendments of the Commonwealth law on such a scale that a fresh referral might be thought warranted. In Thomas, when examining whether the introduction by the Commonwealth, with the agreement of the Council of Australian Governments, of a scheme of control orders for terrorism suspects was supported by the legislative referral three years earlier, Kirby J declined to ‘interpret the provisions of s 51(xxxvii) of the Constitution to permit the parliamentary reference of constitutional power to be achieved without any relevant parliamentary involvement, as by the use of communiqués by heads of government alone’. In practice, however, such a question is always likely to be subsumed within the broader one of determining the scope of either the subject-matter originally referred or of the amending reference accompanying a referral of legislative text.
There is, however, an additional question: if State Parliaments are effectively prohibited from delegating their power of referral to the Executive, then what are we to make of provisions which provide that an effective referral to the Commonwealth will determine after a stipulated period ‘or on a later day fixed by the Governor by proclamation’? Is a statutory authority granted to the executive so that it may extend the life of a referral beyond its scheduled expiration permitted by s 51(xxxvii)? It can hardly be said that this is merely a condition accompanying the reference. The general tenor of conditions is to limit the referral – in scope or duration – not to provide for its continuation past a certain point. Rather than a condition upon the referral received by the Commonwealth, an executive power to extend it is essentially an option for renewal. With the placitum clear that referrals are to issue from Parliament alone, it seems inconsistent to allow this requirement to be a one-off and for the executive to decide whether the referral is to continue after a stipulated period has elapsed. That sense is strengthened by recognition of the almost certainly altered composition of both the Parliament and the government by the time the referral falls due for extension. A significant factor in allaying concern over the executive’s power to extend a referral would be certainty as to whether the State legislature retains its power to revoke it. If so, then the parliament remains the central actor since it could, at least in theory, override the decision of the executive. But, as averred to above, the issue of revocation remains shrouded in doubt.
To summarise, the accepted wisdom on s 51(xxxvii) has established:
Although this part has sought to clarify those aspects of the referral power that are largely free of doubt, inevitably quite complex questions have arisen about the status of the laws which result. In particular, uncertainty over the amendment and termination of laws dependent upon s 51(xxxvii) cannot be seen as independent from the unusual circumstances of their creation. It is to an exploration of these issues that the following two parts now turn.
III AMENDMENT OF COMMONWEALTH LEGISLATION ENACTED PURSUANT TO S 51 (XXXVII)
Even though an Act made pursuant to s 51(xxxvii) is indubitably a Commonwealth law, arguably the truly distinctive nature of such laws emerges more through consideration of their subsequent amendment or termination than creation. Many of the ambiguities of such laws post-enactment were averred to in the preceding part. The central, but by no means exclusive, puzzle is how the referral can be made in such a way that the Commonwealth enjoys the necessary capacity to maintain and enhance the law’s operation through amendment, without this flexibility being exploited to the detriment of State power. Given the scale and complexity of recent examples, a bare referral of legislative text by the States with no accompanying acknowledgment of the need for subsequent amendment of the Commonwealth law which is to result is simply impractical. States have thus settled on the practice of supplementing an ‘initial reference’ of legislative text with an ‘amending reference’ which permits Commonwealth responsiveness to the law’s necessary development. Ensuring the scope of the latter is of meaningful utility to the Commonwealth without enabling it to destroy the integrity of the primary reference is the real challenge in current reliance upon s 51(xxxvii). Its successful resolution is vital to the willingness of both levels of Australia’s federal system to use this constitutional device as a mechanism for co-operative federalism over the long-term.
In this Part of the paper, I consider this challenge in light of the presently favoured approach to drafting such referrals and some High Court consideration of the same in the case of Thomas. A directly related, rather than distinct, question is the relevance and legitimacy of statutory recognition of the intergovernmental processes which underpin laws resulting from State referrals. However, before turning to these issues some comment on the construction of referring legislation is a necessary preliminary.
A The Principles Applicable to Construction of Referrals
In Graham v Paterson, Latham CJ said that, when a State refers a matter to the Commonwealth Parliament, ‘it produces the result of adding to the paragraphs of s 51 a further paragraph specifying the matter referred’. This invites the question: how should the Court determine the scope of the matter referred? Is the correct approach to construe a referring Act in accordance with the principles of interpretation relevant for legislation of the referring jurisdiction? When one considers the potential impact of a State-based Charter of Human Rights, such as that found in Victoria, it is clear that very real differences may arise depending on the State in question. What might the effect of such distinctions be for the general meaning of the Commonwealth law? Alternatively, and in light of Latham CJ’s analogising a referral to effectively a new head of power, should the Court apply those principles of constitutional interpretation it would bring to bear on any other subject-matter grant of legislative power? What might the implications for the scope of the referred matter be if its meaning does not depend exclusively upon ordinary methods of statutory construction of the referring Act? It is clear that Commonwealth laws passed in reliance on s 51(xxxvii) are subject to express and implied constitutional limitations, but does ‘constitutionalising’ the referral itself promote an especially liberal construction of its terms or, conversely, expose it to interpretative considerations which might serve to inhibit it in some way?
Despite indications from lower courts that referrals in State legislation do not require any special or enhanced approach beyond a search for the ‘natural and commonsense meaning’ of their language, some interesting light on the extent to which contextual considerations might affect construction of these State Acts is given in the dissenting opinion of Kirby J in Thomas v Mowbray. In that case the particular ambiguity did not lie in any expression of subject-matter or the descriptors applied to the appended legislative text, but crucially in the scope of the generic term ‘amendment’ in ss 3 and 4(1)(b) of the Terrorism (Commonwealth Powers) Act 2003 (Vic) (Referring Act). The meaning of this word and the extent to which it allows the Commonwealth to deal independently with legislation enacted pursuant to a referral is, of course, the central concern of this Part and the opinions from Thomas which shed light on this are squarely considered in the following section. The focus at this stage is simply on the issue of interpretative method.
The Commonwealth accepted in oral argument that the meaning of the provisions in the Referring Act was context-dependent. Accordingly, Kirby J voiced his opinion that it was significant the referred text had been included as a Schedule to the Referring Act, emphasising the distinction between that approach and that of merely referencing provisions tabled in the Parliament of another jurisdiction – as the Victorian legislature had done with respect to the bill which became the Corporations Act 2001 (Cth). His Honour ascribed the difference, in this instance, to the Parliament’s ‘acute need to ensure greater clarity and precision’ in setting the scope of the reference – and thus containing the Commonwealth’s power to amend the legislation which it sustained.
Additionally, Kirby J seemed to assert that the particular context and specificity of any intergovernmental agreement preceding the referral are relevant to the process of construing a referring Act. His Honour contrasted the ‘restricted nature’ of the referral with respect to terrorism laws from that which enabled the national corporations law, by highlighting the purpose and intent of the State in each instance. He suggested that the terrorism powers were given more guardedly than those designed to facilitate the construction of a broad, uniform framework for corporate regulation. In part he relied upon a consideration of the Victorian Attorney-General’s second reading speeches in respect of the referring legislation for both schemes, but also the basic dimensions of the resultant law itself (a new Part of the Commonwealth Criminal Code as compared to two comprehensive federal statutes).
