• Specific Year
    Any

Blackshield, Tony --- "New South Wales v Commonwealth Corporations and Connections" [2007] MelbULawRw 42; (2007) 31(3) Melbourne University Law Review 1135

CASE NOTE

NEW SOUTH WALES V COMMONWEALTH[1]

CORPORATIONS AND CONNECTIONS

TONY BLACKSHIELD[∗]

[In the WorkChoices Case, the High Court held by a 5:2 majority (Kirby and Callinan JJ dissenting) that the Workplace Relations Amendment (Work Choices) Act 2005 (Cth), substantially reshaping the Workplace Relations Act 1996 (Cth), was valid. The substantive repeal of the older provisions based on s 51(xxxv) of the Constitution was a valid exercise of power under s 51(xxxv); their substantive replacement by new provisions based on s 51(xx) was a valid exercise of the latter power. The result entailed a clear rejection of the ‘narrow’ view of s 51(xx), and to that extent appeared to vindicate the ‘wide’ alternative view. However, the joint judgment did little to clarify the principles involved. This case note argues that despite the width now accorded to s 51(xx), the underlying rationale is still dependent on the need to establish a ‘sufficient connection’ with the business of corporations in order for legislation relying on this head of power to be valid. It also explores the possibility that the same idea of ‘sufficient connection’ might have been used to give stronger support to the dissident arguments against wholesale encroachment on the province of s 51(xxxv).]

CONTENTS

I INTRODUCTION

The 2006 result in New South Wales v Commonwealth (‘WorkChoices Case’)[2] was as widely expected as the 1990 result in New South Wales v Commonwealth (‘Incorporation Case’)[3] had been unexpected. The decision in 1990 established that s 51(xx) of the Constitution (the ‘corporations power’) does not enable the Commonwealth Parliament to enact a comprehensive regime of corporations law, including control of the actual incorporation of corporations. The decision in 2006 establishes that the ‘corporations power’ does enable the Commonwealth Parliament to enact a comprehensive regime of industrial relations law, substantially replacing the traditional regime dependent on s 51(xxxv). The Workplace Relations Amendment (Work Choices) Act 2005 (Cth) (‘WorkChoices legislation’) cut a swathe through the Workplace Relations Act 1996 (Cth) leaving little more than an empty shell, which was then refilled with the new provisions based on s 51(xx). The repeal of the older provisions based on s 51(xxxv) is now upheld — uncontroversially — as itself a valid exercise of power under s 51(xxxv).[4] What remains controversial, at least at a political and emotional level, is the use of s 51(xx) to enact the new provisions.

At that political and emotional level, the dissenting judgments are passionate in protest. Callinan J deplores a reading of s 51(xx) so expansive that it utterly destroys any semblance of ‘federal balance’;[5] Kirby J deplores the departure from our previous adherence ‘over more than a century’ to ‘an industrial process that is uniquely Australian’ — embodying not only ‘an important part of the nation’s institutional history’, but a commitment to ‘egalitarian and idealistic values’ that have hitherto ‘profoundly influenced the nature and aspirations of Australian society’.[6] Yet, however deeply we may sympathise with either or both of these protests, a jurisprudence of regret[7] is not itself a sufficient basis for constitutional argument.

On the other hand, the joint judgment of the majority[8] gives no convincing explanation of why the new regime is supported by s 51(xx). The actual decision rests solely on dicta by Gaudron J in Re Pacific Coal Pty Ltd; Ex parte Construction, Forestry, Mining and Energy Union (‘Pacific Coal’)[9] (considered below); and insofar as the judgment does attempt a discussion of the principles that might inform the interpretation of s 51(xx), much of it appears to hark back to an earlier and more primitive stage of doctrinal development.

Initially, after Strickland v Rocla Concrete Pipes Ltd (‘Concrete Pipes Case’)[10] rescued s 51(xx) from oblivion, judicial opinion polarised between a ‘broad’ and a ‘narrow’ view of its scope — the broad view extending to any law addressed to a constitutional corporation,[11] or taking the presence of such a corporation as the trigger for its operation (so that any law in the form ‘[e]very corporation shall …’ or ‘[n]o corporation shall …’ would be valid); the narrow view taking the distinctive classes of corporations singled out for inclusion (‘foreign’, ‘trading’, ‘financial’) as an index to the subjects of legislation that the power would support.[12] By the time of Re Dingjan; Ex parte Wagner (‘Dingjan’),[13] that dichotomy had been transcended: instead of pursuing the fruitless attempt to settle the ambit of the power definitively once and for all, most members of the Court were content to confine themselves to the immediate question of whether the provision there impugned had a ‘sufficient connection’ with power.

The canon of judicial parsimony underlying that approach has often been reaffirmed.[14] Indeed, the WorkChoices judgment itself effectively reaffirms such a canon,[15] and speaks at times as if the only question is one of ‘sufficient connection’ with a head of power. For example, it recites the standard collection of formulaic canons of characterisation attributed nowadays to Grain Pool of Western Australia v Commonwealth (‘Grain Pool’) — including an insistence that so long as a sufficient connection with a head of power exists, justice, wisdom and proportionality ‘are matters of legislative choice’.[16] Moreover, much of the polemic that follows appears to assume that ‘sufficiency of connection’ is itself a self‑sufficient criterion: for example, the plaintiffs are criticised for seeking to complicate it by ‘an additional filter’.[17] Yet, along with all this (and no doubt in response to the plaintiffs’ arguments), the main structure of the majority judgment reverts to the old dichotomous concern with a ‘broad’ and a ‘narrow’ view of s 51(xx) — now reformulated respectively as an ‘object of command’ test and a ‘distinctive character’ test — and is largely devoted to an extended refutation of the latter. Whether this means that the ‘object of command’ test has now prevailed will be considered below.

II COMPLICATIONS

The argument is complicated by reference to two other conceptions drawn from the decided cases. One is an attempt to confine the scope of s 51(xx) by limiting it to the regulation of activities and relationships ‘external’ to a corporation, as distinct from its ‘internal’ arrangements — a distinction proposed by Isaacs J in Huddart Parker & Co Pty Ltd v Moorehead (‘Huddart Parker’), where he treated such matters as ‘wages and hours’ for employees as questions of ‘purely internal management’,[18] and hence beyond the scope of the power. The joint judgment rejects the whole distinction: the dividing line is lacking in clarity, its consequences are lacking in logic, and the whole idea of excluding legislative interference in a corporation’s ‘internal’ affairs is an inappropriate transplant into constitutional law from the choice‑of‑law rules enjoining comity in the treatment of foreign companies.[19]

The second complicating factor is the test of ‘discriminatory operation’, originally proposed by Brennan J in Actors Equity.[20] Although Brennan J there declined to choose between the ‘broad’ and the ‘narrow’ view of s 51(xx), his approach seemed to favour the ‘object of command’ test, since it treated s 51(xx) as supporting any law that focused on corporate involvement as the discrimen triggering its application. Accordingly, the WorkChoices judgment initially assumes that ‘on its face’, the test of ‘discriminatory operation’ would be ‘satisfied by any law which singled out constitutional corporations as the object of statutory command’.[21]

In Dingjan, however, Brennan J seemed to assimilate his test to that of ‘significance’.[22] Accordingly, the plaintiffs in the WorkChoices Case adopted ‘discriminatory operation’ as an alternative but equivalent version of their own preferred ‘distinctive character’ test.

The majority judgment does not accept this equivalence. Instead, it proposes yet another interpretation, treating the test as one of ‘discriminatory operation’ — that is, as applying ‘chiefly, perhaps only’ to cases where the law applies on its face ‘to constitutional corporations and to other persons indifferently’, but its practical operation has a particular impact on its corporate subjects.[23] The majority reasons that if a law of this kind is brought within power under s 51(xx) by reason of its practical impact on constitutional corporations, a law applicable only to such corporations must be valid a fortiori: that is, the ‘object of command’ test is merely a ‘logical extension’ of the test proposed by Brennan J.[24]

Neither of these new explanations of what Brennan J meant by ‘discrimination’ seems very convincing; but the issue can be put to one side. The practical effect of the plaintiffs’ explanation was that ‘distinctive character’ and ‘discriminatory operation’ became, in effect, alternative versions of the limit which they sought to impose on s 51(xx) — and, however it is expressed, the majority judges decline to impose any such limit.

III FILTERS AND INVERSIONS

The rejection of the plaintiff’s argument rests partly on the objections that have always been raised against attempts to limit the regulatory possibilities permitted by s 51(xx) by relying on the adjectives ‘trading’, ‘financial’ and ‘foreign’: not only would this mean that different enactments relying on the one head of power would be tested by three different criteria, but the issues arising under the rubrics ‘trading’ and ‘financial’ would be different in kind from those arising under the rubric ‘foreign’. The former epithets would invite attention to the activities of corporations, the latter to their ‘status or origin’;[25] and there is ‘no immediately evident reason for … such disconformity’.[26]

Primarily, however, the majority critique of the plaintiffs’ arguments is centred on their apparent inability to accept the idea of ‘sufficient connection’ as an adequate key to characterisation, in need of no further qualification or elaboration. The argument is put in two ways: either the ‘distinctive character’ test interposes ‘an additional filter’, or it entails ‘inverting the proper order of inquiry’.[27] The objection to an ‘additional filter’ rests not so much upon any inherent vice in the content or operation of the particular filter proposed, as upon the extraneous motivation of those who propose any filter at all. Undoubtedly, those judges in earlier times who argued most strongly for a ‘narrow’ view of s 51(xx) were concerned to avoid what they perceived as the unacceptable consequences of allowing the power to expand to its full potential scope.[28] Accordingly, the WorkChoices judgment assumes that, if no such motivation were present, there could be no legitimate reason to limit the scope of Commonwealth power.

The objection to ‘inverting the proper order of inquiry’ is really the same argument in a different guise: partly it means that ‘consideration of the legal or practical operation of the law in question’[29] should precede the consideration of arguments against a sufficient connection, and partly that presuppositions about a ‘federal balance’ should not take priority over the process of construing the constitutional text.

