Kirk, Jeremy --- "Constitutional Implications (I): Nature, Legitimacy, Classification, Examples" [2000] MelbULawRw 26; (2000) 24(3) Melbourne University Law Review 645

Constitutional Implications (I): Nature, Legitimacy, Classification, Examples

JEREMY KIRK[*]

[Much of Australian constitutional law involves the consideration of implications said to be found within the Australian Constitution. The development of the notion of ‘implied rights’ has brought the topic to prominence, but implications are of great significance in all areas of interpretation. This article, the first of a pair, examines how and when constitutional implications may legitimately be recognised. First, the article examines the nature of implications and how they are recognised, and sets out a list of factors to guide their legitimate ascertainment. Secondly, it discusses three systems for classifying implications, all of which offer some insights into the nature and legitimacy of constitutional implications. Finally, it explores leading examples of constitutional implications (especially the implied nationhood power and the implied intergovernmental immunities) to illustrate the propositions asserted. This analysis sets the scene for the second of the articles, in which the legitimacy and merits of a putative guarantee of equality, and of implied democratic rights, are examined.]

INTRODUCTION

The last two decades of the 20th century have seen significant constitutional activity and evolution in Australia. One of the most striking features of this time has been the development of the phenomenon of ‘implied rights’, that is, the judicial recognition of implied guarantees of individual rights and immunities within the Australian Constitution (‘Constitution’). The implied rights decisions challenged previously assumed or accepted legal boundaries. Inevitably, the cases raised foundational questions as to the nature of the judicial task in interpreting constitutions in general, and the Constitution in particular.

One such question, and the focus of this paper, is how and when constitutional implications may legitimately be recognised, both in relation to implied rights in particular and constitutional interpretation in general. The topic is of great importance when it is understood — as shall be shown — that for the Constitution ‘much of the greatest importance is implied’.[1]

The issue was put in stark relief by the radical decision of Deane and Toohey JJ, in dissent, in Leeth v Commonwealth[2] in June 1992. The judges asserted the existence of a far-reaching, implied guarantee of general legal equality.[3] Their method of implication relied on governmental doctrines and common law principles said to underlie the Constitution. The judges argued that the constitutional framers simply assumed that such doctrines and principles were entrenched in the Constitution.[4] They were faced with a powerful expressio unius argument that the series of constitutional provisions dealing with different types of discrimination indicated that no implied general guarantee of equality could have been intended. Yet the judgment purported to turn this negative into a positive, stating that the express provisions could be regarded as manifestations of the broader doctrine.[5]

Shortly thereafter the High Court handed down its landmark decisions in Australian Capital Television Pty Ltd v Commonwealth [No 2][6] and Nationwide News Pty Ltd v Wills,[7] accepting that the Constitution created the Commonwealth as a representative democracy, and recognising an implied limitation on federal legislative power. This limitation protects free communication on political matters as an essential element of this system of government.

The method of deriving implications from the doctrine of representative democracy subsequently became the source of vehement disagreement within the High Court. There was talk of the illegitimacy of deriving constitutional implications from representative democracy as though representative democracy were a ‘free-standing principle’.[8] In an apparent attempt to resolve this debate the Court, in its unanimous, joint judgment in Lange v Australian Broadcasting Corporation, indicated that democratic constitutional implications (and, implicitly, implications generally) could only legitimately be derived from the ‘text and structure’ of the Constitution.[9] However, invocation of such a mantra does little either to answer the difficult questions that arise in this area, or to guide judges in the choices they face when dealing with arguments as to implications. It also overlooks, and is inconsistent with, important tracts of Australian constitutional law.

The aim of this paper — the first of a pair of articles — is to analyse the nature, legitimacy and derivation of constitutional implications. The task is undertaken in the following manner. Part II addresses issues of interpretational theory and legitimacy which arise when considering whether and how to recognise constitutional implications, including what tests and considerations are relevant to the process. Part III analyses three different means of classifying implications, each of which offers some insight into the nature of implications. Part IV explores leading examples of the derivation of implications in Australian constitutional law.

This analysis enables an informed assessment to be made of the scope and merits of both the putative equality guarantee suggested in Leeth and the rights which flow from the recognition accorded to representative democracy. That examination is undertaken in the second of these articles, to be published in the next volume of this journal.[10] The second article will also illustrate the flaws inherent in the attempt to limit reference to constitutional ‘text and structure’ in constitutional interpretation.

II THEORY, LEGITIMACY AND IMPLICATION TESTS

A Defining Implications

The precise difference between express and implied meaning has been the subject of some dispute.[11] The essential distinction is whether or not the communicated idea is conveyed directly. This difference is one of degree,[12] for the directness of a communication will vary. A statement that supports an implication will usually have some other direct meaning which the implied meaning supplements. The indirectly communicated idea may be the important one. For example, to say pointedly to some unwelcome visitors, ‘the rain has stopped’, is to imply that they should leave.[13] Communicating this idea is the main object of the remark; the direct observation on the weather is incidental. In any case, the key characteristic of an implication is that provision X suggests proposition Z without directly stating it, whether or not X also directly communicates Y.

An example of the overlap between express and implied meaning arises from the injunction in s 80 of the Constitution that ‘trial on indictment ... shall be by jury’ for federal crimes. In Cheatle v The Queen the High Court paraphrased the question at issue in that case as whether this guarantee ‘carries with it a requirement that any conviction’ be unanimous.[14] This can be seen as simply a matter of filling out what the words ‘trial by jury’ mean (that is, as a matter of filling out the directly communicated meaning). The question can alternatively be regarded as involving an assessment of the extent to which the features of trial by jury in 1899–1900 were impliedly entrenched by the operation of the guarantee. The guarantee has direct and significant operation without the extra requirement of unanimity.

Sawer argued that matters of implication are synonymous with the resolution of ambiguity.[15] The process of implication frequently involves ambiguity or uncertainty, but ambiguity is neither necessary nor sufficient here. Some implications, especially logical ones, are perfectly plain. If I say ‘all swans are black or white and I saw one that wasn’t white’, I clearly communicate that I saw a black swan. As for sufficiency, if a law provides that ‘protesters must not cause disruption’, the key question is not what this indirectly communicates but what its exact direct meaning is (what constitutes disruption), if anything. Sawer is right to regard the s 92 guarantee that interstate trade ‘shall be absolutely free’ as fundamentally indeterminate on its face.[16] At issue is what idea is directly, if unclearly, communicated. To see the matter as involving implication distracts from this basic question and adds nothing.

B Implication Tests

The lack of any clear express–implied distinction is not of practical significance. The important question is always what the Constitution means; whether the ideas are communicated directly or indirectly is incidental. The distinction would only matter if there was a purported rule of interpretation preventing reliance on implications. Such a rule would be bizarre given the substantial role implications play in conveying meaning.[17] The very difficulty of delineation would make such an approach arbitrary and uncertain.

Despite these difficulties, the courts have invariably purported to treat implications as a special category. The historic common law test in statutory interpretation, applied in Australian constitutional law,[18] is to look for ‘necessary’ implications. This has overtones of claiming judicial objectivity in stating implications. The test might be thought to limit implications to matters of pure logic, but such an extraordinarily narrow approach is not what is meant. As Lord Eldon stated in 1813, ‘necessary Implication means, not natural Necessity, but so strong a Probability of Intention, that an Intention contrary to that, which is imputed ... cannot be supposed.’[19]

In practice, therefore, the traditional test requires stronger and more persuasive reasons to accept a particular implication than normally is required in accepting a possible construction. Dixon J indicated as much when he stated that constitutional implications should be based on ‘considerations’ which are ‘compelling’.[20] So, too, did Mason CJ when he required that such implications be ‘securely based’.[21] Mason CJ’s subtest for ‘structural’ implications — that they be ‘logically or practically necessary for the preservation of the integrity of that structure’[22] — similarly reflects a desire to limit what might be seen as a particularly indeterminate type of implication.

C Judicial Choice

The concern that motivates this restrictive approach to implications relates to preserving legitimacy in interpretation by confining judicial choice. The need for choice can arise here in the same dual manner as for any constitutional question. For the purposes of the discussion in this section it is assumed that one is considering the matter within a substantially originalist interpretational approach, that is, where the meaning of the Constitution is substantially understood or guided by the intended meanings and purposes of the Consitution-makers. Two types of uncertainty can arise whether the relevant Constitution-makers in Australia are regarded either as the constitutional framers or, instead, as the Australian people of 1899–1900 who ratified the Constitution (a matter which is open to normative debate).[23]

First, the communication of any actual intention of the Constitution-makers may be unclear. The further away one moves from the direct meaning of a set of words, the less sure one can be of the intended meaning. Implications are thus, of their very nature, likely to be more uncertain than express statements. Conclusions as to what is implied will invariably involve some degree of judicial choice, except where the implication is supported by pure logic. This association may explain why Sawer sought to define implications in terms of ambiguity.

Secondly, stating implications may involve choice because the Constitution-makers had not fully considered or agreed upon the ramifications or effects of the Constitution, nor on the interpretational issues which would require resolution. Determinate answers cannot then be found in the Constitution, for they do not exist. Such ambiguity can arise as much in relation to what is communicated indirectly as in relation to what is communicated directly.

What does it mean to say here that judges have choices? Clearly judges do not have a choice in the sense of choosing what colours to paint on a canvas. At the very least the putative construction must be a plausible reading of the document, as understood in its historical context. Thus, a guarantee of trial by jury may imply that jury decisions must be unanimous, but it cannot support an implication that parliamentarians must retire at 60.

One might draw an analogy with a scientist developing an hypothesis to explain a set of data, but this goes too far, for there is (generally) some essential physical reality to be discovered by science. In contrast, we simply cannot know whether s 80 was intended to entrench jury unanimity: the historical materials contain no clear answer on point, and it is likely that the issue was simply not considered in any authoritative way. Nevertheless, the Court had to resolve the question when it was asked this in Cheatle.[24] Judges must make a plausible choice which fits the data (that is, the text, along perhaps with certain extrinsic materials). This choice involves some element of creativity in the sense that no determinate answer can be found from that data alone.

Exactly where the judge’s role is placed between those of the scientist and the artist depends on the approach adopted. The respect, if any, paid to precedent may sometimes provide some restriction or direction. Certain factors — of the type set out below — can also be developed to guide the judge’s choice. Nevertheless, neither theories of precedent nor interpretational factors produce determinate answers, for these matters are not mathematical.[25]

A further, crucial, variable here is the basic theory or method of interpretation employed. The scope of legitimate choice increases as one moves away from the narrowest versions of originalism on the interpretative spectrum. If the importance or determinacy of text and/or original intent are downplayed, for whatever reason, there is then less guidance and constraint as to which constructions can reasonably be adopted. The theory that I have advocated — evolutionary originalism — involves the idea that the text generally is best understood as communicating broad ideas and concepts, and not tightly confined, hard-edged intentions.[26] In contrast, for a strict intentionalist the question will always concern the most likely actual intention of the framers, as ascertained from the text and possibly from extrinsic sources. This approach substantially narrows the potential for implications, not least because the framers simply cannot be taken to have considered the potential ramifications of many matters. Yet there will still be choice involved, for the evidence of intent will often be ambiguous, and questions asked must be answered.