However, some context was not seen as relevant. While Kirby J was willing to contrast the intergovernmental agreements underpinning the two schemes, he declined to read the Referring Act under consideration more broadly on account of the fact that it provided for the reference to be terminated with three months notice. Nor did he see the safeguards of the new Commonwealth legislation itself as relevant – namely the existence of a sunset clause and a requirement for review of the operation of the provisions by COAG in 2010. These he dismissed as political, not legal, factors. That may be prudent in respect of the features of the Commonwealth Act, but disregarding the possibility of termination seems rather at odds with his Honour’s insistence that the scope of the reference is to be determined ‘having regard to the terms in which the will of the State Parliament concerned has been expressed and other relevant considerations’.
Amongst those further considerations, Kirby J appeared to suggest the existence of a general principle that referring Acts should be construed narrowly because, of their nature, they are likely to produce a diminution of State power. His Honour said:
The Referring Act was enacted for the purpose of referring legislative power from the State Parliament of Victoria to the Federal Parliament. A failure on the part of this Court to adhere to established principles of interpretation would enlarge federal legislative powers at the expense of those of the States.
This seems to emphasise the inherent difference between referrals and those areas of legislative power conferred on the Commonwealth directly by section 51. For all the familiar rhetoric about treating the former as a simple addition to the latter, Kirby J’s opinion stressed that the enlargement of federal power via a referral involves an act of accommodation by the States rather than a bestowal from the Constitution’s terms and of its own force. Certainly, it is hard to disagree that this different genesis renders inapt any attempt to extend to the construction of referrals the orthodoxy set forth by Justice O’Connor early in the Court’s life that the broader interpretation of expressions in the Commonwealth Constitution is generally to be preferred over the narrow. It is not simply a matter of referrals falling within the qualification O’Connor J was happy to admit when context demanded. Commonwealth power acquired through State references is ultimately power derived in such distinctive circumstances that a suitably different approach to its reading must necessarily follow. That approach must be to always interpret referrals strictly rather than expansively even if this runs counter to how the scope of Commonwealth power is normally viewed by the Court and even inconveniences efforts at co-operative federalism.
At rather greater length, Justice Kirby justified a narrow reading of the referral due to the ‘principle of legality’, the interpretative presumption that legislation is not intended to curtail common law rights or contravene international human rights standards. Choosing to rise above recent philosophical and legal debates about the complex interrelationship between security and liberty, Kirby J simply asserted that because the Victorian Act referred power with respect to terrorism, and since ‘counter-terrorism legislation, of its nature, seriously diminishes liberty’, then the referring Act should be construed narrowly. Interestingly, Kirby J did not appear to confine this approach to instances where the referral takes the form of detailed legislative text, and not a simple statement of subject matter. On the contrary, his Honour strongly suggested that the principle of legality was equally applicable to the interpretation of both forms of referral. This seems problematic since the coercive or intrusive qualities of Commonwealth laws which result from a subject matter referral must be a matter of speculation until the referral is utilised by the national legislature. How can what results from the referring statute be appropriately considered in the construction of its provisions, let alone used to sustain a narrow reading of them? These evident difficulties of applying the principle of legality to referrals are more, not less, distinct in respect of the interpretation of any neutral term they may employ – such as ‘amendment’, the focus of attention in Thomas itself.
The only other member of the Court in Thomas to consider the reference issue in any detail was Hayne J, but his Honour did not elaborate on the extent to which referring Acts were subject to special rules of statutory construction given their status as facilitations of Commonwealth legislative power.
B Amendment or Addition?
When the States refer a subject-matter to the Commonwealth with nothing more, they are presumably content for it to legislate on that topic as it sees fit and also, while staying within the scope of that matter referred, to make such subsequent amendments or modifications to the resultant law as it desires. Such an open-ended reference is really only to be expected when the States neither care to exercise any future power in the area nor anticipate any prospect of Commonwealth use of the referral hurting their interests. As such occasions are understandably rare, it is far more common, at least in contemporary times, for the States to refer legislative text agreed upon with the Commonwealth and then make a separate referral of the power to amend that text. Ideally the latter referral should not operate as a back door through which the Commonwealth is easily able to evade the constraints which are so evidently intended by the precision of the former reference of textual provisions.
As indicated in the preceding section, a particular formula of drafting along these lines has been employed in respect of major referrals of State power in recent times – dealing with corporate regulation and the prevention of terrorism. This received the scrutiny of Kirby and Hayne JJ in Thomas which concerned a challenge to the addition by the Commonwealth to Part 5.3 of the Criminal Code of a new Division 104 providing for the making of control orders against persons where this would prevent a terrorist act from occurring. Mr Thomas’ legal representatives submitted that this scheme, under which an order had been made against their client, was outside the scope of what had been referred by the States. As these provisions have received judicial attention they will be the primary vehicle for discussion, but it should be noted that the reference in respect of corporations was made in virtually identical terms and the arguments below are of generic application, despite any specific contextual differences which might exist between different referrals.
Section 4 of the Referring Act provides:
(1) The following matters are referred to the Parliament of the Commonwealth –
(a) the matters to which the referred provisions relate, but only to the extent of the making of laws with respect to those matters by including the referred provisions in the Commonwealth Criminal Code in the terms, or substantially in the terms, of the text set out in Schedule 1; and
(b) the matter of terrorist acts, and actions relating to terrorist acts, but only to the extent of the making of laws with respect to that matter by making express amendments of the terrorism legislation or the criminal responsibility legislation.
The ‘referred provisions’ of s 4(1)(a) is the body of legislative text found in Schedule 1 of the Act which the Commonwealth is thus empowered to insert in the Code. In sub-s (3) it is stated that the ‘operation of each paragraph of subsection (1) is not affected by the other paragraph’ – in other words that the ‘initial’ and ‘amending’ references are to be understood independently of each other. In sub-s (4), it is recognised that the Commonwealth may make amendments to the text using its legislative powers aside from any dependent upon the references.
Whether the addition of Division 104 could be supported by use of the power referred in s 4(1)(b) hinged on whether it was within the qualification – is it an ‘express amendment’ of the terrorism legislation referred via the Schedule? This term is defined by s 3 of the Referring Act:
express amendment of the terrorism legislation or the criminal responsibility legislation means the direct amendment of the text of the legislation (whether by the insertion, omission, repeal, substitution or relocation of words or matter) by Commonwealth Acts, but does not include the enactment by a Commonwealth Act of a provision that has or will have substantive effect otherwise than as part of the text of the legislation.
