Abstract
This article considers how section 116 of the Australian Constitution applies to executive power. Notwithstanding that the express terms of s 116 apply only to legislative power, we argue that s 116 should be interpreted to also limit the exercise of executive power. A literal interpretation would have the effect that the Commonwealth can circumvent a constitutional limitation on legislative power through the exercise of executive power and would undermine the fundamental constitutional principle of legislative supremacy. An implication that s 116 restricts executive power in the same fashion as legislative power enhances the coherence of the Constitution as creating an integral polity and eliminates disconnect between legislative power and executive power with respect to the same subject matter.
I Introduction
The proper scope of freedom of religion, and the place of religion in public life, remains a hotly contested issue in contemporary Australia. Understanding how the Constitution relates to religion, and the limits it places on governmental interaction with religion, is a perennially important question. Section 116 of the Constitution provides: The Commonwealth shall not make any law for establishing any religion, or for imposing any religious observance, or for prohibiting the free exercise of any religion, and no religious test shall be required as a qualification for any office or public trust under the Commonwealth.
The establishment clause, the prohibition on imposing religious observances and the free exercise clause apply to Commonwealth legislative action, prohibiting the making of laws which establish a religion, impose a religious observance or prohibit the free exercise of any religion. In their terms, they do not apply to executive power. At face value, therefore, these clauses do not operate to prevent Commonwealth executive action which establishes a religion, imposes a religious observance or prohibits the free exercise of any religion. We refer to this as the ‘literal interpretation’.
Why is the clause drafted in this way? Luke Beck has recently argued that: the precise language of section 116 was not the result of a careful drafting choice. Indeed, neither Clark nor Higgins were particularly thoughtful in their choice of language, and neither the 1891 Australasian Convention nor the 1897−8 Federal Convention carefully considered and debated the precise language of the provision. The language used in section 116 is rather haphazard.
1
It is, therefore, unlikely that the restriction of the first three clauses of s 116 to legislation was the result of careful forethought. With this in mind, we argue that s 116 must be interpreted so as to restrict the exercise of executive power which establishes a religion, imposes a religious observance or prohibits the exercise of any religion, as well as restricting the making of laws which do these things. This is because the literal interpretation would have the effect that the Commonwealth was not prohibited from establishing a religion or restricting the free exercise of religion through the exercise of its executive power. However, by virtue of s 116, the Commonwealth would be prohibited from legislating in relation to the exercise of that power, such as by imposing conditions on the exercise of executive power.
The literal interpretation introduces a disconnect between legislative power and executive power in relation to the same subject matter. This would undermine the constitutional principle of legislative supremacy, which is a fundamental principle underlying our Constitution. As such, the literal interpretation should be rejected. As we explain in this article, this issue is important because of the size and reach of the executive arm of government and Commonwealth executive spending. If the Commonwealth is operating beyond the powers granted to it under the Constitution, this has serious implications for many government departments and many citizens. As a matter of principle, this issue also raises important questions of constitutional interpretation.
We acknowledge that our proposal is a gloss on the words of s 116. However, this is not unprecedented. To support our argument, we discuss other contexts and cases in which legislative supremacy has been employed in constitutional interpretation, and decisions relating to s 51(xxxi) which the court has interpreted as imposing restrictions on executive power which are not found in the express words of the constitutional text. Furthermore, we examine recent scholarship and judgements which consider that the implied freedom of political communication also constrains executive power as a function of the Constitution limiting the exercise of legislative power which burdens political communication. Thus, as a matter of constitutional principle, legislative supremacy dictates that where s 116 of the Constitution limits legislative power in a specific situation, any corresponding exercise of executive power relating to the same situation must be similarly limited.
In Part II of the article, we clarify what we mean by the exercise of executive power in the context of freedom of religion and examine the current jurisprudential position regarding the relationship between s 116 and executive power. In Part III, we argue by analogy with the implied freedom of political communication that s 116 should constrain the exercise of executive power and identify the principle of legislative supremacy as the foundation for this argument on the basis that the Constitution creates an integral polity. Part IV provides a detailed explanation of the principle of legislative supremacy and implications for the relationship between the legislature and the executive, including that the Commonwealth cannot avoid limitations on legislative power through exercising executive power. Consequently, we argue that s 116 should be interpreted to limit the exercise of executive power in the same fashion that it limits legislative power. Such an implication would be consistent with the broad interpretation of other constitutional rights and would enhance the coherence of the constitutional instrument itself as creating an integral polity.