Whatever interpretational theory, guiding factors and view of precedent are adopted, the area of judicial choice in constitutional interpretation can be confined but not eliminated.[27]

D The Nature of the Task in Recognising Implications

Judges have struggled with whether or not to acknowledge the uncertainty involved in implication, a struggle reflected in the language employed to describe the process. Dixon J acknowledged the element of judicial choice by talking of ‘making’ implications.[28] This perhaps goes too far in licensing creativity. Windeyer J, in a widely-approved passage,[29] stated: ‘I would prefer not to say “making implications”, because our avowed task is simply the revealing or uncovering of implications that are already there’.[30] Windeyer J’s main point is that implications must be genuinely referable to the Constitution. His view reflects the orthodox originalist search for intended meaning as communicated by the text.[31] This factor also explains his disavowal of constitutional implications ‘of The Moorcock kind’,[32] by which he appears to refer to judges implying terms into documents as a matter of judicial policy, rather than because of what was actually intended by the relevant parties.[33] Windeyer J is in error, however, in so far as his rejection of ‘making’ suggests that there is no necessary and legitimate element of choice involved.

What verbs, then, are appropriate? Words such as ‘assert’, ‘state’ or ‘propose’ do not imply a false degree of neutrality and certainty. Such expressions as ‘recognise’, ‘infer’, ‘ascertain’, ‘draw’, ‘derive’ and ‘identify’ are largely unobjectionable if understood not to carry overtones of determinism.

More fundamentally, the uncertainty here does not justify employing a specific interpretational test, such as ‘necessity’, to ascertain implications. The problem of ambiguity and uncertainty is endemic in constitutional interpretation. There is no justification for setting a higher onus of persuasion for matters deemed to be implications, when much the same issues arise in relation to ambiguous express meanings, such as the meaning of s 92 of the Constitution. Moreover, any such distinction is rather artificial when there is no clear division between express and implied meaning. The fact that uncertainty may tend to arise to a greater degree for implications does not justify treating any particular question differently merely because it may involve implication.

It is not unreasonable to say that implications must be ‘compelling’ or ‘persuasive’ or such like. However, the ultimate question is always whether the purported construction has sufficient support, given its nature and effects, within the context of the overall appropriate approach that is adopted to interpretation.

E Issues of Legitimacy

Certain specific legitimacy concerns arise when seeking to resolve constitutional uncertainty. Understanding these concerns can and should guide the process of interpretation, including the recognition of implications. They can be stated with relative ease here, for they are manifestations of the basic objections to judicial review which I have addressed at more length elsewhere.[34] The concerns arise within the context of originalism, broadly understood (including the theory of evolutionary originalism). These matters militate against recognising implications too readily. To the extent that this sets a higher persuasive threshold for many implications, it is because of the uncertainty attached, not because implication is involved per se.

The most fundamental concern relates to the fact that judicial review which enforces constitutional norms is antidemocratic.[35] This is so even if there is a democratic mandate for such review, although it can be conceded — without wishing to engage in an artificial exercise of fine balancing — that such a mandate does something to mitigate the antidemocratic element. However, implications raise particular problems as they are, by definition, not to be found expressed in the text and, thus, possess no clear and indisputable democratic mandate.[36] This concern arguably is heightened if one views popular sovereignty as a significant foundation of the Constitution’s authority,[37] as the people would have had far less opportunity than the framers or Imperial Parliament to consider and recognise the potential implications in the document that they approved at referenda.

Campbell makes the further argument that to imply rights is especially objectionable because ‘[s]carcely any aspect of political power is more important’ than determining what ‘priority interests’ override other considerations in society.[38] There is force in this objection. It implicitly questions the desirability of judges creating gaps in the reach of governmental power, for that is the ultimate difference in effect between enforcement of constitutional rights and, say, enforcement of federal divisions of power. Creating such gaps is the very purpose of constitutional rights, but in relation to implied rights this point merely raises again the question of whether or not that purpose is clearly manifest in, and mandated by, the text.

To an extent, however, the democratic mandate objection to implications is question-begging. The existence of any democratic mandate depends on the clarity with which any implication arises from the democratically-accepted Constitution. A very clear implication can be said to have a clear democratic mandate. This is a question of degree that correlates to the degree of certainty with which an implication can be stated.

In the end, therefore, what emerges from the democratic objection is that the less clearly an implication appears from the document, and the more restrictive the consequences of a particular limitation (particularly in excluding areas from potential governmental control), the greater the cause for caution.[39]

An overlapping legitimacy concern relates directly to the judicial creativity inevitably involved in recognising implications.[40] This concern is founded on dual objections: that overriding imposition of judicial views is undemocratic, and that such creativity is inconsistent with the accepted view that a legal system should aim to be certain, consistent and principled. Again, the strength of the concern depends on the relative degree of uncertainty and creativity involved. Thus, particular worry has been expressed that some asserted bases for constitutional implications, such as Murphy J’s reliance on the ‘nature of our society’,[41] are so open as to create ‘no logical limit’[42] and provide ‘no guidance or check to judicial aggrandisement or personal predilections’.[43]

This concern speaks both to the recognition and to the later application of implied propositions. The strength of the objection will depend not only on the degree of judicial choice involved in the first derivation of an implication, but also on the extent to which significant later choices are required to fill out the content of the limitation. With implied constitutional guarantees of rights, for example, it is likely that significant further judicial exposition will be required to clarify the content and effect of the implication.

Another concern involves the ‘political’ nature of the issues involved, an issue often said to be of particular relevance to rights guarantees. I have argued that any ‘political questions’ objection to constitutional rights is far from clear cut, and that public controversy is neither necessarily nor peculiarly associated with constitutional rights matters.[44] Nevertheless, at least some implied rights (and possibly other types of implication) may touch on matters that are of great public concern or which are central to disputes regularly fought within the political realm, such as issues of ‘economic individualism and collectivism’,[45] life and death, sexual conduct and privacy. Even leading proponents of constitutional judicial activism concede that judges should tread warily in relation to social and economic rights.[46] Yet these categories are not indisputably different from other types of rights.[47] If greater deference is due for social and economic issues, it is because of relatively low judicial expertise in such matters, the degree of polycentrism involved,[48] and the political centrality of most of the issues involving the allocation of social and economic resources.

For a court to decide on such controversial matters in any context, but especially within the overriding constitutional sphere, will often lead to public and political attacks with which ‘it is ill-fitted to engage’.[49] This may harm the institution.[50] The judiciary’s fear of damage to the rule of law flowing from criticism of the judiciary is probably overstated,[51] but it is not inconsequential. This very fear may lead the High Court to feel constrained in other areas because of the reduction in its political capital due to constitutional controversies. The less clear the legal basis for any controversial decision, the more vehement the public criticism is likely to be.[52] This does not mean that judges should avoid making controversial decisions; rather, the more centrally political a matter is, the more careful judges should generally be to find compelling bases for their decisions.

F Factors in Deriving Implications

The various concerns about legitimacy can be honed into a list of factors relevant to guiding judicial choices and, in particular, to the identification of constitutional implications. The factors will have different significance in different cases; none is definitive. The list is stated within the context of orthodox Australian originalism–literalism.[53] Indeed, the factors could be presented simply as presumptive guides to what was likely to have been intended, although to do so would tend to be artificial given that the need for choice often arises precisely because there was no formed intent.

Reference to these factors does not depend on any sharp distinction between express or implied meaning, nor on the presence of some particular degree of uncertainty or ambiguity. The factors are quite consistent with a genuine search for original meaning; thus, invoking them produces no radical difference in approach. They are useful and appropriate guides simply to the extent that uncertainty exists.

These factors overlap with the considerations I have listed as being relevant to interpretation within a theory of evolutionary originalism.[54] This is no coincidence: both sets provide guidance to judicial choice. Of course, some of the factors there are not listed here, because I do not seek to introduce an evolutionary element in this context. To the limited extent that the fourth factor — consequences — may have that effect, it is well supported by precedent. The legitimacy concerns that ground these factors also arise for evolutionary originalism, although that theory views the area of legitimate judicial choice in slightly wider terms than orthodox Australian literalism.

In assessing any putative implication, then, judges should take account of the following matters:

1. The strength of the positive imperatives supporting the implication (discussed below).
2. Textual manifestation and guidance. The less clearly the implication appears from the Constitution, the greater the caution that should apply.
3. Whether the implication ‘can be precisely defined’.[55] A putative proposition that is inherently vague or of unclear reach and effect should not readily be accepted. This reflects the value of clarity per se, a concern to limit the need for further judicial content-filling in the later application of the implication, and the premise that imprecise implications are less likely to have been intended. The difficulty of formulating the implied general federal immunity, for example, was accepted to count against its recognition, but this ultimately did not prove conclusive.[56] This precedent shows that the factor may sometimes be outweighed.
4. Consequences. The nature and significance of the consequences of adopting one interpretation over another have long been taken into account by the Court where there is uncertainty.[57] It is the general consequences for the constitutional system and the nation that are relevant and not, of course, any desire for one side or other to win the particular dispute. As an important example of the factor, to the extent that an implied limitation creates areas in the legislative (or executive) spectrum where governments may not reach, caution is appropriate. This element may be qualified by the particular nature of the restriction. For example, an implication that strengthens the democratic process is less objectionable on democratic grounds than one that simply imposes a substantive restriction.
5. Judicial administrability.[58] The more polycentric a matter is, or the farther removed from judicial experience and expertise, or the more difficult it would be for courts to oversee the issue (eg because of the complex fact-finding required), then the less ready a court should be to infer an implication covering such a topic. For example, any purported implication requiring allocation of economic resources would require significant support to be accepted. This factor will be relevant only in rare cases.
6. Political centrality of the issues. The more the issues covered by the implication are central to political views strongly held within society, the more careful a court should be to ensure that there is a compelling legal basis for any such decision.

The first factor, really a broad category, requires further explanation. There is a range of different types of arguments that can be used to support implications. These often seek to show that the framers, or the Australian people of

1899–1900, were likely to have intended to communicate such matters. For example, the fact that an issue is the subject of a pre-existing, relevant constitutional principle or doctrine may increase the credibility of an argument that it was intended to affect constitutional interpretation. Other arguments here relate to what it is reasonable or appropriate to see as communicated, even if not originally considered by the Constitution-makers. Relevant changes in modern circumstances would be considered here if a dynamic element was introduced to the applicable theory of interpretation.

One common form of argument within this category is the assertion that some implication is practically necessary, and thus probably intended, to give full and effective operation to the purpose of some provision or provisions.[59] Thus, if the Constitution creates or presupposes a particular institution, it may be reasonable to recognise an implication protecting that institution’s existence and ability to function. There has been some suggestion that such institutional efficacy arguments require a special ‘practical efficacy’ test of implication.[60] In contract law, similarly, there has been some suggestion of a separate ‘business efficacy’ test for analogous implied terms.[61] Yet there is no reason to see such implications as unique, here or in contract law.[62] It is simply one type of positive argument supporting an implication.

III CATEGORISING IMPLICATIONS

Encompassing taxonomies of implications have limited utility. The distinctions made are invariably ones of degree, which prevent the statement of simple conclusions as to what types of purported implications should or should not be recognised. The ultimate question is always what is supportable within the appropriate theory of interpretation. Nevertheless, these systems can offer some useful, if not definitive, insights. In this Part, three different systems of classification are examined.