Hayne J observed that the two parts of the definition of ‘express amendment’ appeared to be ‘contradictory’. Disinclined to interpret the provision as effectively negating any operation which s 4(1)(b) might have as a separate and additional referral from the State to the Commonwealth beyond the text in the Schedule, his Honour accepted the Commonwealth’s argument that the reference would allow the insertion of any new matter falling within the description of a law with respect to ‘terrorist acts, and actions relating to terrorist acts’ so long as ‘that is done by express amendment to the law that was enacted in the form of the scheduled text’. While that removes the ‘contrariety between the two parts of the definition of “express amendment”’, one may be forgiven for finding it a strangely formalistic result. On this reading, the Referring Act firstly provides a set text of provisions which the referred power is to support as a Commonwealth enactment, before proceeding to grant an unlimited discretion to otherwise legislate on the ‘matter of terrorist acts’ accompanied by a requirement only that this must occur ‘as part of the text’ specifically referred. Hayne J supported his interpretation by pointing to the injunction of 4(3) of the Referring Act to read the two referrals in the sub-ss of s 4 separately, but it is arguable that to do so is destructive of the limited nature of the initial reference.
Kirby J took the opposite approach and refused to accept Hayne J’s conclusion that as a result of s 4(3), it followed ‘that “the matter of terrorist acts, and actions relating to terrorist acts” referred to in a 4(1)(b) is not to be read as confined by reference to the particular provisions set out in the scheduled text’. In doing so, his Honour pointed to a further provision, defining the phrase ‘terrorism legislation’ used in both s 4(1)(b) and the definition of ‘express amendment’. As provided in s 3, ‘terrorism legislation’ referred to ‘the provisions of part 5.3 of the Commonwealth Criminal Code enacted in the terms, or substantially in the terms, of the text set out in Schedule 1 and as in force from time to time’. Regard to this provision strengthens the argument that the legislative text anchors not simply the initial reference but also the accompanying amendment reference. On this view, the latter was not intended to allow significant additions to or departure from the anti-terrorism laws as they then stood. Division 104, introducing a scheme of civil preventative orders imposing an array of possible restrictions upon individuals not charged with, nor even necessarily suspected of, criminal activity can certainly be starkly contrasted with the substance of Part 5.3 supported by the earlier text reference.
On his construction of how the provisions were to be read, Kirby J concluded that to be valid under s 51(xxxvii), Division 104 must:
constitute an express amendment of the ‘terrorism legislation’, which is defined as that enacted in the terms, or substantially in the terms, of the text set out in Sched 1 of the Referring Act. Secondly, the express amendment must be a ‘direct amendment’ of the ‘terrorism legislation’, as so defined. Although this may include the ‘insertion’ of text, that term should be construed ejusdem generis with the preceding words ‘direct amendment’, read together with the requirement that the amendment be to the ‘terrorism legislation’. This requires that a more restrictive meaning be given to the term ‘insertion’.
While this narrow reading of the power to ‘amend’ might be seen as convincing on a contextual level, it clearly suffers from a near fatal weakness. It constrains the Commonwealth’s power to amend, but does so without a referable standard. Giving the power to insert ‘words or matter’ into referred legislative text a ‘more restrictive meaning’, leaves us with uncertainty as to the scope of the amending reference with all the potential this carries for legislative paralysis and instability. Although in Thomas Kirby J makes a good case that the new Division 104 which the Commonwealth inserted into Part 5.3 went beyond being simply a natural extension of the initial scheme, others may have disagreed and almost certainly future cases would be likely to be less, rather than more, clear cut. The strong attraction of the more literal interpretation of Hayne J of the same provisions is the avoidance of this uncertainty.
Oddly, the Commonwealth Attorney-General’s Department now seems to regard its power to amend legislation enacted pursuant to the initial textual reference as severely constrained. In March this year, the Attorney-General justified the omission from the National Security Law Amendment Bill 2010 of amendments to Divs 101 and 102 of Part 5.3 of the Criminal Code which had been proposed by an earlier Discussion Paper, by saying they could not be made without the States first amending their referral legislation. This insistence that the Commonwealth cannot amend its law until the States enact an amendment of their referral is not supported by the relevant intergovernmental agreement, the power conferred on the Commonwealth to make an ‘express amendment’ of the initial text, nor the requirements of s 100.8(2) of the Code for State approval of the making of an ‘express amendment’ (discussed in the next section). Prior to the enactment of Anti-Terrorism Act (No 2) 2005 the States were not required to (and nor did they) amend their referral legislation so as to allow the Commonwealth to insert Div 104 – which, as has been discussed, was a substantial new addition to the existing law, rather than the fairly modest amendments to the definition of ‘terrorist acts’ and introduction of a new hoax offence which had been proposed by the government in its Discussion Paper but which it has now omitted from the new Bill. Lastly, a requirement of State legislative amendment as a precondition to the making of changes to the text of the initial referral which even Kirby J would have been likely to view as examples of ‘direct amendment’ of the ‘terrorism legislation’, substantially defeats the utility and certainty of State referrals under s 51(xxxvii) by reducing the co-operative endeavour to an agreement to pass and update mirroring legislation.
C The Viability of Intergovernmental Agreements as a Control on Amendment
In light of the above, it might be argued that the preferable course is for the Commonwealth and the States to make their own arrangements in order to settle when an insertion of new material to a referred law would constitute an acceptable ‘amendment’, thereby relieving the judiciary of the difficult task that Kirby J’s solution would require of them. On several occasions before his appointment as Chief Justice, Robert French endorsed a particularly strong form of this:
A mechanism by which referring or adopting States may deter the Commonwealth from non-consensual amendment would be to make the referral or adoption subject to a condition that it would be revoked in the event that the law were amended otherwise than in accordance with some agreed mechanism for obtaining consensus.
The Victorian Referring Act (and other State equivalents) in issue in Thomas did not attempt to limit the referral of power in this way – possibly due to the States’ failure, when negotiating the earlier corporations law reference, to persuade the Commonwealth to accept a referral similarly conditional on not being used for particular purposes. Instead, section 100.8 of the Criminal Code in the legislative provisions enacted by the Commonwealth and appended to the Referring Acts purports to require the approval of ‘a majority of the group consisting of the States, the Australian Capital Territory and the Northern Territory’, but at least four states, to any ‘express amendment’ of Part 5.3. Such a mechanism is not an unfamiliar element of the intergovernmental agreements that underlie co-operative schemes, but is not usually carried over to the resulting legislation.
The validity or otherwise of such a requirement will be discussed below, but it is clearly a significant piece of the legislative scheme – arguably the cornerstone of the whole dilemma and one which makes the purpose behind some of its thornier provisions rather clearer. In particular, s 100.8 makes apparent that the qualification in the definition of ‘express amendment’ (that it ‘not include the enactment by a Commonwealth Act of a provision that has or will have substantive effect otherwise than as part of the text of the legislation’) serves a purpose beyond mere formalism. The apparent oddity of the amending reference imposing a limit focused on the location of its exercise but not, effectively, as to substance (‘the insertion, omission, repeal, substitution or relocation of words or matter’) makes rather more sense when seen as designed to ensure the approval process of s 100.8 is not evaded. That prospective mechanism provides the means to resolve uncertainty over which express amendments to the terrorism legislation are permissible and which constitute a deviation by the Commonwealth unacceptable to the States. It has obvious advantages over judicial attempts to line draw in an attempt to protect State interests after the Commonwealth has purported to use the amending reference. An appreciation of s 100.8 means that the interpretation of the Referring Act offered by Hayne J not only has the benefit of certainty – it also lets the parties to the underlying intergovernmental agreement control the use of the reference.