II Executive Power and Section 116
A Executive Power and Religion
Before examining the interpretation of s 116, it is necessary to consider what executive powers the Commonwealth possesses in relation to religion. Section 61 of the Constitution incorporates powers derivable from the royal prerogative, namely those prerogatives of the Crown that are applicable to Australia and appropriate to the spheres of responsibility vested in the Commonwealth by the Constitution. 2 Although the prerogative includes powers relating to religion, including supreme authority and jurisdiction as head of the Church of England, 3 power to convene ecclesiastical synods and the right of nomination to bishoprics and other ecclesiastical preferments, 4 as there is no established religion in Australia, those prerogative powers ‘do not exist and have never existed in respect of the Commonwealth’. 5
Recent High Court decisions have significantly curtailed the non-statutory executive power of the Commonwealth. In Pape v Federal Commissioner of Taxation (‘Pape’), the High Court held that Commonwealth power to spend money must be found in s 61 of the Constitution or a valid statute 6 and in Williams v Commonwealth (No 1) (‘Williams (No 1)’) the Court held that s 61 does not include a general power to contract and spend money in relation to matters falling within the heads of Commonwealth legislative power. 7
Given that the scope of non-statutory executive power is shrinking,
8
what scope remains for Commonwealth executive power to affect religion? French CJ noted in Williams (No 1) that ‘[t]here are undoubtedly significant fields of executive action which do not require express statutory authority’.
9
His Honour noted that the executive power referred to in s 61 extends to: • powers necessary or incidental to the execution and maintenance of a law of the Commonwealth; • powers conferred by statute; • powers defined by reference to such of the prerogatives of the Crown as are properly attributable to the Commonwealth; • powers defined by the capacities of the Commonwealth common to legal persons; • inherent authority derived from the character and status of the Commonwealth as the national government.
10
We consider that there are several categories of executive power which may be relevant to religion. These are outlined in turn.
1 Contracts for the Ordinary Services of Government
There is authority that Commonwealth executive power extends to entering into contracts for the ordinary functions of government. In New South Wales v Bardolph, the High Court held that, in the words of Rich J, ‘the Crown has a power independent of statute to make such contracts for the public service as are incidental to the ordinary and well-recognized functions of Government’. 11 The existence of this category has been criticised by Enid Campbell on the basis that it ‘has the effect of enlarging the area of executive authority by prescription’, given that the executive itself decides what the regular activities of government are. 12 Indeed, the precise extent of this power appears to be uncertain, with subtle differences existing among the authorities. Some judgements suggest that the power extends to administering the ordinary or recognised government activities, 13 while others suggest that it extends to ‘the administration of departments of State’. 14
While the outer limits of this power may be contestable, it remains true that the Commonwealth has non-statutory executive power under either or both of ss 61 and 64 to engage in activities relating to the administration of government departments. Under this power, the Commonwealth has power to do things such as enter into contracts, spend money, lease property and so on. It is conceivable that the Commonwealth could exercise this power in such a way as to conflict with the requirements of s 116. Saunders and Yam have argued that government contracting poses a significant challenge for the rule of law, given that ‘governments may impose conditions of a regulatory or public policy nature on private parties in a manner that simulates the effect of a legal rule’. 15 Given the large scale of Commonwealth departmental expenditure, 16 there is the potential for misuse of this category of power to have significant consequences. While it is unlikely that the Commonwealth could establish a religion using this category of power, it could impose a religious observance, or (a more likely possibility) infringe the free exercise of religion, for example, by imposing contractual conditions on contracting parties which have that effect.