A System 1: Intent

Jeffrey Goldsworthy has sought to classify different types of implication within the context of his broader argument for moderate originalism.[63] For him, implications can arise only as manifestations of the framers’ actual intentions. His categorisation relates to different ways in which such intentions are manifest. A clear aim of this project of classification is to limit the legitimate scope for inferring implications from the Constitution.

Goldsworthy distinguishes implications arising from:

• pure logic;
• deficient expression (where a failed attempt was made at express communication but it is still possible to guess what the speaker ‘really meant to say’);[64]
• deliberate and conscious intention (where the speaker knowingly chooses to communicate a proposition by implication); and
• ‘implicit assumptions’.

1 Implicit Assumptions and Related Matters

This category requires some explanation. Goldsworthy is right to identify it as being of some importance. Implicit assumptions are matters not actually considered but simply taken for granted by the speaker. They relate to the context in which words are uttered, and give colour to what is meant. This category serves to emphasise the importance of identifying the time and place in which a statement is made if any more precise meaning is to be identified; otherwise one is left only with the range of dictionary definitions of the words employed, with little means to choose between any of these alternative definitions. If a guest requests a glass of water, then he or she probably implies that the water is desired soon, and that it should be of a quality and quantity fit for human consumption. This is so even if the guest has not considered such matters. These meanings arise from general understandings of human needs, practices and expectations. The implicit assumptions are crucial to successfully communicating the basic idea. If one is searching for the likely intended meaning of the Constitution-makers, as one is within originalist theories, then implicit assumptions will be involved in the process of construction.

A number of Australian dicta might be understood, incorrectly, to suggest that assumptions have no interpretational relevance. Three matters must be distinguished. First, there are those implicit assumptions and background doctrines that do appropriately influence interpretation. For example, it was assumed within Australia that the doctrines of parliamentary supremacy and the rule of law would apply, at least to the extent reconcilable with a rigid constitution. To take a more prosaic example, we can reasonably presume that the reference to ‘any court of competent jurisdiction’ in s 46 of the Constitution refers to courts with relevant authority to address the question rather than to courts which had proved themselves legally skilful or meritorious.

Secondly, there are the mere expectations of the Constitution-makers as to how the Constitution would operate in practice. There is a difference between expecting a system to operate in one way and implicitly requiring it to do so. A court may expect a plaintiff to spend awarded damages on health care, but does not usually require it. This category is different from ‘application intentions’,[65] which may include mere practical expectations, but also cover more directional intentions as to how provisions should operate in relation to particular matters.

It was mere practical expectations to which Dixon J referred in refuting the argument that the Commonwealth’s power under s 51(i) of the Constitution did not extend to Commonwealth participation in trade or commerce. The argument was said to confuse the framers’ ‘unexpressed assumptions’ with the ‘expressed meaning of the power’.[66] The same is true when he distinguished ‘assumptions’ from actual ‘restrictive intentions’ in rejecting the argument that the banking power (s 51(xiii)) did not extend to prohibiting banking.[67] Mason J invoked the same category in dismissing the relevance of the possible ‘expectations’ that the external affairs power (s 51(xxix)) would have a limited operation due to the narrow range of international treaty-making in 1900.[68]

Thirdly, there are matters presupposed by the text. For instance, provision in s 73 of the Constitution for appeal to the High Court from State Supreme Courts presupposes the existence of the States’ courts. Stephen Donaghue argues that such a presupposition is constitutionally protected/required if its existence is necessary to prevent the ‘truth or efficacy’ of the provision being ‘called into question’.[69]

Textual presuppositions may sometimes give rise to implied requirements, but Donaghue’s test is misguided. He appears to regard this test as appropriate for all implicit assumptions. Many such assumptions are by no means logically necessary, yet are still appropriately regarded as affecting interpretation. Parliamentary supremacy is an example.

Furthermore, presupposition does not inevitably imply requirement. The operation of a provision may have been intended simply to be conditional on the continuation of the presupposed matter. A statute granting an allowance to homeless people does not require that some people must be without homes. The presupposition may also be unimportant and incidental. Donaghue argues that the High Court was wrong in Clayton v Heffron[70] to reject the argument that the (then) reference in s 15 to the ‘Houses’ of State Parliament prevented the abolition of Queensland’s upper House.[71] Yet it is unlikely that such an incidental use of the plural term, which happened to be descriptively accurate in 1899–1900, really manifested an intention to require the existence of dual State Houses.

How, then, does one distinguish mere practical expectations from implicit assumptions, and distinguish between those presuppositions that do and do not imply a requirement? There are no easy answers. As always, one must look to what construction is supported by analysis within the context of the appropriate approach to interpretation. Within any originalist theory, one will begin by asking what was likely to have been intended.

2 Deliberate Implications

Goldsworthy argues that it is unlikely that framers of legal documents would deliberately leave their intentions in the uncertain realm of implication.[72] He essentially excludes the possibility of deliberate implications from consideration.[73] They are not so easily dismissed.

Constitutions such as Australia’s tend ‘rather to outline principles than to engrave details’.[74] The Constitution is by no means a legal code; it is reasonable to suppose that many matters were consciously left in the realm of implication, tied to the general principles and framework established by the document. The framers and (some of) the Australian people of 1899–1900 were well aware that implications would be drawn. They also implicitly accepted that this would depend upon some degree of non-deterministic judicial choice. The insertion of the s 116 guarantee of religious freedom reflects both these points, as its main apparent purpose was to allay widespread concerns that the reference to God in the preamble to the enacting Act might be taken, contrary to what was actually intended, to imply a Commonwealth power over religion.[75]

Moreover, the lines between these categories are indistinct. The difference between deliberate implications and implicit assumptions is one of conscious thought. Yet some matters may have been just on the peripheries of the Constitution-makers’ minds. In any case, determining whether the matter has been the subject of conscious attention will often be a matter of surmise. In relation to another of Goldsworthy’s categories, the fact that an implied proposition is conveyed ‘in an elliptical way’ does not mean that deficient expression is involved, as he perhaps implies.[76] Brevity is not necessarily deficiency. Drafters frequently and deliberately rely on elliptical communication for reasons of efficiency.[77]

3 Implications from Internal Tensions

Goldsworthy’s analysis is also incomplete. At least a fifth category can be identified within this system relating to the intentions of the Constitution-makers. A complex set of propositions, when taken together, may contain internal tensions. The complexity of any constitutional system is such that the details and ramifications of all the multifarious relations between constitutional propositions will be far from clear. Resolving such tensions involves considering what is implied as to the priority and consistency of the different imperatives. These implications may be called implications from internal tensions.

To take a simple example, s 7 of the Constitution indicates that the Senate ‘shall be composed of senators for each State’. Section 122 states that the Commonwealth may provide for representation of the Territories in either House of federal Parliament on such terms as it thinks fit, potentially enabling the Parliament to swamp what had been conceived of as the States’ House with territorial Senators. There is an internal tension. When faced with the issue, the Court had to make a choice.[78] It could have approved territorial Senators, disallowed them, or permitted them with restrictions. Any decision required an assessment of what was implied by the two sections in all the circumstances, for example, whether or not s 7 implied that the Senate should be composed only of State Senators, or perhaps that only State Senators could have full voting powers.

The resolution and consequences of such internal tensions may not have been directly considered by the Constitution-makers. Yet neither logical implications nor implicit assumptions are necessarily conceived of by the speaker. Internal-tension implications do involve an element of judicial choice, but these implications are substantially derived from a contextual assessment of the meaning of the words and the nature of the structures created. For that reason they can be seen as real implications, that is, indirect communications. They are not necessarily implications by deficient expression (although this may sometimes be the case), because that category involves a failed attempt at actual communication, whereas here the relevant intentions may well have been only partially formed, or the issue may not have been considered. They are not necessarily deliberate implications, and to suggest that the resolutions were simply assumed would be to sidestep the issue. Internal-tension implications might sometimes be deliberate, or reflect deficient expression, but the point is that such implications cannot adequately be explained within Goldsworthy’s four categories.

Many constitutional implications might be seen as involving internal tensions — the federal intergovernmental immunities are leading examples.[79] Arguably, whenever there is a constitutional imperative that potentially restricts Commonwealth power, such as the requirement that the Parliament be chosen by elections, this exists in tension with the affirmative grants of Commonwealth power as understood in light of the doctrine of parliamentary supremacy. Resolving such tension can be seen as a legitimate, indeed unavoidable, matter of relying on relevant internal-tension implications.

4 Utility of System 1

Overall, therefore, Goldsworthy’s distinctions have some descriptive merit within originalism, but his attempt rigorously to limit the recognition of implications is unsuccessful. His downplaying of deliberate implications is unjustified, his categories are somewhat indistinct, and his taxonomy is incomplete within its own terms.

B System 2: Textual Manifestation

Some useful insights can be gained by classifying implications according to the manner in which the text manifests the implied meaning. No precise set of categories emerges, but it does help to explain certain important matters. This system relates strictly to the reasoning and textual basis supporting an implication, not to the implication itself. Thus, an implication may be supported by alternative textual arguments that fall into different categories here.

Broad, overlapping classes can be set out in these terms:

• mono-provisional;
• multi-provisional;
• inductive; and
• no direct manifestation.

1 Mono-Provisional Implications

Mono-provisional implications are those manifested by a specific constitutional provision. The point here is not that the implication arises from the provision construed in isolation, but that the textual premise supporting the implication substantially arises from one particular provision. The provision’s interpretation may still be coloured by its historical and textual context.

Section 75(v) of the Constitution, for example, states that the High Court shall have original jurisdiction over matters in which the remedies of mandamus, prohibition or injunction are sought against officers of the Commonwealth. It is not yet conclusively resolved whether this jurisdiction extends to the issuing of a writ of certiorari by itself.[80] The express mention of the other remedies might be taken to imply that certiorari is not available. On the other hand, the power to issue certiorari might be implied on the principle of giving substantive and effective operation to the provision of the other remedies and the purpose of the subsection. In either case, the question is one of implication, and the main textual premise supporting the implication arises from s 75(v) alone.

2 Multi-Provisional Implications

This category encompasses those implications manifested by two or more provisions when read together. Obviously, these differ from mono-provisional implications only by degree. One example is the issue of whether or not the State and federal Parliaments can together legislate to confer State jurisdiction on federal courts. This depends significantly on what is implied by the nature of the federal judicial system created by the whole of Chapter III of the Constitution. Part of the reasoning of the majority in Re Wakim; Ex parte McNally was that the detailed specification of jurisdictional matters in Chapter III supported a textual premise that such issues were dealt with exhaustively by express words, implying that federal courts cannot exercise State jurisdiction because of the absence of express provision for this.[81]

3 Inductive

The third category covers implications that are derived from the text in an inductive manner. A number of provisions are taken to be merely particular manifestations of a general doctrine, concept or principle. That doctrine (etc) is then used as a basis for implications. The textual premise arises from a series of provisions, just as for multi-provisional implications, but the nature and purpose of these provisions are characterised at a high level of abstraction. The two key distinguishing features of this approach are that the textual provisions tend to be subsumed within the broader doctrine, and that the doctrine is then employed as a basis for implications which travel beyond what is apparently communicated by the provisions themselves. The technique was employed by Deane and Toohey JJ in Leeth[82] to derive their equality guarantee.