Although it is tempting to insist that the governance arrangements underpinning a co-operative scheme should be ‘enshrined in the national law...in a legally enforceable form that in the event of a breach will trigger legal consequences’, there are substantial constitutional difficulties with the statutory recognition of controls agreed upon in the political deal-making which has produced the intergovernmental agreement. In Thomas, Kirby, Hayne and Callinan JJ were all of the view that s 100.8 of the Criminal Code was invalid – although each for different reasons. Kirby J objected to the provision’s apparent attempt to substitute the approval of the executive members of State and Territory governments for the power of the State legislatures to refer matters under s 51(xxxvii) of the Constitution. Hayne J thought s 100.8 was invalid as a federal law fettering the future actions of the Commonwealth Parliament, while Callinan J focused on the section’s purported subjugation of a State parliament’s powers to the decision of a majority of other States ‘and indeed also the Territories’.
Hayne J presented his objection as the fundamental one rendering consideration of the others unnecessary, but so far as it applies to Commonwealth amendment (distinct from partial or total repeal of a law enacted under s 51(xxxvii)) using the amending reference, it warrants a closer examination. For while his Honour’s complaint against s 100.8 that a federal provision stipulating that ‘certain amending laws may be made only if prior approval is given’ is a fetter imposed by the Parliament on ‘the future exercise of its legislative powers’ reflects the orthodoxy of parliamentary sovereignty, it might be said to acknowledge inadequately the unusual nature of the particular power in question. Given that the power to amend legislative text referred by the States may itself be the subject of a referral, surely the Commonwealth’s legislative capacity is both more complex, and inherently less free, than when utilising the other powers expressly conferred upon it by the Constitution? A possible response to that view is, however, that making the referral conditional in the way favoured by French CJ recognises the unique qualities of the power, while managing to avoid the constitutional difficulty identified by Hayne J of attempting to give direct statutory force to a pre-enactment approval mechanism agreed upon by the parties to the intergovernmental agreement. The States would still retain de facto control of the use by the Commonwealth of the reference, and may presumably still be consulted by the Commonwealth in advance of any proposed changes to the law, without the federal Parliament enacting a provision which represents a formal curb on its own legislative capacity, such as s 100.8.
IV TERMINATION OF COMMONWEALTH LEGISLATION ENACTED PURSUANT TO S 51(XXXVII)
The ambiguities surrounding the termination of Commonwealth laws made pursuant to a State referral have been more regularly canvassed in literature on s 51(xxxvii) than those concerning amendment. That said, judicial guidance has been vague and although there is a contemporary consensus that a State referral can indeed end, disagreement lingers as to precisely under what set of circumstances this may occur and the consequences of it doing so.
A The Commonwealth’s Power of Repeal
A preliminary issue which brooks so little debate as to be almost invisible is that the Commonwealth can, of course, repeal any law it has earlier enacted which relies upon a referral from the States. Although in the preceding part, the ability of the States to constrain the Commonwealth’s power to amend such legislation was considered at some length, there can be no doubt that the Commonwealth Parliament is not obliged to maintain a law simply because the State referral still stands. In this sense, there really is no distinction between s 51(xxxvii) and the other placita – they exist for the Commonwealth Parliament alone to use or not as it wishes. Justice Hayne was undoubtedly correct in Thomas when he objected to the attempt of s 100.8 of the Criminal Code to require State and Territory approval of any ‘express amendment’ of the law by the Commonwealth – but especially so far as that phrase is defined to include repeal of the law. Additionally, the one condition which the States surely cannot attach to a referral must be any attempt to block Commonwealth repeal of the resultant law. Whether it be expressed as a guarantee of the duration of the resulting scheme or the imposition of an approval process before such a step is taken, such a condition, while feasible in an intergovernmental agreement, must simply be beyond the constitutional powers of State legislatures, as an invalid attempt to curtail the Commonwealth’s legislative capacity under s 51(xxxvii).
The position with respect to partial repeal is, however, more complex. Partial repeal of the Commonwealth Act may well take the law outside the scope of the initial referral of legislative text by the States. It will be recalled that such references are generally expressed as enlivening s 51(xxxvii) to ‘the matters to which the referred provisions relate, but only to the extent of the making of laws with respect to those matters by including the referred provisions in Acts enacted in the terms, or substantially in the terms, of the tabled text’. Regardless of the current practice of insisting that the operation of the accompanying amending reference (the power to make ‘express amendments’) is independent from that of the initial referral, partial repeal of the referred text (as opposed to substantial additions of the kind at issue in Thomas) may conceivably result in a clear departure from the terms of the initial reference. So while as a matter of principle the Commonwealth must be free to legislate and then later repeal an enactment based on the referral, qualms may legitimately exist were it to act selectively through partial repeal so as to produce a law substantially distinct from that to which the States gave their initial imprimatur.
B Withdrawal of Referral - ‘Determination’ and ‘Revocation’
In Part II, reference was made to the High Court’s confirmation in Public Vehicles Licensing that State referrals could be limited to ‘a time which is specified or which may depend on a future event even if that event involves the will of the State Governor-in-Council and consists in the fixing of a date by proclamation’. At the same time, the Court declined to address the capacity of State legislatures to repeal a referring Act. Although on no occasion has a clause allowing determination by the State Executive been used to withdraw a referral upon which Commonwealth law was reliant, its recognition as a valid mechanism by which the State is able to withdraw, prompts the question why legislative revocation of the reference would not also be allowed? Not only would the power of the States to repeal a referring Act accord with the doctrine of parliamentary sovereignty, but it would also seem consistent with the overriding purpose of the section to permit the flexible redistribution of power.
While conceding that ‘it appears likely that a State could validly revoke its reference’, Twomey has argued that providing the executive with an express power to determine the referral by proclamation avoids problems that some have argued a simple repeal of the referring enactment might face under s 109 of the Commonwealth Constitution. But that latter position cannot be allowed to stand as any consideration. First, the suggestion that State legislation repealing an earlier referral might be invalid due to inconsistency with a Commonwealth law passed pursuant to the referral confuses the two subject-matters of the different laws. The Commonwealth law deals with the subject matter which is referred (eg terrorism, corporations, etc), while the State law concerns the referral itself. Second, even assuming s 109 is a problem, it is not evaded through use of the Governor’s proclamation as an ‘event’ the occurrence of which determines the life of the State referral since the source of the Governor’s power still clearly arises from State legislation. Thus, the attachment of a condition to the referral enabling its determination by executive order may not be legally significant for any reason other than it has been, unlike statutory revocation, unanimously recognised by the High Court in Public Vehicles Licensing as at least a valid mechanism by which a referral may be brought to an end.