The need to preserve the ‘federal balance’ is the primary ground on which Callinan J founds his dissent. But it presupposes that there is a balance in the federal distribution of power that is not only desirable, but ascertainable; and much of the joint majority judgment is devoted to the rejection of any such presupposition. Primarily this involves an emphatic reaffirmation of Amalgamated Society of Engineers v Adelaide Steamship Co Ltd (‘Engineers’ Case’),[30] since both of the doctrines overturned in that case — implied immunity of instrumentalities and reserved state powers — arose out of a failure to start from ‘the constitutional text, rather than a view of the place of the States that is formed independently of that text’.[31] But the joint judgment also relies on the reasoning of Dixon J in Melbourne Corporation v Commonwealth[32] — first, for his insistence that since the Constitution makes affirmative grants of legislative power only to the Commonwealth, the Commonwealth’s position ‘is necessarily stronger than that of the States’;[33] and secondly because his reasoning is said to presuppose or imply that questions about the states’ legislative powers are entirely separate from questions about their existence as bodies politic. The point here is not that concerns about federalism should be confined to questions of the latter kind, but rather that, once we accept such a separation, we are left with no criteria by which to determine what a desirable federal balance might be. Thus the supposed standard is not merely an unwarranted presupposition, but is lacking in content.

The logic is questionable, and seems to extract much more from Sir Owen Dixon’s reflections than he ever intended; but the conclusion is hardly open to debate. Unless we abandon the Engineers’ Case by reverting to the discredited doctrine of reserved state powers (which Callinan J denies he is doing),[34] there is no apparent way of drawing a line between state and Commonwealth legislative powers such that any Commonwealth trespass over that line can objectively be criticised as disturbing ‘the federal balance’. Indeed, when Callinan J seeks to draw such a line, he appears to fall into precisely the error of confusing questions about legislative power with questions about the states’ existence.[35]

The denial of any legitimate place for an ‘additional filter’ is also largely directed to concerns about disruption of the ‘federal balance’, since that is one kind of unacceptable consequence that led earlier judges to seek a limit on the scope of s 51(xx). But, of course, a more specific kind of unacceptable consequence was envisaged, long before the Engineers’ Case, by the catalogue of extreme possibilities offered by Higgins J in Huddart Parker,[36] and memorably summed up by P H Lane as ‘Higgins’ list of horribles’:

If the argument for the Crown is right, the results are certainly extraordinary, big with confusion. If it is right, the Federal Parliament is in a position to frame a new system of libel laws applicable to newspapers owned by corporations, while the State law of libel would have to remain applicable to newspapers owned by individuals. If it is right, the Federal Parliament is competent to enact licensing Acts, creating a new scheme of administration and of offences applicable only to hotels belonging to corporations. If it is right, the Federal Parliament may enact that no foreign or trading or financial corporation shall pay its employés less than 10s per day, or charge more than 6 per cent interest … If it is right, the Federal Parliament can enact that no officer of a corporation shall be an Atheist or a Baptist, or that all must be teetotallers.[37]

The WorkChoices majority dismisses this kind of apprehension as well. The prospect of ‘possible social consequences’ perceived as ‘absurd or inconvenient’ is no ‘reason to confine the reach of the legislative power’, which ‘should not be given a meaning narrowed by an apprehension of extreme examples and distorting possibilities’.[38] The point is reinforced by invoking Leslie Zines’ insistence[39] that powers should not be ‘construed restrictively’ in order to avoid ‘socially bad or downright ridiculous’ results. It could of course have been reinforced by numerous dicta in other cases[40] — and ultimately by the Engineers’ Case itself, with its insistence that the ‘extravagant’ use of power ‘is a matter to be guarded against by the constituencies and not by the Courts’.[41] Arguably, in the case of the WorkChoices legislation, this is precisely what happened at the federal election of 24 November 2007.

IV PACIFIC COAL

All this, however, is only negative — a philippic on why the ‘distinctive character’ test should not be accepted. For the positive basis on which the majority upholds the legislation as valid, we must return to the dicta of Gaudron J in Pacific Coal.

In the passage initially quoted in the WorkChoices Case, Gaudron J had said:

I have no doubt that the power conferred by s 51(xx) of the Constitution extends to the regulation of the activities, functions, relationships and the business of a corporation described in that sub‑section, the creation of rights, and privileges belonging to such a corporation, the imposition of obligations on it and, in respect of those matters, to the regulation of the conduct of those through whom it acts, its employees and shareholders and, also, the regulation of those whose conduct is or is capable of affecting its activities, functions, relationships or business.[42]

Had she stopped there, it might still have been possible to argue that her analysis did not support the validity of the 2005 legislation — since, thus far, the only reference to laws affecting ‘employees’ is to regulation of their ‘conduct’, and moreover only of their conduct ‘in respect of those matters’ already referred to. But she went on to say that the power ‘extends to laws prescribing the industrial rights and obligations of corporations and their employees and the means by which they are to conduct their industrial relations.’[43]

Clearly, this formulation is apt to describe the legislation enacted in 2005.

Kirby J protests that what Gaudron J said has been taken out of context (and technically has no authority, since she dissented in Pacific Coal).[44] Similarly, Callinan J protests ‘that it is not right to seize, and to rely conclusively, upon a pronouncement made by a dissenting judge on a point not even argued’.[45] So much may be admitted. But a further attempt by Kirby J to show that, in context, her Honour’s observations do not in fact support the WorkChoices result, does not withstand closer inspection. The issue in Pacific Coal related to the validity of the Workplace Relations and Other Legislation Amendment Act 1996 (Cth) (‘WROLA Act’, an earlier stage in the shift away from compulsory arbitration as a means for regulating industrial relations). Specifically, it related to the provision that, after a transitional period of 18 months, the content of most existing awards would cease to have effect. Kirby J points out that, when Gaudron J made the observations now adduced by the WorkChoices majority, she did so for the purpose of holding that the corporations power did not support the termination of award provisions — which for Kirby J implies that she would not have supported the WorkChoices legislation, either.[46] Yet this is far from clear. Her conclusion in Pacific Coal was simply that ‘[t]he only connection’ of the challenged provision with s 51(xx) was ‘that it may have some effect on the rights and obligations of corporations and their employees’; and this was ‘not sufficient’ to give it ‘the character of a law with respect to corporations’.[47] This was so, she said, because the provision in question ‘operates neither to prescribe the industrial rights and obligations of corporations and their employees nor to regulate the means by which they are to conduct their industrial relations.’[48]

Yet the WorkChoices legislation does purport to ‘prescribe the industrial rights and obligations of corporations and their employees’, and to ‘regulate the means by which they are to conduct their industrial relations’. In other words, it does precisely what Gaudron J thought the provision in the WROLA Act did not do.

The real problem is not that the Pacific Coal dicta do not support the majority conclusion; they do. The problem is rather that the judgment assumes that because the words are applicable the words are authoritative, thus falling into the error of mechanically applying the words of a judgment as if they were the words of a statute.[49] What is needed is some exposition of why, as a matter of principle, the view expressed by Gaudron J should be accepted as correct; and this the joint judgment fails to supply.

In its absence, Callinan J assumes that because the joint judgment so emphatically rejects the ‘distinctive character’ test, it endorses the ‘object of command’ test.[50] Yet whether this is so is also unclear. He ascribes the supposed endorsement specifically to the paragraph adopting the dicta of Gaudron J in Pacific Coal.[51] Yet a fuller consideration of the reasoning that lay behind those dicta might rather suggest that Gaudron J, and hence by inference the WorkChoices majority, should after all be understood as espousing a version of the ‘distinctive character’ test.

The traditional objection to that test — an objection now repeated, as we have seen, in the WorkChoices judgment — was that any attempt to focus on the ‘trading’ aspect of trading corporations, the ‘financial’ aspect of financial corporations, and the ‘foreign’ aspect of foreign corporations, would lead to three different kinds of enquiry, making nonsense of the uniformity one expects from a single head of power. Yet the Court’s approach in Dingjan[52] effectively overcame this problem, by seizing upon a focus on corporate ‘business’ (itself derived from the statutory language considered in Actors Equity,[53] and ultimately from the judgment of Menzies J in Concrete Pipes Case)[54] as a way of raising the three disparate features (‘trading’, ‘financial’ and ‘foreign’) to a level of generality that effectively merges them in one coherent conception. A reference to the ‘business’ of a corporation covers both the trading activities of a trading corporation[55] and the financial activities of a financial corporation, while the fact that a foreign corporation does business in Australia is presumably the reason why it is selected as an object of Commonwealth legislative concern. That this was the underlying point is made clear by the way the WorkChoices majority explains the reasoning of Gaudron J in Dingjan, and treats it as the foundation for her subsequent dicta in Pacific Coal:[56]

Her Honour’s reasoning [in Dingjan] proceeded by the following steps. First, the business activities of corporations formed within Australia signify whether they are trading or financial corporations, and the main purpose of the power to legislate with respect to foreign corporations must be directed to their business activities in Australia. Second, it follows that the power conferred by s 51(xx) extends ‘at the very least’ to the business functions and activities of constitutional corporations and to their business relationships. Third, once the second step is accepted, it follows that the power ‘also extends to the persons by and through whom they carry out those functions and activities and with whom they enter into those relationships’.[57]

The result is not that s 51(xx) will support any law whatsoever that takes a constitutional corporation as its ‘object of command’, but rather that, in order to claim validity, a law must have a sufficient connection with a corporation’s ‘business’, since that is the reason why the specified classes of corporation are singled out as legitimate objects of legislative concern. The reasoning that underlay the so‑called ‘distinctive character’ test has been raised to a higher level of generality, but its structure remains the same.

Thus, in principle, the reasoning underlying the majority decision — or at least underlying the position taken by Gaudron J in Dingjan and Pacific Coal,[58] on which the majority relies — might still support judicially articulated limits on the scope of s 51(xx).