2 Employment Practices
The power to appoint and dismiss public servants is a prerogative power which exists in Australia, although many aspects of this power are now regulated by statute. 17 Such a power is likely also to exist in relation to the power to administer a Commonwealth department. This power is relevant to religion because it conceivably could be used to restrict the ability of government employees to freely exercise their religion, analogous to the type of restrictions considered in Comcare v Banerji. 18
3 Intergovernmental Agreements
The Commonwealth has power under s 61 to enter into intergovernmental agreements with the States, which is likely to support ‘a range of agreements and understandings between the Commonwealth and State Executive Governments’. 19 While it is true that ‘the executive power could not be invoked to support an intergovernmental agreement under which the Commonwealth would legislate in contravention of a constitutional guarantee or prohibition’, 20 it is conceivable that the Commonwealth could enter into an agreement which induces one or more States to act in contravention of s 116. Intergovernmental agreements ‘affect many, and perhaps most, areas of governmental activity’ and, even where they are not legally binding, may nevertheless ‘be significant as a form of “soft law,” effectively guiding executive action, often complementing the operation of legislation and sometimes affecting the interests of third parties’. 21
4 General Laws Empowering Executive Action
A final category that is potentially relevant is general laws empowering executive action. That is, it is possible that general powers conferred by a statute, which do not specifically have religion in view, might be exercised in such a way as to infringe s 116. Barwick CJ noted that s 116 is ‘directed to the making of law. It is not dealing with the administration of a law’. 22 Although Barwick CJ emphasised that if a statute was administered in such a way as to amount to the establishment of religion, the statute would likely be void under s. 116, 23 this does not necessarily follow. As a result of the narrow purpose test employed by the High Court in religion to s 116, a law will likely only infringe s 116 if it has the purpose of establishing religion or prohibiting the free exercise of religion. 24 However, a power conferred by a statute which does not have the purpose of establishing religion or prohibiting the free exercise of religion may be exercised in such a way as to have that effect. The following section explains the current position on the extent to which s 116 limits executive power and provides some more specific examples of how the exercise of executive power may restrict freedom of religion.
B Current Position on Section 116 and Executive Power
In its terms, s 116 only prohibits the making of laws, which means it only directly restricts the exercise of legislative power. However, Aroney notes that executive power is ‘indirectly’ affected by virtue of the fact that s 116 can limit laws which confer functions, powers and jurisdictions on executive agencies.
25
This has been understood to mean that s 116 does not apply to executive power unless that power is authorised by statute.
26
For example, Barwick CJ in Attorney-General (Victoria) ex rel Black v Commonwealth (‘DOGS Case’) notes the following: The next observation I wish to make as to s. 116 is that it is directed to the making of law. It is not dealing with the administration of a law. But, if that administration is within the ambit of the authority conferred by the statute, and does amount to the establishment of a religion, the statute which supports it will most probably be a statute for establishing a religion and therefore void as offending s.116. That is so, not because of the manner of the administration but because the statute, properly construed, authorizes it… The manner of [a law’s] administration can have no independent effect. What may lawfully be done in its administration forms part of the consideration of the validity of its enactment: and what can be lawfully done is determined by the construction of the statute, the determination of its meaning and its operation.
27
Moens points out that although Barwick CJ was commenting in the specific context of the establishment clause, there is no reason why that analysis should not equally apply to the other clauses of s. 116. 28 This was also affirmed by Jackson J in the Full Federal Court, who, relying on the above passage by Barwick CJ, held that the application of s 116 to executive acts is on the basis that the relevant authorising legislation cannot permit an executive decision which amounts to a breach of s 116. 29 Justice Jackson further held that Barwick CJ’s remarks apply equally to all parts of s 116. 30 However, there is limited authority to suggest that s 116 imposes a limit on the exercise of non-statutory executive power.
Hobbs and Williams emphasise that, as s 116 applies only to legislative power, ‘the exercise of executive power poses a real risk for interference with religious freedom’.
31
Evans expands upon the potential risks: It is important to recall that executive power is not only statutory. The executive has a range of powers granted explicitly by the Constitution as well as prerogative or common law powers. These powers can be quite extensive (for example... they extended to forcibly preventing the entry of those aboard the MV Tampa into Australia) and include the powers of a legal person with respect to such matters as entering into a contract, property ownership and control, and employment. These are areas where there is real potential for interference with religious freedom.