Interestingly, the Convention Debates contain no mention of the possibility that the States could make a referral determinable by the passage of time or some other condition. Instead, the focus was squarely on the States’ capacity for revocation. The framers disagreed over whether referrals would be revocable, but it appears that a belief that they would not be was integral to support for the inclusion of s 51(xxxvii) in the Constitution. Both Deakin and Isaacs argued strongly against allowing State revocations, with the transcript ambiguous as to whether Barton also agreed. Quick and Symon, in contrast, thought that referrals would be revocable, but this led them both to argue the placitum should be ‘struck out altogether’ from section 51. Tellingly, the discussion ended inconclusively with Glynn attempting to reconcile the purported inability to revoke a reference with the constitutional rule that prevented State Parliaments from abrogating their own powers.
That last difficulty features more prominently in the limited judicial attention to this question. In South Australia v Commonwealth, Latham CJ had affirmed that ‘a State Parliament could not bind itself or its successors not to legislate upon a particular subject matter, not even, I should think, by referring a matter to the Commonwealth Parliament under sec. 51(xxxvii) of the Constitution’. He again gave voice to that view when the referrals power first arose for direct consideration in Graham v Paterson, with support from Webb J. Unanimously, the Court in Public Vehicles Licensing stated that it is ‘the general conception of English law that what Parliament may enact it may repeal’. Just a month earlier, in the First Airlines Case, Taylor J made it clear that it was incongruous to forbid revocation after the Court had upheld determination of a referral made conditionally. Windeyer J’s response to that was to seize upon the significance of an absence of conditions upon a referral:
I entertain a serious doubt whether a reference could be for an indefinite period terminable by the State legislature...If a matter be referred by a State Parliament, that matter becomes, either permanently or pro tempore, one with respect to which the Commonwealth Parliament may under the Constitution make laws. If the Commonwealth Parliament then avails itself of the power, it does so by virtue of the Constitution, not by delegation from, or on behalf of the State Parliament. It is not exercising a legislative power of the State conferred by a State Parliament and revocable by that Parliament. It is exercising the legislative power of the Commonwealth Parliament conferred by s. 51 of the Constitution.
However, this passage ignores the unique qualities of s 51(xxxvii). While that power itself cannot be affected by the States, it nevertheless has no operation independent of the referral – ‘the State Act is the material on which the power operates’. The tendency to overlook that and instead emphasise only the integrity of the Commonwealth’s legislative power was also observed in that part of Hayne J’s judgment in Thomas discussing the validity of State efforts to control the Commonwealth’s power of later amendment of the legislation. It would also seem central to the views of the present Chief Justice on the status of a Commonwealth law after revocation of the initial referral, which are discussed in the next section.
Commentary on s 51(xxxvii) features opinions on both sides of the debate about the permissibility of revocation, though those against it are mainly found in speculation preceding the Court’s endorsement of a referral determinable by condition in Public Vehicles Licensing or have mistakenly invested significance in s 109 as an obstacle. As such, it is probably overstating the case to say that the debate is in any sense a live one. Not only is this because support for the possibility of revocation has the upper hand but also because persistent State wariness on the issue ensures that referrals will continue to be made subject to conditions allowing their determination. Consequently, it is hard to envisage the States ever needing to resort to outright revocation of a referral – thus ironically robbing the question of any chance of conclusive determination.
One further consideration should be briefly mentioned. If, as is argued here, States retain the power to revoke a referral, it may be necessary to guard against the possibility of implied repeal of the referring Act, or part of it, by a later State law inconsistent with the initial reference. On the whole such a prospect seems unlikely – it is difficult to see how a State Act on the same topic as that which has been the subject of an earlier referral could, without more, effectively negate the operation of that prior enactment’s deliberate conferral of authority upon the Commonwealth. But that awkward, if remote, possibility might be best forestalled by the adoption of a simple presumption against implied repeal of a law which refers subject-matter or legislative text for the Commonwealth to enact pursuant to s 51(xxxvii).
C The Consequences of Termination
A degree of ambiguity may also be said to cloud the consequences for Commonwealth law of a referral’s termination, whether determined by time or the occurrence of an event specified in the referral or revoked by State repeal. However, there is substantial support for the view that the Commonwealth legislation must be rendered inoperative, if not invalid, once the referral has terminated. Any suggestion that the contrary might be so was memorably derided as arguing for ‘a miracle of legal levitation or a kind of constitutional variation of the fabulous Indian Rope Trick’.
The few adherents of the ‘Rope Trick’ interpretation of s 51(xxxvii) have never managed to furnish a convincing argument in its favour. At the 1898 Convention, Symon said that if the Federal Parliament has acted pursuant to a reference by legislating, ‘then...that legislation becomes federal legislation, and could not be revoked or interfered with in any way by the State’. But revocation simply is not repeal of Commonwealth law by a State. Instead it removes the ‘essential basis’ for the Commonwealth law in a way which bears strong similarity to the waxing and waning of the power with respect to defence in s 51(vi). As Lane said, with the referral removed, ‘the Commonwealth Act loses its point of departure’.
It would render pointless the accepted validity of references conditioned by time or the happening of some other event if the Commonwealth law survived the eclipse of the former. In the First Airlines Case, Windeyer J suggested as much when he said that ‘[a]ny law made by the Commonwealth Parliament, could, I consider, only operate for the duration of the period of the reference’. It will be recalled that his Honour distinguished indefinite references and argued that these were irrevocable. But if that view is incorrect, as discussed in the preceding section, then logically his remarks concerning determinable references must extend to the general case.
Extra-judicially, the present Chief Justice has taken what Pamela Tate SC
describes as the ‘brave’ position, that the
Commonwealth law made
under s 51(xxxvii) remains valid upon the reference’s
2003 and 2007, French J argued this was so because ‘there is nothing in
the grant of the power which makes the laws under
it self-terminating upon
revocation of the
Justice French said the way to ensure a Commonwealth law made under a referral
does not outlive the referral itself is the inclusion
in that law of a
‘self-executing termination provision operative upon revocation of the
referral by a referring
It is not clear what support – textual or otherwise – may be found
for insisting that without such a clause, Commonwealth
laws persist in operation
once the referral is removed. As Mr Aickin QC argued before the Court in
Public Vehicle Licensing:
A statute which is perpetual in terms must be read as intended to operate so
long as its constitutional foundation exists...If it
is seeking to regulate
things irrespective of the subsistence of the reference it will be invalid. But
from ordinary principles of
construction one would treat it as intended to
operate only so long as the constitutional foundation was
If the Commonwealth did not possess the legislative capacity to enact the particular law prior to the referral, from where does its capacity stem upon the revocation of that referral? By (admittedly unlikely) analogy, if as a result of a successful referendum, section 51 was amended by the deletion of some other head of power, say s 51(vii), would legislation earlier enacted using the now defunct power continue without falter? It cannot be reasonable to suggest that question must be answered positively unless the relevant laws contained a clause specifically providing for their termination in such an event. There seems little appreciable difference between that scenario and termination of a reference utilised by the Commonwealth under s 51(xxxvii). If anything, the case for collapse of the Commonwealth law is stronger when the State referral expires or is retracted since in this instance the Commonwealth was always legislating using power which relied on it ‘borrowing’ from the States.