Whether any such restrictions will be possible in practice may be another matter. Certainly, the majority reasoning in the WorkChoices Case goes very far. For one thing, it repeatedly assumes[59] that s 51(xx) is a ‘person’ power, as the Incorporation Case held.[60] For another thing, the majority’s willingness to uphold the 2005 legislation goes well beyond the basic provisions regulating industrial relationships. For example, the new provisions now inserted in Division 7 of the Workplace Relations Act 1996 (Cth) attempt to erect elaborate statutory protections against the possibility that a workplace agreement might contain ‘prohibited content’.[61] The plaintiffs argued that these provisions had no sufficient connection with power; but the joint judgment upholds them because they form ‘an integral part of a set of provisions directed to forbidding employers and employees from making or seeking to make workplace agreements with prohibited content’.[62]

Again, there was a specific challenge to restrictions on the right of trade union officials to enter a workplace to investigate issues of occupational health and safety. By s 755(1)(a) of the Workplace Relations Act 1966 (Cth), the restrictions are made applicable to any premises ‘occupied or otherwise controlled by’ a constitutional corporation. The plaintiffs argued that this provision was caught by what McHugh J had said in Dingjan — namely, that s 51(xx) will not normally support a law ‘that does no more than make some activity of a s 51(xx) corporation the condition for regulating the conduct of an outsider’.[63] The WorkChoices judgment rejects this description:

[It] gives insufficient significance to the fact that the particular operation of the new Act that is in question is the regulation of a right of entry to premises, and that the premises to which the right of entry is controlled are premises ‘occupied or otherwise controlled by’ a constitutional corporation. This is a sufficient connection with s 51(xx), whether or not the entry that is thus regulated concerns a business being conducted on the premises by that corporation. The connection lies in the controlling of entry to a constitutional corporation’s premises.[64]

This passage seems especially revealing because it appears to sever the connection with corporate ‘business’ that was fundamental to the reasoning of Gaudron J in Dingjan[65] and Pacific Coal.[66]

Perhaps most controversially, the new statutory regime based on s 51(xx), like the old regime based on s 51(xxxv), provides for the registration of trade unions and employers’ associations with the Industrial Relations Commission, and also for the incorporation of organisations so registered. The original provisions had been upheld as reasonably incidental to legislation under s 51(xxxv) as early as Jumbunna Coal Mine NL v Victorian Coal Miners’ Association (‘Jumbunna Coal’);[67] but the trade unions represented in the WorkChoices Case argued strongly that the new provisions could not be upheld as incidental to legislation under s 51(xx). The majority judgment disagrees — relying upon Jumbunna Coal, and also upon Cunliffe v Commonwealth,[68] to illustrate ‘the extent of the power to register persons and organisations, and to incorporate the latter, if there be the sufficient connection with one or more paragraphs in s 51’:

If it be accepted … that it is within the corporations power for the Parliament to regulate employer–employee relationships and to set up a framework for this to be achieved, then it also is within power to authorise registered bodies to perform certain functions within that scheme of regulation. It also is within power to require, as a condition of registration, that these organisations meet requirements of efficient and democratic conduct of their affairs.[69]

In short, the 2005 legislation is upheld in its entirety, and any attempt to set limits to the use of the corporations power is dismissed.

V PIDOTO

Although Callinan J uses the ‘federal balance’ to argue that s 51(xx) could not have the extended scope ascribed to it by the majority,[70] Kirby J makes no such claim. His judgment assumes that if the paragraph were to be considered in isolation, it might extend so far. Instead, he focuses on what he perceives as ‘the central issue’:[71] namely, whether the potential operation of s 51(xx) is limited by the presence within the same section of a power dealing specifically with industrial relations (s 51(xxxv)). He answers that it is so limited, so that s 51(xx) cannot be used to render s 51(xxxv) effectively ‘otiose’.[72]

Callinan J is also willing to adopt this analysis, though only as reinforcing his concerns about ‘federal balance’.[73] Pointing out that s 51(xxxv) is the only grant of power to use ‘the language of industrial affairs’, he concludes that

it can be seen to represent the totality of the Commonwealth’s powers of control of industrial affairs, and to give rise to a negative or restrictive implication of the absence of a conferral of industrial power elsewhere under s 51 … [with exceptions] which it is unnecessary to define in this case. I would not regard this holding, of a negative implication, as different in substance from the holding of Kirby J that s 51(xx) be read down so as to exclude its application to industrial affairs.[74]

The difficulty with this argument is the well established presumption that the grants of power in s 51 must be construed independently, so that the potential use of one power is not to be hedged about or constrained by reference to another power. Moreover, the leading illustration of that presumption, Pidoto v Victoria (‘Pidoto’),[75] held specifically that other grants of power (in that case the defence power, s 51(vi)) were not to be constrained by reference to s 51(xxxv). Yet in fact the principle underlying that holding remains obscure.

It is not enough to distinguish Pidoto (as both Kirby and Callinan JJ seek to do)[76] by reference to its wartime context. The primary holding in Pidoto was simply that regulations extending the jurisdiction of the Commonwealth Court of Conciliation and Arbitration to industrial matters which did not satisfy the requirements of s 51(xxxv) were supported by the defence power (s 51(vi)); and for that holding the wartime context was obviously of crucial significance. But the relevance of Pidoto to the WorkChoices Case depends on a different holding — namely, that no barrier to the primary holding was presented by limitations supposedly derived from s 51(xxxv). And on that point, the authority of Pidoto appears to depend entirely on a passage in the judgment of Latham CJ.[77]

The problem — which only Latham CJ addressed — is what happens when a grant of power is circumscribed by express limitations on the power thereby conferred. That s 51(xxxv) is expressed in a way that does impose limitations on power was common ground in Pidoto: McTiernan J spelled out ‘[t]he limitations … inherent in s 51, pl xxxv’,[78] while Latham CJ also spoke of ‘the limitations which arise from [its] terms’:[79]

It has been held that under this provision the Commonwealth Parliament is limited to making laws for the prevention and settlement of industrial disputes (not for the direct regulation of industrial matters) and only of inter‑State industrial disputes, and for the prevention and settlement of such disputes only by the methods of conciliation and arbitration.[80]

In the frequently‑quoted passage in which Latham CJ disposed of the issue, he said:

Section 51(xxxv) is a positive provision conferring a specific power. The particular terms in which this power is conferred are not, in my opinion, so expressed as to be capable of being so construed as to impose a limitation upon other powers positively conferred.[81]

It is generally acknowledged that the Pidoto principle is not universal. (Nor does the above passage, with its guarded attention to ‘particular terms’, suggest that it is.) The conventional understanding postulates a primary rule that one head of power cannot be used to limit the scope of another; and a secondary rule (by way of exception) that restrictions expressed in one head of power may operate to restrict the scope of other heads of power as well.[82] The primary rule is illustrated most obviously by Pidoto; the secondary rule by Bourke v State Bank of New South Wales (‘Bourke’).[83] Building on the language used by Latham CJ in Pidoto, with its emphasis on ‘the particular terms’ in which a power is ‘expressed’, Leslie Zines has explained the secondary rule as arising ‘where the wording of a particular power expressly extracts from or restricts what otherwise might be included within it’.[84] Bourke is an example because the wording of s 51(xiii) — ‘Banking, other than State banking’ — first appears to grant a broad and general power over ‘Banking’, and then extracts or excepts from that subject matter a subset described as ‘State banking’. It followed, in Bourke, that ‘State banking’ had been excised from Commonwealth legislative power not only under s 51(xiii), but under other heads of power as well.

The explanation is unsatisfactory. In the first place, it is not what Latham CJ said. What he described as absent from Pidoto was a situation where one grant of power is ‘so expressed as to be capable of being … construed’ as imposing a limitation on other positive grants of power.[85] The situation illustrated by Bourke is one where one grant of power is ‘so expressed as to be capable of being … construed’ as imposing a limitation on that grant of power itself. The assumption is that wherever a limitation of the latter kind is found, the limitation will extend to other grants of power as well.[86]

Secondly, while a focus on ‘extractions’ and ‘exceptions’ may explain the actual result in Bourke, the unanimous judgment in Bourke treated that result as a mere illustration of the wider principle formulated by Dixon CJ in Attorney‑General (Cth) v Schmidt (‘Schmidt’):

It is hardly necessary to say that when you have … an express power, subject to a safeguard, restriction or qualification, to legislate on a particular subject or to a particular effect, it is in accordance with the soundest principles of interpretation to treat that as inconsistent with any construction of other powers conferred in the context which would mean that they included the same subject or produced the same effect and so authorized the same kind of legislation but without the safeguard, restriction or qualification.[87]

And Dixon CJ in turn had based his discussion of this principle, as applied to s 51(xxxi), on a passage from his own judgment in Bank of New South Wales v Commonwealth (‘Bank Nationalisation Case’), where he treated the principle merely as an application of the even broader maxim quando aliquid prohibetur, prohibetur et omne per quod devenitur ad illud[88] — often loosely translated[89] as meaning that what cannot be done directly cannot be done indirectly.[90]

Thirdly, Zines’ formulation depends on a differentiation which it is impossible to maintain. It postulates that if a grant of power proceeds by first establishing an area of legislative power, and then using the relevant descriptor to carve out an exception from the area thus established, the descriptor will operate as a restriction on other grants of power as well; but that if the same descriptor is used in the first place to limit (or more accurately to delimit) the area of the grant, it will have no effect on other grants. Yet typically the difference between these two situations will turn not on any difference of substance, but merely on convenience of description — since, logically, the areas of power delineated by these two contrasting methods will be exactly the same. A power in respect of ‘all non‑State banking’ would be exactly the same as a power in respect of ‘all banking except State banking’: yet the former would fall under Pidoto, and the latter under Bourke.[91] The result is a classic example of what Julius Stone identified in 1946[92] and again in 1964[93] as a ‘category of meaningless reference’: if the legal result is made to turn on a verbal distinction that reflects no difference in content, the result lacks any logical basis.

The problem is exacerbated by the ambiguity of the word ‘limitation’, which includes among its primary meanings both ‘[t]he action of determining … boundaries’ and ‘a limiting provision, rule or circumstance’ — or more simply ‘delimitation’ and ‘restriction’.[94] A limitation in the first sense will attract Pidoto; a limitation in the second sense will attract Bourke. One consequence of this ambiguity is that it is often unclear whether the wording of a particular paragraph in s 51 involves a ‘limitation’ in the first or the second sense;[95] and s 51(xxxv) is a good example.