32
The existing cases on s 116 also provide numerous examples of how the exercise of executive power may restrict freedom of religion, including the removal of the chosen leader of a religious community under migration powers, removal of members of a religious community from that community if the community is deemed unsafe, and preventing a religious community from assembling and propagating their religion in a state of emergency. In Minister for Immigration and Ethnic Affairs v Lebanese Moslem Association, the Minister deported an individual appointed to be an Imam of the Lebanese Moslem Association. The Full Federal Court reversed a Federal Court decision that the Minister was not permitted to determine the suitability of a religious leader against the decision of a religious association. Despite the Full Court’s factual finding that the exercise of ministerial power to remove an Imam was not ‘for’ prohibiting the free exercise of religion, they conceded it may still have that effect if this became a regular practice. 33 In Kruger v Commonwealth (‘Kruger’), the plaintiffs argued that a Northern Territory ordinance which authorised the forced removal of Indigenous children from their tribal culture and heritage was invalid as a law prohibiting the free exercise of religion. Toohey J and Gaudron J, in their respective minority judgements, conceded that ‘it may well be that an effect of the Ordinance was to impair, even prohibit the spiritual beliefs and practices of the Aboriginal people in the Northern Territory’. 34
In Jehovah’s Witnesses v Commonwealth (‘Jehovah’s Witnesses’), the National Security (Subversive Association) Regulations 1940 (Cth) prohibited the advocacy of doctrines which were considered to be prejudicial to the Commonwealth’s prosecution of the war. They provided for the dissolution of associations propagating such doctrines and vested their property in the Commonwealth. 35 The Jehovah’s Witnesses case challenged the constitutional validity of the Regulations on the basis that, among other arguments, they prohibited the free exercise of religion. The Court held that the Regulations and subsequent exercise of executive power giving effect to the Regulations did not breach s 116. For example, Williams J claimed that ‘a state of war, therefore, justifies legislation by the Commonwealth Parliament … which makes many inroads on personal freedom …’. 36 Babie argues this reasoning implies ‘that the conduct claimed to be infringed by the relevant legislation would almost certainly be encompassed by the ambit of the right if it extended to cover the effects of legislation’. 37 While the law evinces no express purpose to limit free exercise, ‘in its administrative application [it] makes possible the limitation of free exercise … based upon a finding of dangerous religious belief’. 38 Similarly, McTiernan J indicated that ‘it does not appear that the real object of the Regulations is to arm the Executive with power to prohibit or restrict the exercise of any religion’. 39 In referring to the ‘real object’ and the ‘Executive’, McTiernan J allows ‘the possibility that the stated purpose and the deeper objective of legislation [and executive action] may differ’. 40 Mortensen concludes that ‘the Court in Jehovah’s Witnesses assumed that a facially-neutral regulation directed at the suppression of subversive organizations, burdening religion in its effect, could offend the Free Exercise Clause’. 41 This could occur by virtue of the law itself or the administrative application of the law (ie its effect in terms of an exercise of executive power under statute). As such, it is possible that s 116 also extends to limiting laws which restrict freedom of religion in their effect, not merely in their terms or purpose.
Section 116 therefore limits the exercise of executive power under statute, potentially including the effects of the law and not merely its purpose. As concisely stated by McTiernan J, ‘the section creates a restriction both on legislative and executive power’. 42 The most explicit considerations of the application of s 116 to executive power have occurred in the context of executive powers granted under statute (ie statutory executive power). It appears to be reasonably settled that s 116 limits statutory executive power to the extent that it limits the operation of the law conferring the power. However, in the Jehovah’s Witnesses case, Latham CJ also acknowledged that the non-statutory executive power of the Commonwealth under s 61 extends to ‘obstruction to recruiting, certainly in war-time, and, in my opinion, also in time of peace. Such obstruction may be both punished and prevented’. Moreover, ‘propaganda tending to induce members of the armed forces to refuse duty may not only be subjected to control, but may be suppressed’. 43 This would in effect allow the suppression of doctrines which advocate for religiously based pacifism which could, for example, breach the free exercise clause. Evans has also indicated ways in which non-statutory executive power may burden freedom of religion, as discussed above. 44
In conclusion, we consider that there are significant areas of executive power which may impact religion in the ways specified in s 116. These impacts would not necessarily be prohibited by the current interpretation of s 116, which only clearly limits the exercise of statutory executive power which targets religion in explicit terms. In this article, we take this logic further by arguing that s 116 should also limit the exercise of pure executive power which is not granted by statute (ie non-statutory executive power). To do this, in the first instance we reason by analogy through examining emerging scholarship arguing that the implied freedom of political communication limits both statutory and non-statutory executive power. Since the implied freedom constrains government power using the same structure as s 116 (it is in the form of a limit on legislative and correlative powers rather than in the form of a personal right giving rise to individual action), the analogy is sound. 45
III How the Implied Freedom of Political Communication Limits Executive Power