It is telling that, when leaving unanswered the question of revocability of referrals under s 51(xxxvii), the High Court declared ‘[i]t forms only a subsidiary matter which if decided might throw light on the whole ambit or operation of the paragraph’. A conclusive determination of that issue probably does depend on articulating a cohesive theory as to the meaning and scope of the power more generally. Although it may be argued that the opportunity has not arisen which would necessitate this, in part the wary use of the referrals power is due to the Court’s apparent reluctance, on those few occasions when it has had the chance, to meaningfully explore the contours of the placitum. But, as was argued at the start of this article, the centrality of the power in national schemes and its likely importance to future attempts at substantial federal reform, mean that this ambiguity is less tolerable than it once was.
What has been singularly lacking from most judicial consideration of the power has been clear acknowledgement of its distinctive characteristics as a facilitator of co-operative federalism. That might be thought to result in an interpretation which, even after a referral is made, accords significance to the role that has been played by the States. Due recognition of the continued interest which the referring States have in the subject-matter they have made available for the operation of Commonwealth legislative power, has the potential to affect the construction of both the initial referral and the extent to which the accompanying amending reference may be used to depart from it. Similarly, the capacity for States to revoke a reference and the consequences of termination more generally must take their cue from reading the power in this way.
The views of the present Chief Justice illustrate a particularly sharp, and rather curious, inconsistency in this respect. Extracurially he has posited that ‘the Constitution, while marking out the boundaries of legislative power between the components of the Federation, rests upon an assumption of cooperation between them’,  and has highlighted the reference power as ‘an instrument of co-operative constitutional evolution’ to this end. Yet, as discussed, French CJ has argued that Commonwealth laws survive the termination of a State referral, unless they make specific provision to the contrary. Not only is that view inherently difficult to appreciate as a matter of legislative power, it seems distinctly at odds with reading s 51(xxxvii) in a manner which respects it as a truly ‘co-operative mechanism’. While a surrendering of control by the States is implicit in the use of s 51(xxxvii) to produce what is a Commonwealth law, there seems little reason that should be either total or permanent if federal ‘co-operation’ is the guiding principle.
The issue of revocation has been, in any case, reduced to essentially an
academic question. Instead, the aspect of s 51(xxxvii) which is of great and
practical importance is the capacity of the Commonwealth to amend laws made
under the power and the ability
of the States to constrain this in order to
preserve the limits placed on the initial reference. The deficiencies of the
formulation which is presently favoured were made fairly apparent by
both Justices Kirby and Hayne in Thomas, but this is not to suggest that
any alternative approach is easily determined. The manifest failure in that case
of s 100.8 as a
legislative mechanism reflecting the States’ rights under
the intergovernmental agreement is ample demonstration of the complexities
involved in attempts to directly secure their position in the kind of
co-operative exercise facilitated at by s 51(xxxvii).
Clarification of these outstanding issues should ideally occur in the context of a comprehensive judicial exposition of the referrals power as a whole. In the meantime, the attractions of the power will presumably see it continue to support major national legislative schemes. That increased reliance can only hasten the likelihood that the Court will be required to meaningfully engage with s 51(xxxvii). It is to be hoped that the unique properties and purpose of the power are not lost sight of when that opportunity arises.
[*] Associate Professor and Director, Gilbert + Tobin Centre of Public Law, UNSW. I thank David Hume for his excellent research assistance in the preparation of this article. I also thank the two referees and my colleagues in the Gilbert + Tobin Centre, especially Sean Brennan, for their suggestions on an earlier draft of this work. As ever, I alone am responsible for any flaws.
 Corporations Act 2001 (Cth) (‘Corporations Act’); Australian Securities and Investments Commission Act 2001 (Cth).
 Criminal Code 1995 (Cth) pt 5.3 (‘Criminal Code’).
 Fair Work Act 2009 (Cth).
 Graham v Paterson  HCA 9; (1950) 81 CLR 1, 19 (Latham CJ), likening a referral to simply adding a new placitum to s 51 of the Constitution.
 See Ross Anderson, ‘Reference of Powers by the States to the Commonwealth’ (1951-53) University of Western Australia Annual Law Review 1, 6; Robert S French, ‘The Referral of State Power’  UWALawRw 2; (2003) 31 University of Western Australia Law Review 19, 31.
 Anderson, above n 7, 8. Buchanan says the Commonwealth’s power ‘springs from two sources’ before going on to emphasise that the State Act ‘provides a basis or field for the operation of the power contained in s 51(xxxvii)’: P Buchanan, ‘The Queen v Public Vehicles Licensing Appeal Tribunal of Tasmania; Ex parte Australian National Airways Ltd – Case Note’ (1964-65) 1 Federal Law Review 324, 326.
 Criminal Code sch 1, s.100.8.
  HCA 33; (2007) 233 CLR 307 (‘Thomas’).
 Kartinyeri v Commonwealth  HCA 22; (1998) 195 CLR 337,  (Brennan CJ and McHugh J);  (Gaudron J);  (Gummow & Hayne JJ).
 Anderson, above n 7, 8; Graeme Johnson, ‘The Reference Power in the Australian Constitution’  MelbULawRw 4; (1973-74) 9 Melbourne University Law Review 42, 73-74.
 In this sense, the significance of s 51(xxxvii) is not entirely straightforward. In removing any ‘lingering constitutional uncertainty’ (Daryl Williams and James Renwick, ‘The War Against Terrorism: National Security and the Constitution’ (Summer 2002/2003) Bar News: Journal of the NSW Bar Association 42, 43), referrals may be instrumental in the Commonwealth moving forward with confidence to legislate in an area which the High Court may subsequently confirm that it actually held sufficient power of its own accord: such was the case with terrorism control orders and the decision in Thomas  HCA 33; (2007) 233 CLR 307.
 Craven, above n 4, 285; Anne Twomey, ‘Reforming Australia’s Federal System’ (2008) 36 Sydney Law Review 57, 69-71.
 See John Quick & Robert Garran, The Annotated Constitution of the Australian Commonwealth (1901) 648-649. The original of these was contained in the 1849 report of the Committee of the Privy Council recommending the creation of a General Assembly of the colonies. Amongst its proposed powers, the Assembly was to be able to enact laws in relation to otherwise unmentioned matters ‘on which the General Assembly should be desired to legislate by addresses for that purpose presented to them from the legislatures of all those colonies’: Privy Council, Papers Relative to the Proposed Alterations in the Constitution of the Australian Colonies, reproduced in John Williams, The Australian Constitution – A Documentary History (2005) 7-8. Unsurprisingly, given the nascency of the Federation movement, similar clauses feature in proposals made in the following decade in both New South Wales and Victoria to establish a colonial assembly: Williams, 4.