In the passage already quoted from Pidoto, Latham CJ identified two crucial ‘limitations’ on the grant of power in s 51(xxxv): an industrial dispute attracting the power must extend ‘beyond the limits of any one State’, and it must be dealt with only through the mechanism of ‘conciliation and arbitration’.[96] Predictably, Kirby and Callinan JJ seize on these two ‘limitations’ as extending to Commonwealth legislation under other heads of power as well — and, in this case, to legislation under s 51(xx). Predictably also, Callinan J builds his dissent primarily around the exclusion of single‑state disputes, Kirby J primarily around the insistence on independent machinery for ‘conciliation and arbitration’.[97]

The majority judgment counters these arguments initially by narrowing the Bourke formulation even further, stressing (first) that it applies only where a grant of power is subject to ‘a positive prohibition or restriction’, and (secondly) that even where such a prohibition is found, it may apply only to that particular paragraph and not to Commonwealth powers in general.[98] On this analysis, the majority insists that the limiting expressions — ‘conciliation and arbitration’, and ‘extending beyond the limits of any one State’ — are not formulated by way of ‘prohibition or restriction’, but are merely parts of ‘a compound conception’.[99]

Thus far, the majority judgment falls directly into the fallacy identified by Stone, since the reasoning turns simply on the meaningless distinction between alternative modes of expression. The judgment does not stop there, however. Stone had emphasised two ways in which the emptiness of such distinctions might be transcended; and the majority judgment goes on to introduce both of these. First, Stone conceded that his analysis may not apply where a proposition is ‘embodied in an authoritative form of words’[100] such as a statute (or perhaps a fortiori a constitution), since the draftsman may have chosen that form of words precisely in order to give effect to one intention rather than the other.[101] Secondly, Stone’s less immediate purpose was to argue that the meaningless distinction could and should always be transcended, by basing the ultimate choice on considerations of substance and not merely on the form of words.[102] The joint judgment deploys both of these strategies.

In the first place, it stresses that in any event, s 51(xxxv) deals only with power in respect of ‘industrial disputes’: it does not purport to deal exhaustively with the whole range of ‘industrial matters’.[103] (Thus, whatever s 51(xxxv) might be thought to say with respect to ‘industrial disputes’, it might still be open to the Parliament under other heads of power to deal with other aspects of ‘industrial matters’.) In the second place, even as to ‘industrial disputes’, the judgment canvasses the Convention Debates of the 1890s to show that it is unlikely that the framers would ever have intended to confine the power to deal with such disputes to a single paragraph.[104] For example, in 1898 John Quick suggested that s 98, extending the trade and commerce power to ‘navigation and shipping’, would also extend ‘to labour disputes in connexion with navigation and shipping’.[105] More audaciously, the joint judgment suggests that the Inter‑State Commission envisaged by s 101 of the Constitution might have been intended to play such a role.[106] In the third place, the joint judgment points to an assortment of precedents in which the Parliament has in fact been permitted to deal with ‘industrial disputes’ under other heads of power. Pidoto is only one such case;[107] but the reasoning of Latham CJ in that case ‘is compelling and should be followed’.[108]

These arguments still beg the question. In particular, while the precedents adduced do show that ‘industrial matters’ can be dealt with under other heads of power, they do not involve any direct competition with s 51(xxxv). Either they deal with workers’ compensation[109] in a way that might conceivably compete with or encroach upon state workers’ compensation legislation, but involves no departure from the machinery for conciliation and arbitration established under s 51(xxxv); or they accept the essential features of that established machinery, but seek to expand its jurisdiction.[110] Only in Pidoto is the crucial issue raised directly, and the reasons why its reasoning is ‘compelling’ are not further explained. Nevertheless, these features of the majority judgment do represent an attempt to decide the issue ‘upon considerations of substance and not of form’.[111]

Kirby and Callinan JJ, for their part, also invoke the Convention Debates. For Callinan J, they reveal the framers’ intention ‘that any federal power in relation to industrial affairs was to be confined to those of an interstate character, and that the former colonies were to retain power over internal industrial disputes’.[112] For Kirby J, the Debates reveal that it was ‘only by restricting the grant of power in the two ways stated (compliance with the necessity of interstateness and indirect, independent, decision‑making)’ that a power relating to industrial affairs was able to ‘squeak through to find its place’ in s 51(xxxv).[113] Thus these judges, too, invoke the framers’ intention as a guide to the operative effect of the words.

But, at least as to the requirement of ‘conciliation and arbitration’, Kirby J goes further, relying on its substantive value‑content as well as its history. Where the majority narrows the Bourke rule so that it applies only to a definite ‘prohibition or restriction’, Kirby J raises it to a more general level by reverting to the earlier formulation in Schmidt: any ‘safeguard, restriction or qualification’ imposed on a grant of power will apply to legislation reliant on any other grant of power as well.[114] He explains the narrower rule in Bourke, confined to express exclusions or exceptions from a grant of power, by treating this as only one subcategory of the broader conception underlying Schmidt.[115] He treats the cases on s 51(xxxi) as illustrating a different subcategory, in which the Constitution ‘limits the exercise of [a] power by reference to a constitutional guarantee, protective of the legal rights of those potentially affected by the federal law’.[116] For Kirby J, the insistence on ‘conciliation and arbitration’ falls within this conception, since it protects the parties to industrial conflict against direct political intervention, and ensures an independent procedure for arriving at outcomes representing ‘the interests of [the] parties, the public interest …, due process, transparent negotiations’ and ‘the ideal of “a fair go”.’[117] At least since 1904, the requirement of conciliation and arbitration has evolved into an ‘important guarantee of industrial fairness and reasonableness’.[118] The pervasive theme of his Honour’s judgment — his view of s 51(xxxv) as the historical vehicle for development and transmission of ‘some of the core values that have shaped the evolution of the distinctive features of the Australian Commonwealth, its economy and its society’[119] — is clearly intended to have an emotional force going far beyond its significance as an element in a dry legal argument. But he also seeks to represent some elements of his emotional plea as the content of a value‑laden ‘guarantee’ embedded in s 51(xxxv) and operating by virtue of Schmidt as a restraint on all other Commonwealth powers as well.

He does appear to retain for his ‘guarantee’ category one element of the analysis developed in Bourke for express exclusions or exceptions. On his reasoning, the first step is to establish that the legislation can fairly be characterised as a law with respect to industrial disputes.[120] But once it can be so characterised, it must then observe the ‘safeguards’ upon which the specific grant of power in respect of industrial disputes is conditioned.

The trouble is that this attempt to transmute emotion into logic does not go far enough. It fails to explain how the principle in Schmidt — however broadly we state it, and however widely we extend the category of ‘guarantees’ to which it applies — can be reconciled with Pidoto. Moreover, it fails to acknowledge the settled assumption that the relationship (or lack of it) between grants of legislative power will normally be governed by Pidoto, and that any contrary rule derived from Bourke (or from Schmidt or from Broom’s Legal Maxims)[121] must operate only by way of exception. The Pidoto principle must have supremacy not only because it represents ‘law taken for granted’,[122] but because it accords with general principles of characterisation.

Yet it is perhaps precisely in the general principles of characterisation that it might have been possible to find a better solution, giving apodeictic rather than merely emotive force to Kirby J’s protest[123] that the majority view would, in effect, render s 51(xxxv) ‘otiose, or at least optional for most purposes, effectively consigning it to … insignificance’.

On every question of characterisation, the ultimate test is that of ‘sufficient connection’. But this is not a self‑contained or self‑executing test. The Oxford English Dictionary defines ‘sufficient’ as meaning ‘[o]f a quantity, extent, or scope adequate to a certain purpose or object’.[124] That is, the relevant assessment of ‘adequacy’ can only be made by reference to the ‘purpose or object’ to be served. The connection that is required between a law and its supporting head of legislative power need not be ‘close’ or ‘direct’; it need only be sufficiently ‘close’ or ‘direct’ to support that particular law. But what is sufficient to support one law may not suffice to support another: a particular law may ‘go too far’ for the sufficiency of the connection to be sustained.

There is some analogy here with the test laid down in Commonwealth v Bank of New South Wales for impermissible interference with freedom of interstate trade: a law will not be invalid merely because it results in ‘some indirect or consequential impediment which may fairly be regarded as remote’.[125] In SOS (Mowbray) Pty Ltd v Mead, Barwick CJ explained that test by using the metaphor of widening ripples from a stone cast into a pond.

[A]ll that the law produces in its application to the facts is within its operation until [the] point of remoteness is reached; or it can be seen that the effect is but a consequence. All that impact which is not remote or mere consequence is direct and immediate in the relevant sense.[126]

In short, everything is sufficiently direct until it is too remote — that is, until it ‘goes too far’.

The majority judgment does in effect consider two ways in which a law might be said to ‘go too far’ for the sufficiency of its connection with power to be maintained. One is ‘Higgins’ list of horribles’:[127] the catalogue of anomalous consequences that might follow if s 51(xx) were not confined within justiciable limits. The other is a perceived distortion of the ‘federal balance’. The joint judgment rejects the first of these yardsticks by insisting that constitutional interpretation ought not to be swayed by ‘extreme examples or distorting possibilities’.[128] It rejects the second by treating the notion of ‘federal balance’ as merely a presupposition, erroneously smuggled in as antecedent to the task of characterisation, and in any event having no objective content against which the sufficiency of connection with power can be determined.