Substantial legitimacy concerns arise here. The category represents a movement away from according primacy to what is communicated by the text, although the text is not abandoned altogether. This move allows significant scope for judicial choice in stating what the doctrine implies, particularly in identifying which version of the particular concept is involved and to what extent it is incorporated. Indeed, substantially different doctrines, or different versions of a doctrine, may sometimes be consistent with the set of provisions relied upon.[83]

This category cannot simply be rejected, however. The method has, it seems, previously been employed in Australian constitutional law. Arguably, the doctrine of responsible government, which has been invoked and relied upon by judges in interpretation,[84] is derived inductively from the ‘machinery’[85] provisions dealing with aspects of the doctrine in ss 62 and 64 (there is not space to develop the argument here). The general doctrine of the separation of powers also appears to rely on an inductive approach.[86]

The method is plausible. If provisions make little sense unless interpreted in the light of some more general doctrine, then it is reasonable to attribute interpretational significance to that doctrine. If the Constitution-makers had wished to incorporate a particular doctrine, they could have done so in clearer and more detailed ways, yet they may not have fully considered the ramifications of the imperatives they created.

Nevertheless, because of the substantial legitimacy concerns and the relatively low probability of induction having been intended (assuming an originalist approach to interpretation), any claim that an external doctrine is incorporated by induction will require compelling justification.

4 No Direct Manifestation

Some implications may legitimately arise although they have no direct and meaningful textual manifestation. This category covers both implicit assumptions and some deliberate implications (to hark back to aspects of the first system of classification). Examples include the doctrine of parliamentary supremacy and aspects of the rule of law. Neither of these is manifest in the text, yet both have shaped its interpretation.[87] Needless to say, if no particular text can be pointed to in support of a claimed proposition, then the merits of the claim must be assessed with some wariness. However, the Constitution-makers did not and could not express all their multifarious intentions in the text.

5 ‘Structural’ Implications

There has been a recent tendency in the High Court to speak of ‘structural’ implications, particularly in relation to federal intergovernmental immunities[88] and democratic implications.[89] Mason CJ argued in the Political Advertising Case, in a much-approved passage,[90] that structural implications ‘must be logically or practically necessary for the preservation of the integrity of that structure.’[91]

But what structure is being referred to? In so far as judges are referring to implications arising from the general internal structuring (meaning the ordering) of the Constitution itself,[92] they speak of multi-provisional or inductive implications. The label has also been used when talking of requirements that are implied in order to protect or maintain particular structures (meaning institutions or systems) created by the Constitution, such as federalism or representative democracy. This sense refers not to the textual manifestation of the implication but to a positive reason for arguing that the implication should be recognised.[93] The two usages overlap because such institutions or systems as federalism or representative democracy are often established by a series of provisions. Nevertheless, the usages involve different issues.

C System 3: The Relative Importance of Text

The third system of classification is closely related to the second. It addresses the degree to which the text itself plays a role in answering an interpretational question, that is, the importance of the particular words in reaching a conclusion as to the meaning of the Constitution on any point. This covers all aspects of constitutional interpretation, not just implication. The second system of classification addresses the manner in which an implication is manifest by the text (if at all), in the sense of whether it is none, one, or a series of provisions which support the implication. Yet, however the implication is supported, this does not directly address the issue of the extent to which it is the actual words of the text which fill out the relevant meaning and content of the constitutional proposition and thus answer the interpretational question being considered. That issue is the focus of this third system.

Broad, overlapping categories can be stated as follows:

• text-dominant;
• purposive;
• incorporated doctrines; and
• no significant role.

1 Text-Dominant

For the first category, it is the words of the text that provide the main focus for answering the particular interpretational question. This is not to deny that some judicial choice may be involved, nor that the words may be seen as communicating some broader purpose.

An example is the interpretational question in New South Wales v Commonwealth:[94] whether the Commonwealth’s power over trading and financial corporations (s 51(xx)) extended to the incorporation of such companies. This issue hinged on whether the constitutional phrase ‘formed within the limits of the Commonwealth’ was meant in a temporal and/or spatial sense. Thus finding the answer to the question centred on the meaning of the particular words employed.

2 Purposive

Purposive interpretation involves seeing a provision in light of its apparent purpose, as ascertained from the words themselves and the general background context (including, potentially, from reference to extraneous materials). This category substantially overlaps with the first. However, engaging in purposive interpretation sometimes involves giving a ‘strained construction’ to the words employed.[95] Alternatively, the words may communicate very little. For instance, the protracted dispute over the meaning of the prohibition in s 90 of the Constitution of non-federal ‘excise’ duties has centred on the provision’s claimed purpose,[96] for the word itself has no clear connotation.[97] In either case, the actual wording becomes secondary to the purpose in answering the question posed.

3 Incorporated Doctrines

The third category describes situations in which a doctrine, concept or principle, which has some independent external existence, is to some extent incorporated into the Constitution by reference. In addressing relevant interpretational questions it is then that doctrine, more than the text, which is the focus. A doctrine might be incorporated in its past incarnation, or in an evolving form.

This category differs from purposive interpretation only by degree. In both cases the particular wording of the text has limited significance. Here, however, the significance of what is referred to is rather larger, and a doctrine will have some prior independent existence in law, politics, economics, philosophy and so forth. A mere purpose tends to be both more confined, and have no significant, separate existence apart from the matter in question.

A doctrine can be incorporated by express reference or allusion. For example, s 80 entrenches the institution of jury trials for federal indictable offences. Guidance as to what is meant by a ‘trial by jury’, in relation to such matters as representativeness and unanimity of the jury, is filled out not by the words but by the historical concept and the pre-existing common law. Whether these characteristics are themselves best regarded as express or implied is a moot point. Another possible example is the s 92 guarantee that interstate trade ‘shall be absolutely free’; though which external doctrine is incorporated by these ambiguous words has been the subject of much dispute throughout the 20th century.

A doctrine could be incorporated simply by expressing all its relevant aspects in the text. However, interpretational questions would then be answered substantially by reference to the text, thus tending to fall into the first and second categories of this third system. It would not be an incorporated external doctrine in the sense employed here.

Doctrines can also be incorporated by implication, textually manifest in any of the first four ways described above in the second system. Inductive argument necessarily involves incorporated doctrines, for the notion of inductive argument, as defined above,[98] involves referring to external doctrines. The reverse is not true, however, for a doctrine can be incorporated expressly or by implication in a number of ways.[99]

The idea of referring to impliedly incorporated external doctrines has caused much unease and confusion in Australia over the course of the last century. Such reference was seemingly the target of the High Court’s statement in Lange v ABC that democratic requirements were only entrenched to the extent required by the Constitution’s ‘text and structure’.[100] Yet such doctrines have played a fundamental role in Australia in relation to, inter alia, the implied nationhood power, the federal intergovernmental immunities and the separation of powers.[101] If the Constitution-makers chose to indicate (expressly or impliedly) that a pre-existing doctrine should be given constitutional effect, though not fully set out in the terms of the Constitution, then originalism requires that their intention be given operation. In contract law, analogously, the incorporation of terms by reference to another document is well accepted.[102]

Brennan J has stated in relation to the implied federal immunities that ‘[i]t is impermissible to construe the terms of the Constitution by importing an implication from extrinsic sources when there is no federation save that created by the express terms of the Constitution itself’.[103] This statement has been quoted with approval by Dawson J, who was concerned to prevent implications from ‘extrinsic circumstances’.[104] Yet it is ambiguous. Brennan J might have been suggesting that constitutional questions can and must be resolved only by reference to sources within the four corners of the Constitution. Such a strict interpretivist position is unsustainable, for the Constitution provides few determinate answers.[105] He may have been seeking to rule out reference to external doctrines such as federalism, at least in so far as they are not clearly set out in the text. If by this he meant that no reference can made to external doctrines to fill out the text, then his view is inconsistent with important parts of Australian constitutional law. If he meant to say that external doctrines cannot be referred to unless there is a textual justification for doing so — a view McHugh J has expressed[106] — then his statement is unobjectionable, subject to a possible dispute in relation to incorporation by assumption.

McHugh J has asserted that ‘[u]nderlying or overarching doctrines may explain or illuminate the meaning of the text or structure of the Constitution but such doctrines are not independent sources’ of powers and rights.[107] This points to a purported distinction, emphasised by others, between taking account of a doctrine as an aid to constitutional interpretation and giving direct effect to that doctrine.[108]

If an external doctrine is invoked in order to resolve ambiguity, or to fill out the text, or to influence interpretation in some way, then that doctrine is given some effect to the extent that it makes a difference. The Court in Cheatle[109] did not merely take account of the historical requirement of unanimity in jury verdicts; it gave this aspect of the concept hard effect. The particular construction would not have been adopted but for the influence of the external doctrine. The common law and historical concept of the jury trial was treated as incorporated in the Constitution to that extent. One could argue over whether the historical concept was an ‘independent source’ of limitations, but the important question is whether the doctrine is incorporated to the relevant extent, so that particular aspects of it are given effect. If an external doctrine has little causal effect on interpretation, being employed just to confirm a construction adopted on other grounds, then perhaps one could say that it was merely taken into account. However, any distinction here is really one of degree relating to the influence of the doctrine, and thus to the extent to which the doctrine is incorporated.

Distinguishing between the implied incorporation of a doctrine in multi-provisional and in inductive ways can be difficult. The distinction is one of degree. In both instances a series of provisions is claimed to support a premise that an external doctrine is given some constitutional status. There are two practical differences. First, with induction the supporting provisions are treated as mere manifestations of a general doctrine. The text forms less of a limit than for multi-provisional incorporation because it is possible, using the inductive argument, to view the supporting provisions within a wide range of abstraction. It is then relatively easy to assert that all or much of the relevant doctrine is impliedly incorporated, depending on how the supporting provisions are characterised. Similarly, there is likely to be more room for arguing that a broader, substantive conception of a particular doctrine is incorporated. Secondly, for inductive reasoning the supporting provisions are likely to have reduced significance, as their specific effect tends to be subsumed by the broader doctrine. This is a characteristic of Deane and Toohey JJ’s implied equality guarantee, for example.

In assessing an incorporated doctrine argument, therefore, the wider and more substantive the doctrine that is incorporated, and the less the manifesting provisions are left with a meaningful role, the more likely it is that inductive reasoning is involved. The key questions relating to any incorporated doctrine are which particular version of the doctrine is incorporated and to what extent. The more one moves away from an attachment to the text, the less clear the answers to these questions will be. Thus it is that substantial legitimacy concerns arise when a doctrine is said to be incorporated by induction.

4 No Significant Role

The last category, where the text plays no significant role in answering a question, is interconnected with the category of ‘No Direct Manifestation’ in the second system.[110] It covers some implicit assumptions and deliberate implications (to use terms from the first system).

IV IMPLICATIONS IN AUSTRALIAN CONSTITUTIONAL LAW

Implication plays a substantial role in conveying meaning. As Dixon J stated, a prohibition on implication ‘would defeat the intention of any instrument, but of all instruments a written constitution seems the last to which it could be applied.’[111] Certain leading examples of implication in Australian constitutional law are examined in this Part, both in order to illustrate the distinctions and arguments made above and to prove the ubiquitous nature of implications.