 See French, above n 7, 25; Johnson, above n 13, 43.
 Bill to Empower the Legislatures of the Australian Colonies to Form a Federal Assembly (NSW) 1857, cl 2.
 French, above n 7, 34, who finds the prospect ‘not particularly edifying’.
 Ibid 34.
 Official Record of the Debates of the Australasian Federal Convention, Melbourne, 27 January 1898, 221 (Richard O’Connor and Sir John Downer, respectively) (‘Federal Convention Debates’).
 Ibid 222 (Isaac Isaacs).
 Ibid 220 (Sir John Downer).
 French, above n 7, 34.
 Federal Convention Debates, above n 22, 215-17 (Alfred Deakin).
 Ibid 219 (Josiah Symon).
 Ibid 221 (Sir John Downer).
 Anderson, above n 7, 6.
 Graham v Paterson  HCA 9; (1950) 81 CLR 1, 22 (McTiernan J). On this point, Johnson discussed the express limitations on Commonwealth power found in ss 92, 99 and 116 of the Constitution: above n 13, 63-4, but of course implied limitations also pertain: Thomas  HCA 33; (2007) 233 CLR 307, 462 (Hayne J).
 Federal Convention Debates, above n 22, 222 (Isaac Isaacs). See also Anderson, above n 7, 6-7.
 Johnson discusses the reluctance of the High Court in Graham v Paterson to embrace language such as ‘denominated’, ‘transferred’ or ‘delegated’ when considering the meaning of ‘referred’: above n 13, 62. Accordingly, it is rare to find use of such terminology in connection with the power. However, cf Federal Convention Debates, above n 22, 223-24 (Edmund Barton) and, presumably without significance, Thomas  HCA 33; (2007) 233 CLR 307, 510 (Callinan J).
 See, eg, Graham v Paterson  HCA 9; (1950) 81 CLR 1, 19 (Latham CJ), and also contemporary commentary in Anderson, above n 7, 5; K H Bailey, ‘Fifty Years of the Australian Constitution’ (1951) 25 Australian Law Journal 314, 335.
  HCA 9; (1950) 81 CLR 1, 22. See also  HCA 9; (1950) 81 CLR 1, 25 (Williams J). Section 107 of the Commonwealth Constitution preserved the legislative powers of colonial parliaments upon their conversion to Statehood, subject only to those exclusive powers now vested in the Commonwealth.
 Federal Convention Debates, above n 22, 221 (Sir John Downer), on the expectation that States would be reluctant to lose power via the referral mechanism.
 Anne Twomey, The Constitution of New South Wales (2004) 811; WA Wynes, Legislative, Executive and Judicial Powers in Australia (5th ed, 1976) 172.
 Pamela Tate SC, ‘New Directions in Co-operative Federalism: Referrals of Legislative Power and their Consequences’ (Paper delivered at the Gilbert + Tobin Centre of Public Law 2005 Constitutional Law Conference, Sydney, 18 February 2005) .
 R v Credit Tribunal; Ex parte General Motors Acceptance Corporation  HCA 34; (1977) 137 CLR 545, 563 (Mason J).
 See Cheryl Saunders, ‘A New Direction for Intergovernmental Agreements’ (2001) 12 Public Law Review 274, 284, discussing the scheme adopted to this end by the Corporations Act 2001 (Cth) pt 1.1A. See also Fair Work Act 2009 (Cth) pt 1-3.
  HCA 15; (1964) 113 CLR 207 (‘Public Vehicles Licensing’).
 Ibid 225.
 Commonwealth Powers (Air Transport) Act 1952 (Tas) s 2.
 Craven, above n 4, 287.
 The law under challenge in Thomas  HCA 33; (2007) 233 CLR 307 is a strong demonstration of this point.
 Craven, above n 4, 287.
 See, eg, Corporations (Commonwealth Powers) Act 2001 (NSW) s 5.
 See, eg, Terrorism (Commonwealth Powers) Act 2002 (NSW) s 5.
 Public Vehicles Licensing  HCA 15; (1964) 113 CLR 207, 226.
 Federal Convention Debates, above n 22, 224 (Edmund Barton). See also Thomas  HCA 33; (2007) 233 CLR 307  (Kirby J).
 Thomas  HCA 33; (2007) 233 CLR 307 .
 Wynes, above n 37, 171.
 See, eg, Corporations (Commonwealth Powers) Act 2001 (NSW) s 5(1). Note that this is the reverse of referring a matter subject to an executive proclamation determining the duration of the referral. That approach, discussed below in Part IV(B), is clearly constitutionally permissible.
  HCA 9; (1950) 81 CLR 1, 19.
 Charter of Human Rights and Responsibilities 2006 (Vic).
 French, above n 7, 19.
 Smith v St James (1996) 135 FLR 296, 311.
 See  HCATrans 078 at 13865-13871, referred to in Thomas  HCA 33; (2007) 233 CLR 307  (Kirby J).
 Thomas  HCA 33; (2007) 233 CLR 307 .
 Ibid  (Kirby J).
 Ibid .
 Ibid -.
 Ibid  (Kirby J).
 Ibid  (Kirby J).
 Ibid .
 Jumbunna Coal Mine NL v Victorian Coal Miners’ Association (1908) 6 CLR 309, 368.
 See Pape v Commissioner of Taxation (2009) 238 CLR 1, - (Heydon J).
 See Murray Gleeson AC, ‘The Meaning of Legislation: Context, Purpose and Respect for Fundamental Rights’ (2009) 20 Public Law Review 26, 33.
 Thomas  HCA 33; (2007) 233 CLR 307  (Kirby J).
 Ibid .
 Ibid  (Kirby J).
 Justice Robert French, ‘Horizontal Arrangements: Competition Law and Cooperative Federalism’ (Speech delivered at the Competition Law Conference, Sydney, 5 May 2007) ; French, above n 7, 33.
 The more recent referrals to extend the application of the Fair Work Act 2009 (Cth) to private sector employees of unincorporated businesses utilise a similar structure to that of the corporations and terrorism referrals. However, it should be noted that the scope of the Commonwealth’s power to act pursuant to the referrals is less ambiguously constrained due to the clear articulation by the States of lists of both those specific topics which are referred and those matters that are expressly excluded from their individual references: see Industrial Relations (Commonwealth Powers) Act 2009 (NSW), ss 3(1) (definitions of ‘excluded subject matter’ and ‘referred subject matters’) and 6.
 Thomas  HCA 33; (2007) 233 CLR 307, .
 Thomas  HCA 33; (2007) 233 CLR 307, .
 Ibid .
 Ibid .