Yet one can conceive of other ways in which a law might ‘go too far’ which do provide an objective yardstick. One is the extent to which a law offends against traditional notions of civil liberties (such as freedom of speech) or the rule of law. It was on such a ground that in Davis v Commonwealth a unanimous High Court held that federal legislative power did not extend to the restricted lexical choices allowed by the Australian Bicentennial Authority Act 1980 (Cth), because the restrictions were ‘grossly disproportionate’ and ‘reach too far’.[129] Similarly, it was on such a ground that the ‘deeming’ provision in Actors Equity, imputing responsibility to trade unions for the acts of their officers, was held to have (as Mason J put it) ‘a very remote connexion with corporations, a connexion so remote that the provision cannot be characterized as a law with respect to corporations’.[130]

This does not mean that legislation can never interfere with basic common law values. On the contrary, it constantly does so. What is involved is a double question of degree: the impugned law is measured both for the degree of strength or directness or obviousness of its connection with the head of power relied upon, and for the extent and severity of its adverse impact on traditional values. The question is whether this degree of connection is sufficient to support this degree of adverse impact.[131]

In the same way, one might argue that a law enacted in purported reliance on one head of power might ‘go too far’ because of the extent and severity of its adverse impact on another head of power. The argument would be that where the impact of legislation under one head of power entails such an extensive incursion into the territorial ‘field’ of another as to leave it with no operative content (or very little operative content), the incursion will only be valid in a case where the connection with the head of power relied on is sufficiently ‘strong’ or ‘close’ to support that incursion. Such an argument would be fully compatible with recognition that legislation under other heads of power (including s 51(xx)) may sometimes operate validly and even extensively on the subject matter of s 51(xxxv).[132]

Obviously, such an argument would also be compatible with the result in Pidoto. The actual decision in that case involved no incursion into the sphere of s 51(xxxv) at all, since the claim was one which could not have attracted jurisdiction under s 51(xxxv). More broadly, the judgments also upheld the validity of regulations expanding the jurisdiction of the Court of Conciliation and Arbitration not only without regard to the ‘limitations’ imposed by s 51(xxxv), but specifically for the purpose of avoiding them. But the additional jurisdiction conferred was of a temporary nature, clearly related to the conduct of the war and limited to its duration, and conditioned on the need for a determination that any specific dispute thus brought within jurisdiction had a sufficient connection with the wartime need for ‘industrial peace and national security’.[133] Above all, the regulations approved in Pidoto did nothing to modify what Kirby J calls ‘the protective machinery of conciliation and arbitration’;[134] they simply extended that machinery to a limited number of additional cases. Thus, on any view, the relevance to defence purposes was close and obvious, and the incursion into the area of conciliation and arbitration was relatively minor.

Nor would the argument here envisaged give rise to any conflict with the actual language used in Pidoto, since it does not depend on any suggestion that the ‘particular terms’ of one head of power are ‘capable of being so construed as to impose a limitation upon other powers’.[135] The argument entails no definitional limit on the ambit of any head of power, but only a limit on the extent to which a power can sometimes be used.

It is unclear whether, in order to be persuasive, such an argument would have to be shown to hold good for the relations between all heads of Commonwealth power. One possibility is that the issues arising from particular powers, or particular combinations of powers, are so different that it is simply a mistake to seek a single uniform rule to govern all possible cases. On the face of it, for example, the above argument would be difficult to reconcile with the settled understanding of the relationship between placitum (xxix) of s 51 and placita (x) and (xxx), the former relating to ‘external affairs’ and the latter respectively to ‘fisheries in Australian waters beyond territorial limits’ and ‘the relations of the Commonwealth with the islands of the Pacific’. In both cases, it seems clear that laws validly enacted under s 51(xxix) might so comprehensively ‘cover the field’ of the other paragraphs referred to as to leave them with no effective operation at all. On the other hand, the proposed understanding of ‘sufficient connection’ might extend to these cases, since any such laws would have such a strong or obvious connection with ‘external affairs’ as to satisfy the suggested criterion.[136]

Whether such an argument would have sufficed to change the result in the WorkChoices Case is (to say the least) open to doubt. Yet it might at least have enabled the dissenters to mount a more persuasive case.

VI OTHER MATTERS

Although the majority judgment dismisses the idea of ‘federal balance’ as vague and lacking in content, it gives equally short shrift to any doctrines seeking to give that idea more specific content. One such doctrine is, of course, the Melbourne Corporation principle.[137] The plaintiffs invoked it particularly in relation to s 117 of the amended Workplace Relations Act 1996 (Cth), which empowers the Australian Industrial Relations Commission to restrain a state industrial authority from dealing with any matter pending ‘before the Commission under this Act’. It was also said that any order made under this provision would purport to restrain a state authority from applying or enforcing state laws,[138] or impede ‘an organ of State government’ in its ‘constitutional functions’, and would thus infringe s 106 of the Constitution by impairing the operation of state constitutions.[139] More generally, it was said that Commonwealth powers cannot be used ‘to interfere with the functions of State executives’.[140]

The majority judgment points out that none of these objections has ever been raised against previous versions of s 117, which gave similar powers to the Commission and its predecessors in the exercise of their functions under s 51(xxxv) of the Constitution. On the contrary, there are several cases in which it has been said or assumed that such provisions were valid. While those cases have never involved arguments based on Melbourne Corporation (or on s 106), ‘the novelty of the arguments may be seen as a badge of their lack of merit’.[141] The argument based on s 106 would require an assumption that the laws establishing a state’s industrial tribunals are part of its constitution;[142] and whether or not that assumption is plausible, the plaintiffs did not establish it. As for the Melbourne Corporation argument, any orders made under s 117 would normally be confined to orders directed against ‘a State industrial authority’; and although such an order might conceivably extend to other state courts, it certainly could not extend to ‘their ordinary civil and criminal jurisdiction’.[143] Thus s 117 was not ‘directed at the core of State judicial systems’.[144] More generally, its ‘relatively minor’ interference with state concerns could not be said to affect a state’s capacity to ‘function as a government’ or to ‘exercise constitutional functions’.[145]

Another doctrine derived from the general idea of ‘federal balance’ is that of Evatt J that the Commonwealth cannot ‘create’ or ‘manufacture’ inconsistency:[146] either s 109 operates, or it does not, and the Commonwealth cannot manipulate the outcome by legislation explicitly designed to trigger its operation. Later cases, while never wholly repudiating this notion, have effectively confined its possible operation to what Dixon J in Wenn v Attorney‑General (Vic) (‘Wenn’) called ‘a bare attempt to exclude State concurrent power’[147] in an area where the Commonwealth has made no law of its own (as distinct from a case where the Commonwealth passes a law, and expresses an intention to ‘cover the field’). On that basis, elaborate provisions excluding the operation of specific state laws have repeatedly been upheld,[148] so long as they are attached to a Commonwealth regulatory scheme. There was thus little prospect that such an argument might succeed in the WorkChoices Case.

The argument focused on the new s 16(1), which declares an intention that the Workplace Relations Act 1996 (Cth) (as amended) shall apply to exclude any operation, ‘in relation to an employee or employer’, of any ‘State or Territory industrial law’ — as well as any other state or territory law that ‘deals with leave other than long service leave’; or provides for orders relating to equal pay ‘for work of equal value’; or provides for the variation of unfair contracts of employment; or ‘entitles a representative of a trade union to enter premises’.

The expression ‘State or Territory industrial law’ is defined by s 4. It specifically includes the Industrial Relations Act 1996 (NSW), Industrial Relations Act 1999 (Qld), Fair Work Act 1994 (SA), Industrial Relations Act 1984 (Tas) and Industrial Relations Act 1979 (WA). It also includes a state or territory law applying ‘to employment generally’ whose main purposes include any of a lengthy list of employment‑related matters. The definition also extends to subordinate legislation, and to any other state or territory laws that may be prescribed. However, s 16(2) and (3) create exemptions for state or territory laws dealing with the prevention of discrimination, equal opportunity and other ‘non‑excluded matters’ (such as superannuation, workers compensation, child labour, long service leave and industrial action affecting essential services).[149]

The plaintiffs first submitted that s 16 is not supported by any head of power. The majority treats this submission as depending largely on a question of construction, since one argument was that the expressions ‘employer’ and ‘employee’ in s 16 are excessively broad. But for purposes of the amended Act as a whole, ss 5 and 6 confine those expressions to a series of definitions each invoking a particular head of Commonwealth power.[150] By this device the Act is confined to a series of operations relating to different kinds of ‘employers’, each supported by a particular grant of power. The majority concludes that the words ‘employer’ and ‘employee’ in s 16 are similarly confined, and hence that the various heads of power invoked by ss 5 and 6 suffice to support that provision, just as they do for the substantive provisions.

The joint judgment adverts briefly to the further submission that, even on that interpretation, s 16 cannot be characterised ‘as a law with respect to corporations’, but says simply: ‘The submission must be rejected.’ No reasons are given. Nor is any reference made to Construction, Forestry, Mining and Energy Union (New South Wales Branch) v Newcrest Mining Ltd, where the New South Wales Industrial Commission appeared to support such a submission by holding (in relation to the 1996 provisions) that ‘there are limits to the extent to which the corporations power can be used to support a law to exclude a State award or State agreement’, and that the power does not extend to laws that ‘extinguish the power of a State Industrial Authority … to resolve an industrial dispute’.[151] The elaborate attention to the preliminary question of construction, combined with the cursory dismissal of the question of characterisation, is distressingly typical of the current tendency to seize on questions of statutory construction as a way of avoiding any serious discussion of constitutional principles.

The plaintiffs then argued that s 16 constitutes ‘a bare attempt to limit or exclude State legislative power … rather than to comprehensively regulate a particular field of activity to the exclusion of any State law which also regulates that field of activity’.[152] In particular, Western Australia argued that, while the amended Workplace Relations Act 1996 (Cth) covers some of the matters on which the operation of state laws is excluded, there are other such matters for which it makes no provision[153] — so that, at least in these areas, s 16 seeks to effect ‘a bare exclusion of State law’.[154] The Commonwealth replied that, so long as its law evinces an intention to ‘cover the field’, it need not make ‘its own detailed provisions about every matter within that field’.[155] Again, the joint judgment concludes that: ‘The Commonwealth’s submissions are to be preferred.’[156] Just as the Commonwealth may enact ‘a scheme involving a more detailed form of regulation than State law provides’, so it may choose to enact ‘a scheme involving less detailed regulation than State law provides’ — or, as here, it may provide ‘a more detailed scheme than State law in some respects and a less detailed scheme in other respects’.[157]

The one issue on which the plaintiffs elicit some degree of sympathy from the majority — though not enough for a positive holding — is the extent to which legislative power can be delegated to the executive. The essential battle over this issue was fought and lost as long ago as Victorian Stevedoring & General Contracting Co Pty Ltd and Meakes v Dignan (‘Dignan’s Case’);[158] but the outcome was not unqualified. Dixon J warned that a delegation might not be valid ‘however extensive or vague the subject matter may be’, since there may be ‘such a width or such an uncertainty of the subject matter to be handed over that the enactment attempting it is not a law with respect to any particular head or heads of power.’[159] Evatt J suggested that the Commonwealth cannot ‘abdicate’ its legislative powers: a provision empowering the executive government to ‘make regulations having the force of law upon the subject of trade and commerce with other countries or among the States’ would not be valid, since it would not be ‘a law with respect to trade and commerce with other countries or among the States … but a law with respect to the legislative power to deal with [that] subject’.[160]

These reservations were invoked by the plaintiffs as a basis for their submission that certain provisions in the amended Workplace Relations Act 1996 (Cth) were invalid. In particular, though a series of provisions (notably ss 3589 and 3656) is designed to prevent a workplace agreement from containing ‘prohibited content’, that expression was nowhere defined by the 2005 legislation. Instead s 356 provided simply: ‘The regulations may specify matters that are prohibited content for the purposes of this Act.’