A The Limits of Powers

Issues of implication are crucial in resolving the exact limits of Commonwealth powers. Such implications usually emerge from single provisions, and the substantive source of guidance is generally the wording or the perceived purpose of these provisions. They can arise in relation to the central areas of power, although these issues will often involve borderline express/implied meanings. One example is the now-accepted view that the external affairs power, s 51(xxix), extends to the legislative implementation of international treaties. In Victoria v Commonwealth the majority treated this aspect as arising because it was ‘at least implicit’ that the power would authorise federal legislative implementation of treaties entered into by the Commonwealth, given that the power historically extended to federal implementation of treaties entered into on behalf of Australia by Britain.[112]

Questions of incidental power more clearly fall within the implication arena. The accepted general proposition that grants of power extend to matters incidental or ancillary to the main operation of the power, even aside from the express incidental power (s 51(xxxix)), is itself an implication. The preceding English common law saw the matter as one of implication,[113] and that is how Australian commentators have regarded it.[114] In fact, given that the principle was ‘so well understood in English law that it would have been superfluous to incorporate it in an express provision’,[115] it is likely that deliberate implication was involved.

The specific extent of the incidental aspect of a power will thus also generally be a question of implication. For example, the power over taxation (s 51(ii)) impliedly extends to many matters beyond the simple imposition of a tax, such as provisions ameliorating the effects of a tax.[116] The grant of a power to tax can reasonably be taken impliedly to extend to such matters so as to give the Commonwealth full and effective control of its power.

In practice modern judges have rarely acknowledged the implied nature of the incidental area of powers.[117] This illustrates a reluctance to admit the element of uncertainty involved; to accept that one has entered the dim realm of implication is to admit the potential for judicial choice.

B The Implied Nationhood Power

Some Commonwealth powers arise only by way of implication. A leading example is the ‘implied nationhood power’.[118] It has been widely accepted that the Commonwealth has power, derived from its character and status as a national government, to (1) protect itself against subversion,[119] and (2) engage in or support ‘enterprises and activities peculiarly adapted to the government of a nation’,[120] such as cultural or scientific endeavours or the celebration of national events and symbols.[121]

Dixon J saw the power over subversion as an implied legislative power.[122] The then prevailing view, however, was that it arose under the executive power in s 61, which ‘extends to the execution and maintenance of this Constitution’.[123] Such executive power can be given legislative expression by the express incidental power (s 51(xxxix)).

In relation to the less regulatory second aspect, the general view has been that it is, at least, an implied aspect of the executive power,[124] but may also be an implied legislative power.[125] Yet Wilson and Dawson JJ,[126] with whom Toohey J generally agreed,[127] went to some lengths in Davis v Commonwealth to argue that there was no implication involved, apparently motivated by a wariness of implications per se, and a quasi-interpretivist refusal to accept that powers can be affected by matters beyond ‘the confines of the Constitution’.[128]

It is certainly not unreasonable to see the subversion aspect as covered by the express extension of the executive power to the ‘maintenance of this Constitution’. The celebratory-cultural aspect of the power does not easily fall within this phrase, however. To see it as a part of the ‘executive power’ granted in s 61 is feasible, but it is not practically or incidentally required by those two words and it is not communicated directly by them.

In any case, Wilson, Dawson and Toohey JJ appeared to accept that the celebratory aspect is supported by the character and status of the Commonwealth as a national government.[129] This acceptance implicitly concedes that the words of s 61 by themselves do not support the power. The trappings of nationhood are neither spelt out in the Constitution nor arise as an inherent and determinate part of being a nation state. Instead, that notion must be developed by reference to history, practice, custom and expectations in Australia, Britain and the world. The implied nationhood power is therefore derived from an understanding of nationhood that is external to the Constitution. In other words, in terms of the third system of classification, the interpretational work is done by reference to an external concept which is taken to be incorporated to some extent in the Constitution.

In relation to the second system of classification, one could strive to present the celebratory aspect of the power as finding its textual support in s 61 alone. Yet this allocation seems more to reflect a desire to tie it to some particular section, of which s 61 is the most plausible, than to be because of the character of s 61 itself. The implication really flows from what are deemed to be the incidents of nationhood. And in the words of Brennan J, ‘[t]he Constitution summoned the Australian nation into existence’,[130] and it was the Constitution as a whole that did so. This aspect of the implied nationhood power, therefore, could be seen as a multi-provisional implication, but given the lack of a clear connection to any particular provisions it may be more accurate to say that it has no direct manifestation in the text (the fourth category in the second system). Whilst this does raise legitimacy questions, the consequences of the celebratory-cultural aspect are limited, for it has proved an innocuous power.[131]

C Federal Immunities

The issue of implied intergovernmental immunities, a dominating theme of the first two decades of Australian federal constitutional jurisprudence, was not quieted by the Engineers’ Case[132] for long. Beginning with Melbourne Corporation v Commonwealth[133] in 1947, it has become well accepted that there are two related limbs of State intergovernmental immunity implied in the Constitution.[134] First, the Commonwealth may not discriminate against the States, either individually or as a group. An approximately reciprocal prohibition seemingly restricts the States in relation to the Commonwealth.[135] Secondly, the Commonwealth may not act so as to destroy or curtail the existence of the States or their capacity to function as governments (the ‘general immunity’).

Two cases have resulted in the invalidation of federal laws for discrimination. The second, QEC v Commonwealth[136] (in 1985), concerned a law requiring that a particular industrial dispute, involving a Queensland government instrumentality, be dealt with in a distinctive manner by the federal Conciliation and Arbitration Commission. In the first case, Melbourne Corporation v Commonwealth,[137] a law prohibited private banks from providing banking services to State governments in an attempt to force the States to bank with the Commonwealth. Dixon, Rich and Starke JJ held that this breached the implied prohibition. Latham CJ also invalidated the law, but apparently sought to avoid relying on implications. Although he viewed the measure as sufficiently connected to the banking power (s 51(xiii)) on the ordinary approach to constitutional characterisation,[138] the fact that it could be characterised as a law with respect to the States in particular was taken to preclude validity.[139] This type of approach also seems to have been employed by Williams J.[140] Barwick CJ and Owen J applied this approach in the Payroll Tax Case.[141] However, the whole Court has subsequently rejected the method and accepted the implication route.[142]

This characterisation detour illustrates again how far some judges have gone to avoid being seen to enter the realm of implication. Yet the adoption of a special approach to characterisation when the States were adversely affected must have depended on something in the Constitution that indicated that they were deserving of special attention. There is no express provision to that effect. The approach must therefore have been based on an implication sounding from the federal system, as Williams J admitted.[143]

The other, general limb of immunity, though difficult to formulate, has proved capable of remarkably precise application. The perennial question has been the extent to which the Commonwealth may regulate the employment conditions of State government employees. A six judge majority in Re AEU[144] drew a detailed line between what did and did not constitute an interference in the States’ ability to function as governments, addressing what types of matter could be regulated for what types of State employee.

A case can be made against both of the implied federal immunities on expressio unius grounds. Section 114 of the Constitution prevents the two levels of government taxing property belonging to each other. Further, ss 51(ii), 51(iii) and 99 prohibit the Commonwealth from discriminating between States or parts thereof in laws relating to taxation, bounties, revenue, trade and commerce. Though of general effect, these limitations protect the States themselves to some extent. Such provisions support an argument that implied limitations were not intended beyond the extent expressly provided.[145] Of all the judges who have accepted the implied immunities, none has directly addressed this expressio unius point.[146]

A strong argument can be made, however, that the Constitution does not expressly resolve the issue of intergovernmental immunities. Section 114 is directed only to the question of the two levels of government taxing each other, not to the broader issue of whether and to what extent they can bind each other (although, admittedly, it does suggest that they can). The other prohibitions do not directly address intergovernmental immunities.

In the Engineers’ Case the majority derided implication ‘formed on a vague, individual conception of the spirit of the compact, which is not the result of interpreting any specific language to be quoted, nor referable to any recognised principle of the common law’.[147] Yet neither the specific language nor the common law resolves the issue. Governments are not ordinary persons, in either the natural or legal sense. They are unique, as is reflected in the traditional canon of statutory construction that Acts are presumed not to bind the Crown. A natural legal reading of any Commonwealth power, therefore, cannot simply be presumed to extend to the States. Further, there was no clear legal position on this issue to be found in the common law or elsewhere in 1899–1900. The fact that the British Parliament could bind the British Crown provides no relevant answer. There is a fundamental difference between Parliament’s binding its own executive in a unitary system and one Parliament binding a government of another jurisdiction within a federation.[148] As for the American case law, this was neither consistent nor clear, and involved implied immunities in any case.[149]

There is no neutral position. Either the different governments can bind each other — absolutely or with some limitations — or they cannot. In the absence of any relevant express position on these issues, the matter must be resolved by reference to implications. In doing so one must refer to some model of, or doctrine on, how different levels of government can and should interact in a federation. That doctrine will be significantly determined by the details of the Australian federation established by the Constitution, but those details alone do not supply the answers needed here; some external concept is required. To the extent that such a doctrine is given constitutional effect, it is impliedly incorporated. The Court indicated in the Engineers’ Case that the Commonwealth generally could bind the States. The later recognition of certain immunities qualified this principle. Whatever the position, these decisions necessarily rely in part on external notions. Thus the decision in the Engineers’ Case itself, that archetype of literalism, depends on an impliedly incorporated doctrine.

It is precisely because the Constitution provides no express answers to these issues that the Court has had such difficulty in establishing a consistent position over the last century. With incorporated doctrines, it is frequently difficult to be certain which version of the doctrine is entrenched and to what extent. That external notions of federalism are doing the relevant work is emphasised by the fact that the majority of judgments on the implied immunities have invoked no specific text. Judges have contented themselves with references to the purported requirements of the Constitution’s ‘frame’ or its federal ‘system’, ‘conception’, ‘nature’, ‘compact’ or ‘structure’.[150] The history of disagreements on the issue indicates that these requirements are far from self-evident.

How then, within the terms of the second classification system, is this federalism doctrine impliedly incorporated? There have been some recent attempts, led by Brennan J in QEC v Commonwealth, to tie the implied State immunities to s 106 of the Constitution.[151] This is similar to the attempt to link the implied nationhood power to s 61. Section 106, which indicates that State constitutions shall ‘continue as at the establishment of the Commonwealth’, does not provide any significant assistance. First, given that it is expressed to be ‘subject to this Constitution’, it is far from clear how it can support implied limitations on the expressly granted Commonwealth powers. Secondly, one key element of the Engineers’ Case decision was a rejection of the view that s 107, a provision closely related to s 106, preserved particular State powers from Commonwealth interference.[152] Drawing the line between the preservation of State constitutional structures (per s 106) and State powers (per s 107) is not easy.[153] Finally, if the words of the section are crucial in this area, why did Brennan J himself not bother to quote them?

The key textual basis supporting the implications is the presupposition of numerous provisions[154] that the States will continue to exist and function as independent units of government. The implied federal immunities are, therefore, best regarded as emerging from the text in a multi-provisional manner. Given the presence of s 114 and the related provisions, an inductive argument could have been made, but none of the judges relied on such an approach.[155] That they ignored the expressio unius argument is a fault, but they did not seek to turn the difficulty into a purported virtue by claiming that the relevant provisions invoked a broader doctrine along the same lines.