 Additionally, Kirby J quoted the Victorian Attorney-General’s second reading speech in support of the law effecting the reference of power for Part 5.3, wherein the Minister indicated his understanding that the referral was ‘limited to only that necessary to enact terrorism offences in the same form, or substantially the same form, as the present Commonwealth terrorism offences and to amend them as required’: Thomas  HCA 33; (2007) 233 CLR 307, .
 Thomas  HCA 33; (2007) 233 CLR 307,  (Gummow and Crennan JJ).
 Ibid .
 Commonwealth, National Security Legislation -Discussion Paper on Proposed Amendments, July 2009.
 Commonwealth, Parliamentary Debates, House of Representatives, 18 March 2010, 2922 (Robert McClelland, Attorney-General).
 Commonwealth and States and Territories Agreement on Terrorism and Multi-Jurisdictional Crime, 5 April 2002, cl 3; which merely requires that ‘amendment based on the referred power will require consultation with and agreement of States and Territories’.
 Terrorism (Commonwealth Powers) Act 2003 (Vic), s 3; and Criminal Code, s 100.1, respectively.
 Quite apart from the matter of referrals, the Commonwealth seems oblivious to the consequences of the endorsement of a majority of the High Court in Thomas of its broad capacity to legislate on terrorism using its power of defence under s 51(vi).
 French, above n 7, 34; French, above n 78, .
 BM Selway QC, ‘Hughes Case and the Referral of Powers’ (2001) Public Law Review 288, 299. That impasse was subsequently broken by a deal between the Commonwealth and New South Wales and Victoria that any such limitation on the referral ‘should be contained in the non-justiciable agreement rather than the reference Act’ (at 300).
 For the negotiations on State/Territory approvals to amendments to the Corporations Law, see ibid 300.
 George Williams, Working Together – Inquiry into Options for a New National Industrial Relations System (Final Report), State Government of New South Wales, 2007, 89.
 Thomas  HCA 33; (2007) 233 CLR 307, , .
 Thomas  HCA 33; (2007) 233 CLR 307, . See Dennis Rose, ‘Uniform Personal Property Security Legislation for Australia – Constitutional issues’ (2002) 14 Bond LR 26, 29.
 Thomas  HCA 33; (2007) 233 CLR 307, . In 1986, the Advisory Council on Intergovernmental Relations recognised that majority voting controls raised concerns for the sovereignty of both individual States and the Commonwealth: see Martin Painter, Collaborative Federalism – Economic Reform in Australia in the 1990s (1998) 108.
 The statutory provisions themselves are not expressly so limited, with 100.8 imposing the need for State/Territory approval simply on the making of any ‘express amendment’ (as that phrase is defined in s 100.1 of the Code and s 3 of the Referring Act). A further acknowledgment of the Commonwealth’s capacity to amend the legislation using powers aside from s 51(xxxvii) is found in the Referring Act, s 4(4). Twomey suggests such an ambit restriction on amendment using any Commonwealth power is ineffective: Twomey, above n 37, 814, fn 102. That seems correct, but it invites speculation as to whether the States might make an initial referral conditional upon the Commonwealth using the States’ amendment reference exclusively to make subsequent changes to the resultant Commonwealth law.
 Thomas  HCA 33; (2007) 233 CLR 307, .
 See First Airlines Case  HCA 2; (1964) 113 CLR 1, 53 (Windeyer J). See also Buchanan, above n 8, 326-7.
 Gerard Carney, The Constitutional Systems of the Australian States and Territories (2006) 15. In Kartinyeri v Commonwealth  HCA 22; (1998) 195 CLR 337, , Gummow and Hayne JJ said that ‘An amendment may take the form of, or include, a repeal. Thus, if a section is deleted it can be said that it has been repealed whilst the statute itself has been amended’. Not simply any deletion will amount to a partial repeal giving rise to the problem Carney has highlighted; inevitably it will be a matter of degree.
 See Referring Act, s 4 quoted in main text above and also Corporations (Commonwealth Powers) Act 2001 (Vic), s 4(1)(a); Fair Work (Commonwealth Powers) Act 2009 (Vic), s 4(1)(a).
 See Guy Aitken and Robert Orr, Sawer’s The Australian Constitution (3rd ed, 2002) 218-20.
 Buchanan, above n 8, 327.
 Twomey, above n 37, 810. As an example of the s 109 argument itself, see Wynes, above n 37, 171-72, while more recently Saunders was content to highlight lingering uncertainty over the issue by saying there is ‘some chance that the principle of paramountcy might preclude revocation of an otherwise unlimited reference’: above n 41, 282.
 ‘The terms of section 109 would not appear to be sufficient to preclude a State Parliament from exercising its constitutional power of repealing its own legislation’: Anderson, above n 7, 9.
 Carney, above n 103, 15-16.
 As Craven said, ‘the difference between revocation and determination may not amount to much’: above n 4, 287.
 Federal Convention Debates, above n 22, 217 and 223 respectively.
 Ibid 218.
 Ibid 218 and 219 respectively.
 Ibid 225.
  HCA 14; (1942) 65 CLR 373.
 Ibid 416.
 Graham v Paterson  HCA 9; (1950) 81 CLR 1, 18 and 25 respectively.
  HCA 15; (1964) 113 CLR 207, 226.
  HCA 2; (1964) 113 CLR 1, 38. See also Anderson, above n 7, 9; Johnson, above n 13, 70.
 Buchanan, above n 8, 326-7.
 See Senex, above n 49.
 Wynes, above n 37, 171-72.
 The development of a qualified approach to the implied repeal of statutes in particular constitutional contexts is not without some precedent: see Thoburn v Sunderland City Council  EWHC 195; (2003) QB 151 (Laws LJ).
 Senex, above n 49, 326; cf Wilbur Ham KC cited in Senex, above n 49, 325.
 Federal Convention Debates, above n 22, 219.
 Anderson, above n 7, 8.
 P H Lane, above n 110.
  HCA 2; (1964) 113 CLR 1, 52-3.
 Tate, above n 39, .
 French, above n 7, 33; French, above n 78, . In the latter, French J argued that this scenario was distinct from State adoption of the Commonwealth law since ‘the reference in s 51(xxxvii) to States whose parliaments “afterwards adopt the law” arguably provides for the extension of the law to those States only during the currency of their adoption of it.’
 French, above n 78 at .
 Public Vehicles Licensing  HCA 15; (1964) 113 CLR 207, 226.
 French, above n 78, .
 French, above n 7, 21.
 Ibid; French, above n 78, .
 Saunders, above n 41, 285.
 These were succinctly stated, in the context of the corporations reference, by Ian Govey and Hilary Manson, ‘Measures to Address Wakim and Hughes: How the Reference of Powers Will Work’ (2001) 12 Public Law Review 254, 259. As argued above, even when it transpires that a reference was not actually needed (as in Thomas), the power plays a significant role in forging Commonwealth-State co-operation as a precursor to the Commonwealth law’s enactment.