The majority judgment does not extend a judicial imprimatur to this drafting technique, but is not prepared to strike it down, either. The judgment points out that the content that may be prohibited is not left wholly to the discretion of the executive government, since other provisions specify various examples of ‘required content’ which a workplace agreement must contain.[161] (That is, whatever is ‘required content’ can obviously not be ‘prohibited content’.) The range of matters that might be specified as ‘prohibited content’ is also constrained by s 172, which provides that the ‘key minimum entitlements’ enshrined in the new ‘Pay and Conditions Standard’ must prevail over workplace agreements to the extent to which they represent a more favourable outcome for the employee; and by s 173, which provides that any term in a workplace agreement purporting to exclude the Pay and Conditions Standard shall be of no effect. Again, it would not be possible to specify such a wide range of ‘prohibited content’ ‘as to undermine the possibility of making a workplace agreement capable of practical operation’.[162] More generally, the joint judgment treats the regulatory power to specify ‘prohibited content’ as sufficiently constrained by the general requirement in s 846(1)(b) that regulations be ‘necessary or convenient … for the purposes of this Act’. Yet the judgment does emphasise that none of this amounts to judicial approval:

It should be said that the technique employed in s 356 is an undesirable one which ought to be discouraged. For one thing it requires the lawyers (and the many non‑lawyers) who have to work with the new Act to look outside it in order to apply it: identifying what regulations are in force is a task which many inquirers have found difficult. And it creates difficulty in assessing whether particular regulations made under the legislation are intra vires. However, to make these criticisms is one thing; to conclude that there is constitutional invalidity is another.[163]

Kirby J takes a less indulgent view. He finds it wholly inadequate for the majority judges to ‘content themselves’ with this degree of ‘chastisement’:

With all respect, this is an inadequate response … Under the Constitution, it is the duty of this Court to uphold the law‑making and supervisory powers of the Parliament. We should not sanction still further erosion of those powers and their effective transfer to the Executive Government … There comes a point when a regulation‑making power becomes so vague and open‑ended that the law which establishes it ceases to be a law with respect to a subject of federal law‑making power …

The plaintiffs’ challenge … should be upheld both to defend the proper constitutional role of the Federal Parliament and to discourage future similar measures. The impugned provisions border on an endeavour to enact an abdication of the Parliament’s responsibilities. This Court should say so and forbid it.[164]

In response to these criticisms, s 356 has now been amended to define ‘prohibited content’, though still allowing for further definition by regulation. How far Kirby J’s more fundamental objections to the legislation will be satisfied by the amendments to be made in 2008 remains to be seen.


[1] [2006] HCA 52; (2006) 229 CLR 1 (‘WorkChoices Case’).

[∗] LLM (Syd), LLD (Macq); Founding member, Faculty of Law, The University of New South Wales; Emeritus Professor of Law, Macquarie University.

[2] [2006] HCA 52; (2006) 229 CLR 1.

[3] (1990) 169 CLR 482.

[4] Cf Kartinyeri v Commonwealth [1998] HCA 22; (1998) 195 CLR 337, 356 (Brennan CJ and McHugh J) (‘Hindmarsh Island Bridge Case’): ‘the power which supports a valid Act supports an Act repealing it’. See also at 376 (Gummow and Hayne JJ): ‘what the Parliament may enact it may repeal’.

[5] WorkChoices Case [2006] HCA 52; (2006) 229 CLR 1, 333–4.

[6] Ibid 215, 218.

[7] See Jeremy Webber, ‘The Jurisprudence of Regret: The Search for Standards of Justice in Mabo[1995] SydLawRw 1; (1995) 17 Sydney Law Review 5.

[8] Gleeson CJ, Gummow, Hayne, Heydon and Crennan JJ.

[9] (2000) 203 CLR 346, 375.

[10] (1971) 124 CLR 468.

[11] That is, a corporation falling into one of the categories referred to in s 51(xx): a foreign corporation, or a trading or financial corporation formed within Australia.

[12] See especially Actors and Announcers Equity Association v Fontana Films Pty Ltd (1982) 150 CLR 169 (‘Actors Equity’); Commonwealth v Tasmania [1983] HCA 21; (1983) 158 CLR 1 (‘Tasmanian Dam Case’).

[13] [1995] HCA 16; (1995) 183 CLR 323.

[14] Notably by Gummow, Hayne and Heydon JJ in Singh v Commonwealth [2004] HCA 43; (2004) 222 CLR 322, 383: ‘The task of the Court is not to describe the metes and bounds of any particular constitutional provision; it is to quell a particular controversy’.

[15] [2006] HCA 52; (2006) 229 CLR 1, 75–6 (Gleeson CJ, Gummow, Hayne, Heydon and Crennan JJ): ‘the plaintiffs’ submissions require consideration of what is meant by a law “with respect to” the subject matter of constitutional corporations, rather than identification of the metes and bounds of the subject matter of the relevant head of power.’

[16] (2000) 202 CLR 479, 492 (Gleeson CJ, Gaudron, McHugh, Gummow, Hayne and Callinan JJ).

[17] WorkChoices Case [2006] HCA 52; (2006) 229 CLR 1, 104 (Gleeson CJ, Gummow, Hayne, Heydon and Crennan JJ).

[18] [1909] HCA 36; (1908) 8 CLR 330, 396.

[19] WorkChoices Case [2006] HCA 52; (2006) 229 CLR 1, 87–8 (Gleeson CJ, Gummow, Hayne, Heydon and Crennan JJ), citing Bateman v Service (1881) 6 App Cas 386, and piquantly relying on choice‑of‑law doctrine to argue that constitutional arguments should not rely on choice‑of‑law doctrine.

[20] (1982) 150 CLR 169, 216–18.

[21] [2006] HCA 52; (2006) 229 CLR 1, 115 (Gleeson CJ, Gummow, Hayne, Heydon and Crennan JJ).

[22] [1995] HCA 16; (1995) 183 CLR 323, 337–9.

[23] WorkChoices Case [2006] HCA 52; (2006) 229 CLR 1, 114 (Gleeson CJ, Gummow, Hayne, Heydon and Crennan JJ).

[24] Ibid 116 (Gleeson CJ, Gummow, Hayne, Heydon and Crennan JJ).

[25] Ibid 104 (Gleeson CJ, Gummow, Hayne, Heydon and Crennan JJ).

[26] Ibid 112 (Gleeson CJ, Gummow, Hayne, Heydon and Crennan JJ).

[27] Ibid 103 (Gleeson CJ, Gummow, Hayne, Heydon and Crennan JJ).

[28] The joint judgment notes particularly the concerns of Gibbs CJ in Actors Equity (1982) 150 CLR 169, 182 and of Dawson J in Dingjan [1995] HCA 16; (1995) 183 CLR 323, 345–6. (But obviously, a rejection of the motivations for espousing a particular doctrine does not necessarily mean that the resulting doctrine should itself be rejected.)

[29] WorkChoices Case [2006] HCA 52; (2006) 229 CLR 1, 104 (Gleeson CJ, Gummow, Hayne, Heydon and Crennan JJ). Another in the Grain Pool collection of canons (see above n 16 and accompanying text) is Kitto J’s insistence in Fairfax v Federal Commissioner of Taxation [1965] HCA 64; (1965) 114 CLR 1, 7, that the character of a law be determined by reference to ‘the rights, duties, powers and privileges’ which it creates — now oddly attributed, with some muddling of the Hohfeldian language, to the Hindmarsh Island Bridge Case [1998] HCA 22; (1998) 195 CLR 337, 352–3 (Brennan CJ and McHugh J).

[30] [1920] HCA 54; (1920) 28 CLR 129.

[31] WorkChoices Case [2006] HCA 52; (2006) 229 CLR 1, 118 (Gleeson CJ, Gummow, Hayne, Heydon and Crennan JJ).

[32] [1947] HCA 26; (1947) 74 CLR 31, 82–3.

[33] WorkChoices Case [2006] HCA 52; (2006) 229 CLR 1, 118 (Gleeson CJ, Gummow, Hayne, Heydon and Crennan JJ). Kirby J’s more nuanced account of federalism also emphasises this point: at 226.

[34] Ibid 271–2. (But Callinan J sees the Engineers’ Case as a ‘somewhat unsatisfactory … early instance of judicial activism’: at 369.)

[35] Ibid 321: ‘the exercise of different powers of varying importance by each of the Commonwealth and the States, but not so that, relevantly for present purposes, the essential functions and institutions of the States … are obstructed, impeded, diminished, or curtailed’.

[36] [1909] HCA 36; (1908) 8 CLR 330, 409–10.

[37] P H Lane, The Australian Federal System (2nd ed, 1979) 160.

[38] WorkChoices Case [2006] HCA 52; (2006) 229 CLR 1, 117 (Gleeson CJ, Gummow, Hayne, Heydon and Crennan JJ).

[39] Leslie Zines, ‘Characterisation of Commonwealth Laws’ in H P Lee and George Winterton (eds), Australian Constitutional Perspectives (1992) 33, 59.

[40] See, eg, Western Australia v Commonwealth [1975] HCA 46; (1975) 134 CLR 201, 271 (Mason J) (‘First Territory Senators Case’).

[41] [1920] HCA 54; (1920) 28 CLR 129, 151 (Knox CJ, Isaacs, Rich and Starke JJ).