As for an originalist classification, Goldsworthy unenthusiastically suggests that the immunities may be implicit assumptions.[156] Given that there was no clear pre-established legal position, it is impossible to say quite what the framers or Australian people of 1899–1900 might have taken for granted. Instead, the immunities fall into the fifth, internal-tension category. The Constitution creates an internal tension in arming the new government with powers which might substantially affect the States, yet also presupposing that the States would continue to exist and function (and, to a lesser extent, vice versa).

This internal tension points to why the two limbs of implied State immunity, at least, have become well accepted amongst judges since Melbourne Corporation v Commonwealth.[157] A tension exists, and whilst the appropriate reconciliation is not obvious, the need for some resolution is unavoidable. The presupposition of the States’ continuing existence gives rise to a strong positive imperative to maintain the institutional efficacy and functioning of the federal structure.[158] The relative lack of controversy is aided by the limited restrictive consequences of the immunities.

It is noteworthy, however, that relatively little justification has been provided for the discrimination limb (which is not to say that it cannot be justified). The general limb, protecting the States’ existence and functioning, flows directly from the positive imperative just mentioned. However, the law invalidated in QEC v Commonwealth[159] imposed a relatively minor procedural burden on one State that in no way threatened its existence. Judges have apparently taken the view that a Commonwealth legislative purpose of damaging or controlling the States can never be legitimate.[160] There has been a related suggestion that the States are obliged and entitled to engage with the general legal system on the same basis as others.[161] These propositions merely restate the conclusion, however; they do not support it. The Court could have opted to assess whether the effect of any discriminatory legislation was such as to threaten the States’ existence and functioning as governments. The relevant legislative purpose is, after all, tested substantially by reference to a law’s effect.[162] The lack of persuasive judicial argument for this implied immunity is worth bearing in mind when considering some of the vociferous criticisms by judges in the implied rights field.

V RECOGNISING CONSTITUTIONAL IMPLICATIONS

The analysis in Part IV illustrates that the process of ascertaining implications is a fundamental and frequent part of constitutional interpretation. One legacy of the Engineers’ Case has been a marked judicial reluctance to admit that this is the case. The clearest acceptance of the role of implications came only where it was hard to deny, with the federal immunities. Even then, some judges sought to sidestep the issue, as others did for the implied nationhood power.

By way of exception to this trend, Dixon J stated in Australian National Airways Pty Ltd v Commonwealth: ‘I do not see why we should be fearful about making implications.’[163] He is right. There are good reasons for approaching the task with caution but constitutional implications cannot be scorned. Implied meanings are, after all, only different from express meanings by a matter of degree, namely, the directness of the communication involved.

Legitimacy issues do arise whenever judicial choice is involved, and judicial choice is almost invariably inherent in recognising implications. A concern about these matters is reflected in the traditional restrictive ‘necessary implication’ test, although such a separate test for implications cannot ultimately be justified. Instead, the question of whether or not to recognise an implication depends on the general approach adopted to constitutional interpretation, combined with such other guides as a theory of stare decisis. Within an originalist approach, factors can be stated to guide the necessary judicial choices. An awareness of the different types of implication may also offer some insight and assistance.

It is in light of this complex picture that the controversial implied rights judgments relating to equality and democracy must be analysed. That is the task to be undertaken in the second of these two articles.

[*] BA, LLB (Hons) (ANU), BCL, D Phil (Oxon); Barrister and Solicitor of the Supreme Court of New South Wales. Both this article and one to follow in the next volume of this journal (Jeremy Kirk, ‘Constitutional Implications (II): Doctrines of Equality and Democracy’ (2001) 25 Melbourne University Law Review (forthcoming)) are drawn from Jeremy Kirk, ‘Implied Rights’ in Constitutional Adjudication by the High Court of Australia since 1983 (D Phil thesis, Oxford University, 1998). A copy of the thesis is held in the Bodleian Law Libary, Oxford. Thanks are due to my doctoral supervisors, Professor John Finnis and Sir Anthony Mason, for their comments on previous versions of these papers. The views expressed, and any errors, are my own.

[1] McGraw-Hinds (Aust) Pty Ltd v Smith [1979] HCA 19; (1979) 144 CLR 633, 668 (Murphy J).

[2] (1992) 174 CLR 455 (‘Leeth’).

[3] Ibid 485–6.

[4] Ibid 484–7.

[5] Ibid.

[6] (1992) 177 CLR 106 (‘Political Advertising Case’).

[7] [1992] HCA 46; (1992) 177 CLR 1 (‘Nationwide News’).

[8] Eg McGinty v Western Australia [1996] HCA 48; (1996) 186 CLR 140, 234, 251 (McHugh J), also 169 (Brennan CJ), 188 (Dawson J) (‘McGinty’).

[9] [1997] HCA 25; (1997) 189 CLR 520, 566–7 (‘Lange v ABC’).

[10] Jeremy Kirk, ‘Constitutional Implications (II): Doctrines of Equality and Democracy’ (2001) 25 Melbourne University Law Review (forthcoming).

[11] See Jeffrey Goldsworthy, ‘Implications in Language, Law and the Constitution’ in Geoffrey Lindell (ed), Future Directions in Australian Constitutional Law (1994) 150, 153–4.

[12] Acknowledged in Levy v Victoria [1997] HCA 31; (1997) 189 CLR 579, 607 (Dawson J).

[13] Glanville Williams, ‘Language and the Law — IV’ (1945) 61 Law Quarterly Review 384, 400.

[14] [1993] HCA 44; (1993) 177 CLR 541, 548 (emphasis added) (‘Cheatle’).

[15] Geoffrey Sawer, ‘Implication and the Constitution: Part I’ (1948) 4 Res Judicatae 15, 17.

[16] Ibid.

[17] Illustrated below in Part IV.

[18] Eg Amalgamated Society of Engineers v Adelaide Steamship Co [1920] HCA 54; (1920) 28 CLR 129, 155 (‘Engineers’ Case’); see also references gathered in Stephen Donaghue, ‘The Clamour of Silent Constitutional Principles’ [1996] FedLawRw 5; (1996) 24 Federal Law Review 133, fn 166.

[19] Wilkinson v Adam [1812] EngR 144; (1813) 1 Ves & B 422, 466; [1812] EngR 144; 35 ER 163, 180; approved in James v South Australia [1927] HCA 32; (1927) 40 CLR 1, 24 (Isaacs ACJ and Powers J).

[20] Melbourne Corporation v Commonwealth [1947] HCA 26; (1947) 74 CLR 31, 82. See similarly B (a minor) v DPP [2000] UKHL 13; [2000] 1 All ER 833, 839 (Lord Nicholls): ‘“Necessary implication” connotes an implication which is compellingly clear.’

[21] Political Advertising Case (1992) 177 CLR 106, 134. Note also, in a contractual context, Codelfa Construction Pty Ltd v State Rail Authority of NSW (1982) 149 CLR 337, 346 (Mason J) (‘Codelfa’).

[22] Political Advertising Case (1992) 177 CLR 106, 135.

[23] As to which, see Jeremy Kirk, ‘Constitutional Interpretation and a Theory of Evolutionary Originalism’ (1999) 27 Federal Law Review 323, 325–6, 342–3, 346–8, 362–3.

[24] [1993] HCA 44; (1993) 177 CLR 541.

[25] Note Justice M H McHugh, ‘The Judicial Method’ (1999) 73 Australian Law Journal 37, 48–9, and at 46, quoting Justice Harlan Stone, ‘The Common Law in the United States’ (1936) 50 Harvard Law Review 4, 20; see also Julius Stone, Precedent and Law: Dynamics of Common Law Growth (1985) ch 1.

[26] Kirk, ‘Constitutional Interpretation’, above n 23, 358–66 and generally.

[27] Cf Goldsworthy, ‘Implications’, above n 11, 162–3.

[28] Australian National Airways Pty Ltd v Commonwealth [1945] HCA 41; (1945) 71 CLR 29, 85. See also, in a statutory context, Gaudron, Gummow and Callinan JJ in Katsuno v The Queen [1999] HCA 50; (1999) 199 CLR 40, 56.

[29] Eg Political Advertising Case (1992) 177 CLR 106, 134 (Mason CJ); Nationwide News [1992] HCA 46; (1992) 177 CLR 1, 42 (Brennan J); Theophanous v Herald & Weekly Times Ltd [1994] HCA 46; (1994) 182 CLR 104, 197 (McHugh J) (‘Theophanous’); McGinty [1996] HCA 48; (1996) 186 CLR 140, 184 (Dawson J).

[30] Victoria v Commonwealth [1971] HCA 16; (1971) 122 CLR 353, 402 (‘Payroll Tax Case’).

[31] See Kirk, ‘Constitutional Interpretation’, above n 23, 324–31.

[32] Payroll Tax Case [1971] HCA 16; (1971) 122 CLR 353, 401.

[33] See The Moorcock [1889] UKLawRpPro 8; (1889) 14 PD 64 (UK CA), as explained by G H Treitel, The Law of Contract (9th ed, 1995) 194; see also Williams, above n 13, 401–4.

[34] Jeremy Kirk, ‘Rights, Review and Reasons for Restraint’ (2001) 23 Sydney Law Review (forthcoming).

[35] Ibid, see section entitled ‘The Democratic Objection’.

[36] Noted in Kable v DPP (NSW) [1996] HCA 24; (1996) 189 CLR 51, 66 (Brennan CJ dissenting); Building Construction Employees and Builders’ Labourers Federation of NSW v Minister for Industrial Relations (1986) 7 NSWLR 372, 405 (Kirby P) (‘BLF Case’); George Winterton, ‘Extra-Constitutional Notions in Australian Constitutional Law’ (1986) 16 Federal Law Review 223, 234; Geoffrey Lindell, ‘Recent Developments in the Judicial Interpretation of the Australian Constitution’ in Geoffrey Lindell (ed), Future Directions in Australian Constitutional Law (1994) 1, 38.

[37] Discussed in Kirk, ‘Constitutional Interpretation’, above n 23, 340–2, 346–8.

[38] Tom Campbell, ‘Democracy, Human Rights, and Positive Law’ [1994] SydLawRw 16; (1994) 16 Sydney Law Review 195, 205.

[39] Acknowledged in John Pfeiffer Pty Ltd v Rogerson [2000] HCA 36; (2000) 172 ALR 625, 643–4 (Gleeson CJ, Gaudron, McHugh, Gummow and Hayne JJ).

[40] Note discussion in Kirk, ‘Rights, Review and Reasons for Restraint’, above n 34, section entitled ‘Legitimacy, Values and Preferences’; also Kirk, ‘Constitutional Interpretation’, above n 23, 348–50.

[41] McGraw-Hinds (Aust) Pty Ltd v Smith [1979] HCA 19; (1979) 144 CLR 633, 670.

[42] BLF Case (1986) 7 NSWLR 372, 405 (Kirby P); see also Nationwide News [1992] HCA 46; (1992) 177 CLR 1, 44 (Brennan J); Winterton, above n 36, 234, 239.

[43] Leslie Zines, Constitutional Change in the Commonwealth (1991) 51; see also Theophanous [1994] HCA 46; (1994) 182 CLR 104, 194 (Dawson J); Campbell, above n 38, 209–10.