[42] Pacific Coal (2000) 203 CLR 346, 375 (emphasis added).

[43] Ibid.

[44] WorkChoices Case [2006] HCA 52; (2006) 229 CLR 1, 206–7.

[45] Ibid 366.

[46] Ibid 206–7.

[47] (2000) 203 CLR 346, 375 (emphasis added).

[48] Ibid.

[49] Cf Caltex Oil (Aust) Pty Ltd v Feenan [1981] 1 NSWLR 169, 173 (Lord Diplock):

To speak of ‘construing’ the words in which judges have chosen to express the reasons for their judgment involves … a misuse of language that is all too common and reflects a mistaken approach to the use of judicial precedent. The only words that require to be ‘construed’ are those of [a] statute … The language used by judges to … apply to the particular circumstances of the case under consideration, is chosen with those particular circumstances in mind and is not intended as … necessarily appropriate to all other circumstances.

[50] WorkChoices Case [2006] HCA 52; (2006) 229 CLR 1, 375–6. It follows in Callinan J’s opinion that the Incorporation Case ‘may well now be effectively overruled’, since the ‘object of command’ test ‘is not reconcilable with’ that decision: at 376.

[51] Ibid 366.

[52] [1995] HCA 16; (1995) 183 CLR 323.

[53] (1982) 150 CLR 169.

[54] (1971) 124 CLR 468, 511: ‘I am not prepared to attempt to define the limits of the power … I content myself with saying that a law … governing the conduct of its business by a trading corporation formed within the limits of the Commonwealth is within the power of the Parliament by virtue of s 51(xx).’

[55] But see Brennan J’s warning in Actors Equity (1982) 150 CLR 169, 221, that ‘business and trade are not coextensive’.

[56] (2000) 203 CLR 346, 375.

[57] WorkChoices Case [2006] HCA 52; (2006) 229 CLR 1, 114 (Gleeson CJ, Gummow, Hayne, Heydon and Crennan JJ).

[58] (2000) 203 CLR 346, 375.

[59] WorkChoices Case [2006] HCA 52; (2006) 229 CLR 1, 104, 117, 159 (Gleeson CJ, Gummow, Hayne, Heydon and Crennan JJ).

[60] (1990) 169 CLR 482, 497 (Mason CJ, Brennan, Dawson, Toohey, Gaudron and McHugh JJ).

[61] Thus s 365 makes it an offence to seek to include such content in an agreement and s 366 makes it an offence to represent falsely (and recklessly) ‘that a particular term does not contain prohibited content’.

[62] WorkChoices Case [2006] HCA 52; (2006) 229 CLR 1, 143 (Gleeson CJ, Gummow, Hayne, Heydon and Crennan JJ).

[63] [1995] HCA 16; (1995) 183 CLR 323, 370.

[64] [2006] HCA 52; (2006) 229 CLR 1, 146 (Gleeson CJ, Gummow, Hayne, Heydon and Crennan JJ).

[65] [1995] HCA 16; (1995) 183 CLR 323.

[66] (2000) 203 CLR 346, 375.

[67] [1908] HCA 95; (1908) 6 CLR 309.

[68] [1994] HCA 44; (1994) 182 CLR 272.

[69] WorkChoices Case [2006] HCA 52; (2006) 229 CLR 1, 153 (Gleeson CJ, Gummow, Hayne, Heydon and Crennan JJ).

[70] Ibid 320–1.

[71] Ibid 190.

[72] Ibid 190, 202.

[73] Conversely (or reciprocally), Kirby J is willing to invoke the ‘federal character’ of the Constitution as an ‘additional consideration’ supporting his principal argument: ibid 222–9.

[74] Ibid 350. See also his Honour’s reasoning (at 344) and conclusion (at 384–5).

[75] [1943] HCA 37; (1943) 68 CLR 87.

[76] WorkChoices Case [2006] HCA 52; (2006) 229 CLR 1, 230–3 (Kirby J), 337 (Callinan J).

[77] For McTiernan J the only issue was whether the use of s 51(vi) infringed s 107 of the Constitution by invading ‘a field of State legislative power’: [1943] HCA 37; (1943) 68 CLR 87, 123. Williams J made only the obvious point that power to deal with some industrial issues under s 51(xxxv) does not preclude legislation on other industrial issues under a different head of power: at 127. Starke J, dissenting as to s 51(vi), did not refer to s 51(xxxv) at all, while Rich J was content to express his ‘substantial agreement’ with Latham CJ and Williams J: at 115.

[78] Ibid 122 (emphasis added).

[79] Ibid 99 (emphasis added).

[80] Ibid 100 (emphasis added).

[81] Ibid 101.

[82] See Tony Blackshield and George Williams, Australian Constitutional Law and Theory: Commentary and Materials (4th ed, 2006) 784.

[83] [1990] HCA 29; (1990) 170 CLR 276.

[84] Leslie Zines, The High Court and the Constitution (4th ed, 1997) 24.

[85] Pidoto [1943] HCA 37; (1943) 68 CLR 87, 101.

[86] One aspect of the majority judgment in the WorkChoices Case (unnecessary to the decision) is explicitly to deny this assumption: even if the relevant kind of limitation were to be found in s 51(xxxv), the joint judgment would still propose a further enquiry as to whether it would govern the use of other legislative powers as well: [2006] HCA 52; (2006) 229 CLR 1, 127–8 (Gleeson CJ, Gummow, Hayne, Heydon and Crennan JJ). The criteria that might guide this further enquiry are not explained, however.

[87] [1961] HCA 21; (1961) 105 CLR 361, 371–2, as quoted in Bourke [1990] HCA 29; (1990) 170 CLR 276, 285–6 (Mason CJ, Brennan, Deane, Dawson, Toohey, Gaudron and McHugh JJ).

[88] [1948] HCA 7; (1948) 76 CLR 1, 349–50.

[89] See, eg, Herbert Broom, Broom’s Legal Maxims (4th ed, 1854) 314. Cf D K Singh, ‘What Cannot Be Done Directly Cannot Be Done Indirectly’ (1959) 32 Australian Law Journal 374; 33 Australian Law Journal 3.

[90] As to that proposition, WorkChoices Case [2006] HCA 52; (2006) 229 CLR 1, 131 (Gleeson CJ, Gummow, Hayne, Heydon and Crennan JJ) quotes what was said by Gleeson CJ in Pacific Coal (2000) 203 CLR 346, 360:

there is no principle that Parliament can never do indirectly what it cannot do directly. Whether or not Parliament can do something indirectly, which it cannot do directly, may depend upon why it cannot do it directly … The true principle is that ‘it is not permissible to do indirectly what is prohibited directly’.

See Caltex Oil (Aust) Pty Ltd v Best [1990] HCA 53; (1990) 170 CLR 516, 522 (Mason CJ, Gaudron and McHugh JJ).

[91] See Tony Blackshield, ‘Engineers and Implied Immunities’ in Michael Coper and George Williams (eds), How Many Cheers for Engineers? (1997) 92.

[92] Julius Stone, The Province and Function of Law (1946) 171–3.

[93] Julius Stone, Legal System and Lawyers’ Reasonings (1964) 241–2.

[94] J A Simpson and E S C Weiner (eds), The Oxford English Dictionary (2nd ed, 1989) vol VIII, 966.

[95] Thus the limiting words in s 51(i) (‘with other countries, and among the States’) and in s 51(xxxi) (‘on just terms’) are in each case part of ‘a compound conception’, but are construed as imposing a limit in the second, restrictive sense. To subject other grants of power to the limit in placitum (xxxi) is now commonplace; to subject them to the limit in placitum (i) would result in the fallacy of reserved state powers.

[96] See above nn 7981 and accompanying text.

[97] WorkChoices Case [2006] HCA 52; (2006) 229 CLR 1, 346.

[98] Ibid 127 (Gleeson CJ, Gummow, Hayne, Heydon and Crennan JJ).

[99] On this reading, since s 51(xxxv) of the Constitution contains no ‘prohibition or restriction’, there is no need to consider how far any such prohibition might have extended to other Commonwealth powers.

[100] Stone, The Province and Function of Law, above n 92, 172.

[101] Stone, Legal System and Lawyers’ Reasonings, above n 93, 243: in such a case ‘the order of words is a kind of code adopted to convey the meaning. Reliance on the distinction is then entirely proper for it is no longer meaningless.’

[102] See especially Julius Stone, ‘Burden of Proof and the Judicial Process’ (1944) 60 Law Quarterly Review 262, making the point through a detailed analysis of Joseph Constantine Steamship Ltd v Imperial Smelting Co Ltd [1942] AC 154. See also Anthony R Blackshield, ‘The Legacy of Julius Stone’ [1997] UNSWLawJl 22; (1997) 20 University of New South Wales Law Journal 215, 228–35, showing how Dixon CJ dealt with similar problems by finding ‘substantial reasons’ for choosing one verbal formulation rather than the other (Alford v Magee [1952] HCA 3; (1952) 85 CLR 437, 451–2); or by basing the choice ‘upon considerations of substance and not of form’ (Dowling v Bowie [1952] HCA 63; (1952) 86 CLR 136, 140); or by applying the conventional distinction ‘in a less technical manner’, depending ‘not so much upon form as upon substantial considerations’ (Vines v Djordjevitch [1955] HCA 19; (1955) 91 CLR 512, 519).

[103] WorkChoices Case [2006] HCA 52; (2006) 229 CLR 1, 123 (Gleeson CJ, Gummow, Hayne, Heydon and Crennan JJ).

[104] Ibid 124–6 (Gleeson CJ, Gummow, Hayne, Heydon and Crennan JJ).

[105] Ibid 128 (Gleeson CJ, Gummow, Hayne, Heydon and Crennan JJ), citing Official Record of the Debates of the Australasian Federal Convention (Third Session), Melbourne, 27 January 1898, vol IV, 182 (John Quick).

[106] Ibid 128–9 (Gleeson CJ, Gummow, Hayne, Heydon and Crennan JJ).

[107] See also Australian Steamships Ltd v Malcolm [1914] HCA 73; (1914) 19 CLR 298 (‘Malcolm’), as to ss 51(i), 98; Re Maritime Union of Australia; Ex parte CSL Pacific Shipping Inc (2003) 214 CLR 397 (‘CSL’), also as to s 51(i); Telstra Corporation Ltd v Worthing (1999) 197 CLR 61 (‘Worthing’) as to s 51(v).

[108] WorkChoices Case [2006] HCA 52; (2006) 229 CLR 1, 129 (Gleeson CJ, Gummow, Hayne, Heydon and Crennan JJ).

[109] See Malcolm [1914] HCA 73; (1914) 19 CLR 298; Worthing (1999) 197 CLR 61.

[110] See CSL (2003) 214 CLR 397; Pidoto [1943] HCA 37; (1943) 68 CLR 87.

[111] See Stone, The Province and Function of Law, above n 92.

[112] WorkChoices Case [2006] HCA 52; (2006) 229 CLR 1, 275.

[113] Ibid 216.

[114] Ibid 206, citing Schmidt [1961] HCA 21; (1961) 105 CLR 361. See also above n 87 and accompanying text.

[115] Namely, to the general principle of interpretation that a constitution, like any other document, must be read as a whole ‘as one coherent instrument of government’, so that any particular provision must be read in the context of the whole: see WorkChoices Case [2006] HCA 52; (2006) 229 CLR 1, 208, 212.

[116] Ibid 210.

[117] Ibid 217.

[118] Ibid 218 (Kirby J).

[119] Ibid 244.

[120] Ibid 207–8, 226. Indeed his Honour goes further, though the above formulation would be sufficient to found his argument. He insists (at 226) that s 51(xx) does not support ‘laws that, in truth, relate to industrial disputes’ (emphasis added); and that ‘properly characterised, the Amending Act is one with respect to the prevention and settlement of industrial disputes’: at 229 (emphasis added). Although he seeks earlier in his judgment (at 203–4) to dissociate himself from the fallacy of a single objective ‘true’ characterisation, the majority (at 72) rightly condemns the italicised expressions as ‘fundamental constitutional error’. Kirby J (at 203) responds unrepentantly: ‘To have heresy alleged by those who participated in the joint reasons [in Combet v Commonwealth (2005) 224 CLR 494] … is an accusation to be borne with an easy heart.’

[121] Above n 89.

[122] See O’Connell v The Queen [1844] EngR 880; (1844) 11 Cl & F 155, 372–3; [1844] EngR 880; 8 ER 1061, 1143 (Lord Denman).

[123] WorkChoices Case [2006] HCA 52; (2006) 229 CLR 1, 190.

[124] Simpson and Weiner, above n 94, vol XVII, 128 (emphasis added).

[125] [1950] AC 235, 310 (Lord Porter).

[126] (1972) 124 CLR 529, 551.

[127] See above n 37 and accompanying text.

[128] WorkChoices Case [2006] HCA 52; (2006) 229 CLR 1, 118 (Gleeson CJ, Gummow, Hayne, Heydon and Crennan JJ).

[129] [1988] HCA 63; (1988) 166 CLR 79, 100 (Mason CJ, Deane and Gaudron JJ).

[130] (1982) 150 CLR 169, 211. Ironically, a similar provision now crops up, apparently unchallenged, as s 24 of sch 1 to the amended Workplace Relations Act 1996 (Cth).

[131] Another example is Australian Communist Party v Commonwealth [1951] HCA 5; (1951) 83 CLR 1, where the reasoning of Dixon and Fullagar JJ makes it clear that their estimate of the actual threat to defence might have been sufficient to support a less drastic infringement of rule‑of‑law values, while perception of a more serious threat might have sufficed to support even the draconian legislation in fact enacted.

[132] The argument would be similar in effect to the ‘degree of invasion’ test used as an aid to characterisation in decisions on the Constitution of India. Despite differences in the constitutional context (three lists of legislative powers rather than one) and in methods of characterisation (‘pith and substance’ rather than ‘sufficient connection’), the essential argument seems fairly transferable. As laid down by the Privy Council in Prafulla Kumar Mukherjee v Bank of Commerce Ltd, Khulna (1947) LR 74 Ind App 23, the idea is that, in determining whether impugned legislation falls within one head of legislative power, it is relevant to consider ‘the extent of the invasion’ of another head of power. This does not mean that ‘the validity of an Act can be determined by discriminating between degrees of invasion’; validity still depends on characterisation, but ‘degree of invasion’ is a relevant factor in the process of characterisation. Thus, so long as legislation is characterised as pertaining to one field of power, it will still be valid even if it ‘encroaches practically on the whole’ of another field of power: Bombay v Narottamdas Jethabhai [1950] INSC 40; AIR 1951 SC 69, 77 (Patanjali Sastri J). But if its impact on that other field ‘is so heavy and devastating as to wipe out or appreciably abridge’ that field, that may affect its characterisation: Gujarat University, Ahmedabad v Krishna Ranganath Mudholkar [1962] INSC 67; AIR 1963 SC 703, 721 (Subba Rao J).

[133] See generally Pidoto [1943] HCA 37; (1943) 68 CLR 87.

[134] WorkChoices Case [2006] HCA 52; (2006) 229 CLR 1, 218.

[135] [1943] HCA 37; (1943) 68 CLR 87, 101 (Latham CJ).

[136] Alternatively, these cases might simply reflect the particular history of s 51(x) and (xxx) of the Constitution. Thus, in New South Wales v Commonwealth [1975] HCA 58; (1975) 135 CLR 337, 471 (‘Seas and Submerged Lands Case’), Mason J explained as to placitum (x) ‘that control and regulation of fisheries beyond territorial limits was regarded as having such importance as to require its specific mention in s 51’: because the power had been included in the Federal Council of Australasia Act 1885 (Imp), its omission might have had ‘untoward significance’. A similar point can be made for placitum (xxx). Indeed, in 1898, Edmund Barton justified the overlap of placitum (xxix) with placita (x) and (xxx) in precisely those terms: see Official Record of the Debates of the Australasian Federal Convention (Third Session), Melbourne, 21 January 1898, vol IV, 30 (Edmund Barton).

[137] The dissenting judges do not explore this aspect of the plaintiffs’ case in detail, rather relying on Melbourne Corporation [1947] HCA 26; (1947) 74 CLR 31 as illustrating the continued need for a ‘federal balance’ (at 321–5, 330–43 (Callinan J)), or at least ‘the essential federal character of the Australian Commonwealth’ (at 226–9 (Kirby J)).

[138] WorkChoices Case [2006] HCA 52; (2006) 229 CLR 1, 169–72 (Gleeson CJ, Gummow, Hayne, Heydon and Crennan JJ).

[139] Ibid 171 (Gleeson CJ, Gummow, Hayne, Heydon and Crennan JJ).

[140] Ibid.

[141] Ibid 172 (Gleeson CJ, Gummow, Hayne, Heydon and Crennan JJ).

[142] Ibid 173–4 (Gleeson CJ, Gummow, Hayne, Heydon and Crennan JJ).

[143] Ibid 171–2 (Gleeson CJ, Gummow, Hayne, Heydon and Crennan JJ).

[144] Ibid 173 (Gleeson CJ, Gummow, Hayne, Heydon and Crennan JJ).

[145] Ibid 174 (Gleeson CJ, Gummow, Hayne, Heydon and Crennan JJ).

[146] Initially in Stock Motor Ploughs Ltd v Forsyth [1932] HCA 40; (1932) 48 CLR 128. See also Blackshield and Williams, above n 82, 391–402.

[147] [1948] HCA 13; (1948) 77 CLR 84, 120.

[148] See, eg, Botany Municipal Council v Federal Airports Corporation [1992] HCA 52; (1992) 175 CLR 453; Bayside City Council v Telstra Corporation Ltd [2004] HCA 19; (2004) 216 CLR 595.

[149] In addition, the supremacy of federal awards or workplace agreements is provided for by s 17 — a variant of the former s 152, as to which see Blackshield and Williams, above n 82, 389–91 — while s 18 stipulates that the preceding sections do not exclude the possibility of additional operations of s 109 of the Constitution.

[150] Thus, s 6(1)(a) invokes s 51(xx) of the Constitution by defining ‘employer’ to include a constitutional corporation. Similarly, s 6(1)(e) and (f) invoke the territories power (s 122 of the Constitution) by defining ‘employer’ to include an employer incorporated in a territory, or carrying on in a territory any ‘activity’ for which it employs or usually employs any individual. The majority holds that both these devices for extending the operation of the new regime are valid: at 156–9. Kirby J is disposed to agree: at 195–6.

[151] [2005] NSWIRComm 23; (2005) 139 IR 50, 64, 67 (Walton V‑P, Boland and Staff JJ).

[152] WorkChoices Case [2006] HCA 52; (2006) 229 CLR 1, 160 (Gleeson CJ, Gummow, Hayne, Heydon and Crennan JJ).

[153] For example, the sweeping exclusion of state laws dealing with unfair contracts and trade union rights of entry was contrasted with the very limited provisions on those subjects in the Commonwealth law.

[154] WorkChoices Case [2006] HCA 52; (2006) 229 CLR 1, 166 (Gleeson CJ, Gummow, Hayne, Heydon and Crennan JJ).

[155] Ibid.

[156] Ibid.

[157] Ibid 166–7 (Gleeson CJ, Gummow, Hayne, Heydon and Crennan JJ). The judgment goes on (at 167) to observe that s 16 ‘strongly resembles’ the provision considered in Wenn, and concludes (at 168) that Wenn is therefore ‘directly applicable’.

[158] [1931] HCA 34; (1931) 46 CLR 73. For Dixon J (at 98–102) the battle had already been lost in Roche v Kronheimer [1921] HCA 25; (1921) 29 CLR 329.

[159] Dignan’s Case [1931] HCA 34; (1931) 46 CLR 73, 101.

[160] Ibid 119–20.

[161] WorkChoices Case [2006] HCA 52; (2006) 229 CLR 1, 178–80 (Gleeson CJ, Gummow, Hayne, Heydon and Crennan JJ).

[162] Ibid 180 (Gleeson CJ, Gummow, Hayne, Heydon and Crennan JJ).

[163] Ibid 175 (Gleeson CJ, Gummow, Hayne, Heydon and Crennan JJ).

[164] Ibid 197–8.