[44] Kirk, ‘Rights, Review and Reasons for Restraint’, above n 34, section entitled ‘Political Questions’.

[45] Zines, Constitutional Change in the Commonwealth, above n 43, 54.

[46] Eg T R S Allan, ‘The Common Law as Constitution: Fundamental Rights and First Principles’ in Cheryl Saunders (ed), Courts of Final Jurisdiction (1996) 146, 155, 162.

[47] Robert McCloskey, ‘Economic Due Process and the Supreme Court: An Exhumation and Reburial’ [1962] Supreme Court Review 34, 45–54; K D Ewing, ‘Social Rights and Constitutional Law’ [1999] Public Law 104, 113–16, 119–23.

[48] See further Kirk, ‘Rights, Review and Reasons for Restraint’, above n 34, section entitled ‘Capability’.

[49] Kable v DPP (NSW) [1996] HCA 24; (1996) 189 CLR 51, 66 (Brennan CJ dissenting).

[50] Acknowledged in a related context by Deane J: Theophanous [1994] HCA 46; (1994) 182 CLR 104, 185.

[51] Kirk, ‘Rights, Review and Reasons for Restraint’, above n 34, section entitled ‘Public Confidence’.

[52] See further Richard Posner, Overcoming Law (1995) 186–7.

[53] Kirk, ‘Constitutional Interpretation’, above n 23, 324–5.

[54] Ibid 360.

[55] Re Residential Tenancies Tribunal (NSW); Ex parte Defence Housing Authority (1997) 190 CLR 410, 504 (Kirby J dissenting) (‘Defence Housing Authority’). A similar concern has been manifest regarding implied contractual terms: eg Ansett Transport Industries (Operations) Pty Ltd v Commonwealth [1977] HCA 71; (1977) 139 CLR 54, 62 (Gibbs J); Codelfa (1982) 149 CLR 337, 347 (Mason J), regarding the fourth of five conditions necessary to ground the implication of a contractual term (‘it must be capable of clear expression’); D W Grieg and J L R Davis, The Law of Contract (1987) 557–8.

[56] Payroll Tax Case [1971] HCA 16; (1971) 122 CLR 353, 411 (Walsh J), 424 (Gibbs J); Re Australian Education Union; Ex parte Victoria (1995) 184 CLR 188, 228 (Mason CJ, Brennan, Deane, Toohey, Gaudron and McHugh JJ) (‘Re AEU’); see further below Part IV(C).

[57] See Kirk, ‘Constitutional Interpretation’, above n 23, fn 30; Eastman v The Queen [2000] HCA 29; (2000) 172 ALR 39, 50 (McHugh J).

[58] This is a legitimate consideration in constitutional interpretation which overlaps with, but does not itself constitute, a ‘political questions’ or justiciability doctrine: see Geoffrey Lindell, ‘The Justiciability of Political Questions: Recent Developments’ in H P Lee and George Winterton (eds), Australian Constitutional Perspectives (1992) 180, 206–15.

[59] Discussed in Theophanous [1994] HCA 46; (1994) 182 CLR 104, 197–8 (McHugh J). See further below

Part IV(C).

[60] Discussed, and rejected, in Goldsworthy, ‘Implications’, above n 11, 168–70.

[61] Treitel, above n 33, 185–6; note Codelfa (1982) 149 CLR 337, 346–7 (Mason J).

[62] Treitel, above n 33, 186.

[63] Goldsworthy, ‘Implications’, above n 11, 152, 154–61.

[64] Ibid 154.

[65] Discussed in Jeffrey Goldsworthy, ‘Originalism in Constitutional Interpretation’ [1997] FedLawRw 1; (1997) 25 Federal Law Review 1, 20, 30–1.

[66] Australian National Airways Pty Ltd v Commonwealth [1945] HCA 41; (1945) 71 CLR 29, 81.

[67] Bank of NSW v Commonwealth [1948] HCA 7; (1948) 76 CLR 1, 332 (‘Bank Nationalisation Case’).

[68] Commonwealth v Tasmania [1983] HCA 21; (1983) 158 CLR 1, 126–7 (‘Tasmanian Dams Case’).

[69] Donaghue, above n 18, 170. See also at 165.

[70] [1960] HCA 92; (1960) 105 CLR 214, 248–9 (Dixon CJ, McTiernan, Taylor and Windeyer JJ).

[71] Donaghue, above n 18, 170.

[72] Goldsworthy, ‘Implications’, above n 11, 164, 172; see also Sawer, above n 15, 88.

[73] Goldsworthy, ‘Implications’, above n 11, 164–5, 174.

[74] Tasmania v Commonwealth [1904] HCA 11; (1904) 1 CLR 329, 348 (Barton J); see also Baxter v Commissioners of Taxation (NSW) [1907] HCA 76; (1907) 4 CLR 1087, 1105 (Griffith CJ, Barton and O’Connor JJ); McGraw-Hinds (Aust) Pty Ltd v Smith [1979] HCA 19; (1979) 144 CLR 633, 668 (Murphy J).

[75] Official Record of the Debates of the Australasian Federal Convention, Melbourne 1898,

7–8 February 1898, 654–64; Official Record of the Debates of the Australasian Federal Convention, Melbourne 1898, 2 March 1898, 1769–79.

[76] Goldsworthy, ‘Implications’, above n 11, 172.

[77] F A R Bennion, Statutory Intepretation: A Code (3rd ed, 1997) 384–6.

[78] Western Australia v Commonwealth [1975] HCA 46; (1975) 134 CLR 201; Queensland v Commonwealth [1977] HCA 60; (1977) 139 CLR 585.

[79] See below Part IV(C).

[80] Note Re Jarman; Ex parte Cook (1997) 188 CLR 595, 604 (Brennan CJ), 617–18 (Toohey and Gaudron JJ dissenting), 645 (Kirby J).

[81] (1999) 198 CLR 511, 555–63 (McHugh J), 574–82 (Gummow and Hayne JJ, with Gleeson CJ agreeing at 540 and Gaudron J agreeing at 546), 625–6 (Callinan J), cf 605–11 (Kirby J).

[82] (1992) 174 CLR 455. See further Kirk, ‘Constitutional Implications (II)’, above n 10,

Part III(A).

[83] Laurence Claus, ‘Implication and the Concept of a Constitution’ (1995) 69 Australian Law Journal 887, 888, 904; Donaghue, above n 18, 160–1.

[84] Eg Victorian Stevedoring & General Contracting Co Pty Ltd v Dignan [1931] HCA 34; (1931) 46 CLR 73,

101–2 (Dixon J), 114 (Evatt J); R v Kirby; Ex parte Boilermakers’ Society of Australia [1956] HCA 10; (1956) 94 CLR 254, 275–6 (Dixon CJ, McTiernan, Fullagar and Kitto JJ) (‘Boilermakers’ Case’); Defence Housing Authority (1997) 190 CLR 410, 441 (Dawson, Toohey and Gaudron JJ); Lange v ABC [1997] HCA 25; (1997) 189 CLR 520, 558–9, 561.

[85] G J Lindell, ‘Book Review: George Winterton, Parliament, the Executive and the Governor-General’ [1983] UNSWLawJl 22; (1983) 6 University of New South Wales Law Journal 261, 267.

[86] See Jeremy Kirk, ‘Implied Rights’ in Constitutional Adjudication by the High Court of Australia since 1983 (D Phil thesis, Oxford University, 1998) ch 6.1.1.

[87] In relation to the rule of law, note Theophanous [1994] HCA 46; (1994) 182 CLR 104, 196 (McHugh J), citing Australian Communist Party v Commonwealth [1951] HCA 5; (1951) 83 CLR 1, 193 (Dixon J) (‘Communist Party Case’); also Defence Housing Authority (1997) 190 CLR 410, 443–4 (Dawson, Toohey and Gaudron JJ), 507 (Kirby J); Kartinyeri v Commonwealth [1998] HCA 22; (1998) 195 CLR 337, 381 (Gummow and Hayne JJ); Abebe v Commonwealth [1999] HCA 14; (1999) 197 CLR 510, 560–1 (Gummow and Hayne JJ), 586 (Kirby J); Cheng v The Queen [2000] HCA 53; (2000) 175 ALR 338, 356–7 (Gaudron J).

[88] Eg R v Coldham; Ex parte Australian Social Welfare Union [1983] HCA 19; (1983) 153 CLR 297, 313; Re State Public Services Federation; Ex parte A-G (WA) [1993] HCA 30; (1993) 178 CLR 249, 272 (Mason CJ, Deane and Gaudron JJ), 275 (Brennan J). Note also, in relation to the different question of the interrelationship between the States within a federal structure, Lipohar v The Queen [1999] HCA 65; (1999) 168 ALR 8, 26, 35 (Gaudron, Gummow and Hayne JJ); John Pfeiffer Pty Ltd v Rogerson [2000] HCA 36; (2000) 172 ALR 625, 643–4 (Gleeson CJ, Gaudron, McHugh, Gummow and Hayne JJ).

[89] Eg Theophanous [1994] HCA 46; (1994) 182 CLR 104, 121 (Mason CJ, Toohey and Gaudron JJ), 149 (Brennan J).

[90] Theophanous [1994] HCA 46; (1994) 182 CLR 104, 149–50 (Brennan J); McGinty [1996] HCA 48; (1996) 186 CLR 140, 168–9 (Brennan CJ), 231 (McHugh J); Kruger v Commonwealth [1997] HCA 27; (1997) 190 CLR 1, 152 (Gummow J); Defence Housing Authority (1997) 190 CLR 410, 503 (Kirby J dissenting).

[91] (1992) 177 CLR 106, 135.

[92] Eg Boilermakers’ Case [1956] HCA 10; (1956) 94 CLR 254, 275 (Dixon CJ, McTiernan, Fullagar and Kitto JJ); A-G (Cth) v The Queen [1957] HCA 12; (1957) 95 CLR 529, 537–8 (PC).

[93] See above Part II(F). Some judges have been careful to indicate that this is what they mean by ‘structural’ implications: eg Payroll Tax Case [1971] HCA 16; (1971) 122 CLR 353, 405 (Walsh J); Kruger v Commonwealth [1997] HCA 27; (1997) 190 CLR 1, 155 (Gummow J).

[94] (1990) 169 CLR 482 (‘Incorporation Case’).

[95] Kingston v Keprose Pty Ltd (1987) 11 NSWLR 404, 422 (McHugh JA).

[96] Eg Ha v New South Wales [1997] HCA 34; (1997) 189 CLR 465, 491–8 (Brennan CJ, McHugh, Gummow and Kirby JJ), 506–11 (Dawson, Toohey and Gaudron JJ dissenting).

[97] Matthews v Chicory Marketing Board (Vic) [1938] HCA 38; (1938) 60 CLR 263, 293 (Dixon J).

[98] See above Part III(B)(3).

[99] Cf Claus, who appears to equate notions of inductive reasoning and incorporated doctrines: above n 83.

[100] [1997] HCA 25; (1997) 189 CLR 520, 566–7.

[101] Regarding the first two doctrines, see further below Part IV.

[102] Treitel, above n 33, 175–6, 197–201.

[103] Queensland Electricity Commission v Commonwealth [1985] HCA 56; (1985) 159 CLR 192, 231 (‘QEC v Commonwealth’); see also Nationwide News [1992] HCA 46; (1992) 177 CLR 1, 44; Theophanous [1994] HCA 46; (1994) 182 CLR 104, 143–4.

[104] Political Advertising Case (1992) 177 CLR 106, 181; see also McGinty [1996] HCA 48; (1996) 186 CLR 140, 184.

[105] Discussed in, eg, McHugh, above n 25, 42, 44–5.

[106] Theophanous [1994] HCA 46; (1994) 182 CLR 104, 198; McGinty [1996] HCA 48; (1996) 186 CLR 140, 231–2.

[107] McGinty [1996] HCA 48; (1996) 186 CLR 140, 231–2.

[108] Lindell, ‘Book Review’, above n 85, 267; Winterton, above n 36, 224, 226, 234; Goldsworthy, ‘Implications’, above n 11, 181; Claus, above n 83, 888–90.

[109] [1993] HCA 44; (1993) 177 CLR 541.

[110] See above Part III(B)(4).

[111] West v Commissioner of Taxation (NSW) [1937] HCA 26; (1937) 56 CLR 657, 681.

[112] (1996) 187 CLR 416, 483 (Brennan CJ, Toohey, Gaudron, McHugh and Gummow JJ) (‘Industrial Relations Act Case’).

[113] A-G ex rel Hutchings v Great Eastern Railway Co [1880] UKHL 2; (1880) 5 App Cas 473, 478–9 (Lord Selborne LC), 481–2 (Lord Blackburn), 486–7 (Lord Watson); Small v Smith (1884) 10 App Cas 119, 129, 133–5 (Earl of Selborne LC), 139–40 (Lord Watson).

[114] Eg Gary Rumble, ‘Section 51(xxxix) of the Constitution and the Federal Distribution of Powers’ (1982) 13 Federal Law Review 182; R D Lumb and G A Moens, The Constitution of the Commonwealth of Australia: Annotated (5th ed, 1995) 288, 293; Leslie Zines, The High Court and the Constitution (4th ed, 1997) 37–8.

[115] Le Mesurier v Connor [1929] HCA 41; (1929) 42 CLR 481, 497 (Knox CJ, Rich and Dixon JJ).

[116] G G Crespin & Son v Colac Co-Operative Farmers Ltd [1916] HCA 13; (1916) 21 CLR 205, 212 (Griffith CJ), 214 (Barton J), 219 (Isaacs J), 222–3 (Higgins J), 224 (Rich J); see also Moore v Commonwealth [1951] HCA 10; (1951) 82 CLR 547.

[117] Prominent exceptions being Australian National Airways Pty Ltd v Commonwealth [1945] HCA 41; (1945) 71 CLR 29, 85 (Dixon J), on a quasi-incidental issue; Victoria v Commonwealth [1975] HCA 52; (1975) 134 CLR 338, 414 (Jacobs J) (‘AAP Case’); Sillery v The Queen [1981] HCA 34; (1981) 35 ALR 227, 234 (Murphy J).

[118] Another example is the Commonwealth’s power to create federal courts: Waterside Workers’ Federation of Australia v J W Alexander Ltd [1918] HCA 56; (1918) 25 CLR 434, 465 (Isaacs and Rich JJ); Re Governor, Goulburn Correctional Centre; Ex parte Eastman (1999) 165 ALR 171, 179–80 (Gaudron J), 186 (Gummow and Hayne JJ).

[119] Burns v Ransley [1949] HCA 45; (1949) 79 CLR 101; R v Sharkey [1949] HCA 46; (1949) 79 CLR 121; Communist Party Case [1951] HCA 5; (1951) 83 CLR 1.

[120] AAP Case [1975] HCA 52; (1975) 134 CLR 338, 397 (Mason J).

[121] Ibid 397–8 (Mason J); Davis v Commonwealth [1988] HCA 63; (1988) 166 CLR 79, 111 (Brennan J).

[122] Burns v Ransley [1949] HCA 45; (1949) 79 CLR 101, 116; R v Sharkey [1949] HCA 46; (1949) 79 CLR 121, 148–9; Communist Party Case [1951] HCA 5; (1951) 83 CLR 1, 187–8. See also Davis v Commonwealth [1988] HCA 63; (1988) 166 CLR 79, 117–19 (Toohey J); Re Wakim; Ex parte McNally (1999) 198 CLR 511, 581–2 (Gummow and Hayne JJ).

[123] Burns v Ransley [1949] HCA 45; (1949) 79 CLR 101, 109–10 (Latham CJ); R v Sharkey [1949] HCA 46; (1949) 79 CLR 121, 135 (Latham CJ), 157–8 (McTiernan J), 163 (Webb J); Communist Party Case [1951] HCA 5; (1951) 83 CLR 1, 156 (Latham CJ), 211–12 (McTiernan J), 230–2 (Williams J), 233–4 and 247 (Webb J), cf 266 (Fullagar J), 275 (Kitto J).

[124] AAP Case [1975] HCA 52; (1975) 134 CLR 338, 362 (Barwick CJ), 397–8 (Mason J), 411–12 (Jacobs J); Tasmanian Dams Case [1983] HCA 21; (1983) 158 CLR 1, 108–9 (Gibbs CJ); Davis v Commonwealth [1988] HCA 63; (1988) 166 CLR 79, 93–5 (Mason CJ, Deane and Gaudron JJ), 110–11 (Brennan J).

[125] Davis v Commonwealth [1988] HCA 63; (1988) 166 CLR 79, 93, 95 (Mason CJ, Deane and Gaudron JJ).

[126] Ibid 101–4.

[127] Ibid 117–19.

[128] Ibid 104.

[129] Ibid 103–4, 119.

[130] Ibid 110.

[131] See Tasmanian Dams Case [1983] HCA 21; (1983) 158 CLR 1, 106–7 (Gibbs CJ), 203–4 (Wilson J), 252–3 (Deane J), 322–3 (Dawson J); Re Wakim; Ex parte McNally (1999) 198 CLR 511, 581–2 (Gummow and Hayne JJ, Gleeson CJ and Gaudron J agreeing); R v Hughes (2000) 171 ALR 155, 188–9 (Kirby J).

[132] [1920] HCA 54; (1920) 28 CLR 129.

[133] [1947] HCA 26; (1947) 74 CLR 31.

[134] See, eg, Re AEU (1995) 184 CLR 188, 231 (Mason CJ, Brennan, Deane, Toohey, Gaudron and McHugh JJ).

[135] Defence Housing Authority (1997) 190 CLR 410, 424–8 (Brennan CJ), 439–44 (Dawson, Toohey and Gaudron JJ). Kirby J (dissenting) proposed an explicitly reciprocal immunity: at 507–9.

[136] [1985] HCA 56; (1985) 159 CLR 192.

[137] [1947] HCA 26; (1947) 74 CLR 31.

[138] Ibid 50.

[139] Ibid 61–2; see also West v Commissioner of Taxation (NSW) [1937] HCA 26; (1937) 56 CLR 657, 668–9.

[140] Melbourne Corporation v Commonwealth [1947] HCA 26; (1947) 74 CLR 31, 97–100.

[141] [1971] HCA 16; (1971) 122 CLR 353, 372–3, 405. See also at 385–6 (McTiernan J).

[142] R v Coldham; Ex parte Australian Social Welfare Union [1983] HCA 19; (1983) 153 CLR 297, 313; Re AEU (1995) 184 CLR 188, 227 (Mason CJ, Brennan, Deane, Toohey, Gaudron and McHugh JJ).

[143] Melbourne Corporation v Commonwealth [1947] HCA 26; (1947) 74 CLR 31, 99.

[144] (1995) 184 CLR 188, 232–3 (Mason CJ, Brennan, Deane, Toohey, Gaudron and McHugh JJ).

[145] Melbourne Corporation v Commonwealth [1947] HCA 26; (1947) 74 CLR 31, 61 (Latham CJ); Payroll Tax Case [1971] HCA 16; (1971) 122 CLR 353, 369 (Barwick CJ); Sawer, above n 15, 18–19.

[146] Dixon J did discuss s 114 in Essendon Corporation v Criterion Theatres Ltd [1947] HCA 15; (1947) 74 CLR 1, 17–18, but not directly on point.

[147] [1920] HCA 54; (1920) 28 CLR 129, 145.

[148] Defence Housing Authority (1997) 190 CLR 410, 457 (McHugh J), cf 443–4 (Dawson, Toohey and Gaudron JJ).

[149] See Claus, above n 83, 890–4.

[150] Eg Melbourne Corporation v Commonwealth [1947] HCA 26; (1947) 74 CLR 31, 70, 74 (Starke J), 81, 83 (Dixon J), 99 (Williams J); Payroll Tax Case [1971] HCA 16; (1971) 122 CLR 353, 386 (Menzies J), 402 (Windeyer J), 405 (Walsh J), 417–19 (Gibbs J); QEC v Commonwealth [1985] HCA 56; (1985) 159 CLR 192, 205 (Gibbs CJ), 212, 214, 218 (Mason J), 260 (Dawson J).

[151] [1985] HCA 56; (1985) 159 CLR 192, 231–6, see also 245 (Deane J); Political Advertising Case (1992) 177 CLR 106, 241–2 (McHugh J); Re AEU (1995) 184 CLR 188, 226–30 (Mason CJ, Brennan, Deane, Toohey, Gaudron and McHugh JJ).

[152] Engineers’ Case [1920] HCA 54; (1920) 28 CLR 129, 154 (Knox CJ, Isaacs, Rich, Starke JJ).

[153] Zines, The High Court and the Constitution, above n 114, 338–40.

[154] In the Constitution ch I pt II alone, consider ss 7–15, 21.

[155] Though Geoffrey Sawer took the view that the implied immunities did depend on inductive argument: above n 15, 20.

[156] Goldsworthy, ‘Implications’, above n 11, 173.

[157] [1947] HCA 26; (1947) 74 CLR 31.

[158] Eg ibid 83 (Dixon J).

[159] [1985] HCA 56; (1985) 159 CLR 192.

[160] Melbourne Corporation v Commonwealth [1947] HCA 26; (1947) 74 CLR 31, 81–5 (Dixon J); Re AEU (1995) 184 CLR 188, 239–40 (Mason CJ, Brennan, Deane, Toohey, Gaudron and McHugh JJ). See Zines, The High Court and the Constitution, above n 114, 321, 327–8.

[161] Melbourne Corporation v Commonwealth [1947] HCA 26; (1947) 74 CLR 31, 84 (Dixon J); QEC v Commonwealth [1985] HCA 56; (1985) 159 CLR 192, 215, 217 (Mason J), 258, 262 (Dawson J); for other purported justifications, see QEC v Commonwealth at 231–5 (Brennan J), 247–8 (Deane J).

[162] Re AEU (1995) 184 CLR 188, 239–40 (Mason CJ, Brennan, Deane, Toohey, Gaudron and McHugh JJ).

[163] [1945] HCA 41; (1945) 71 CLR 29, 85, quoted and approved in Lamshed v Lake [1958] HCA 14; (1958) 99 CLR 132, 144–5 (Dixon CJ; Williams, Webb, Kitto and Taylor JJ agreeing); see also Payroll Tax Case [1971] HCA 16; (1971) 122 CLR 353, 401 (Windeyer J).