The implied freedom of political communication has been derived by the High Court from the provisions of the Constitution which establish a system of representative government, especially ss 7 and 24 which provide that members of Parliament shall be ‘directly chosen by the people’. 46 This requirement assumes that people are provided the ability to make an informed choice, which entails the free flow of information and opinion pertaining to political matters. To enable the system of representative government to be effective, there must be freedom to communicate on matters of politics and government. 47 It has been held that freedom of political communication is therefore necessarily implied by the constitutional establishment of the system of representative government. Like s 116, the implied freedom is not a personal right actionable by individuals which may give rise to damages in the event of breach; it is a limit on the power of the Commonwealth, State and Territory governments. 48
Forrester, Zimmermann and Finlay emphasise that the implied freedom is a ‘restriction on legislative and executive powers’. 49 However, in Wotton v Queensland (‘Wotton’), the High Court held that the exercise of power by the executive under a statute impugned by the implied freedom of political communication is not a question of constitutional law. 50 As confirmed in Palmer v Western Australia in the context of s 92, the relevant constitutional question is the validity of the provisions of the authorising statute rather any particular exercise of power under that statute. 51 Forrester, Zimmermann and Finlay argue that such an approach to the implied freedom ‘does not properly account for its importance in Australia’s constitutional order. A more systematic approach is required, and one that accords significant weight to the implied freedom’. 52 They outline this approach in detail, arguing that since the implied freedom is an indispensable incident to representative and responsible government, and responsible government entails the exercise of executive power, ‘there is no reason in logic or principle why the implied freedom does not also limit the executive powers found in Chapter II’. 53 They observe that the source of Commonwealth executive power is found in the Constitution itself (s 61), and both the legislature and the executive are limited by the Constitution. As such, the implied freedom limits both the exercise of statutory executive power and non-statutory executive power. 54 The test is the same as that articulated in McCloy v New South Wales, which explicitly refers to ‘administrative acts’. 55
Forrester, Zimmermann and Finlay proceed to explain that the implied freedom still restricts executive power even though certain provisions of the Constitution (such as s 61) are not prefaced by the words ‘subject to this Constitution’. They note that the Constitution creates an integral polity where executive action may affect political communication, and therefore, the implied freedom should be applied such that ‘Australia’s constitutional arrangements create not only one system of jurisprudence but also one polity’.
56
Similarly, the implied freedom also applies to grants of legislative power which lack the prefaced words ‘subject to this Constitution’.
57
They conclude: Section 61 not only obliges the executive to execute laws. It is also a source of the Crown’s non-statutory executive powers, including the prerogative powers. For the reasons noted above, the implied freedom applies to the entirety of s 61. Further, and also for the reasons noted above, the implied freedom extends to State and Territory non-statutory executive power. This means, for example, certain exercises of the prerogative power may impermissibly infringe the implied freedom.
58
Carney also picks up the issue of the implied freedom and non-statutory executive power, arguing that the implied freedom ought to restrict ‘any exercise of non-statutory executive power whether an exercise of royal prerogative power or other non-statutory capacity of the executive’. 59 He notes explicit judicial recognition of the fact that the implied freedom would restrict executive power in the leading cases. 60 This includes non-statutory executive power because some aspects of non-statutory executive power have the capacity to detrimentally affect the legal rights and duties of citizens, and in this sense, the exercise of such power may undermine the principles of representative and responsible government. 61 For example, though non-statutory executive power cannot modify legislation or the common law, 62 it can affect the legal rights and duties of others by acts of coercion; the Royal Prerogative allows the Crown to decide and alter the terms of service of its servants and to destroy property in wartime. 63 The implied nationhood power and s 61 of the Constitution also allows the executive to prevent the entry of aliens into Australia and may extend to the expulsion of aliens present in Australia. 64 The executive can also act non-coercively by being able to do anything not prohibited by law, such as entering into contracts, starting a business or disseminating information. 65 The implied freedom might be restricted in particular through the imposition of gag orders on ministers, through decisions made by the executive regarding freedom of information or through confidentiality in contracts to which the executive is a party. 66 It is also important to emphasise that, at least in the context of executive power, the burden should be framed as a limit on the exercise of power or an immunity, rather than as an imposition on an individual right. 67 Carney’s analysis presciently points to the issues raised in the subsequently decided case of Comcare v Banerji. 68
The case involved Michelle Banerji, an employee of the Department of Immigration who posted messages critical of the Department on Twitter under a pseudonym. Her employment was consequently terminated for breaching s 13 of the Public Service Act 1999 (Cth) (‘APS Code of Conduct’), which provides that members of the Australian Public Service (‘APS’) must uphold the APS values, including that the APS is apolitical and performs its functions in an impartial manner. The primary constitutional issue was whether the relevant provisions of the Public Service Act infringed the implied freedom of political communication. The High Court noted that, in order for a law to impermissibly burden the implied freedom, the law must burden political communication as a whole, not merely an individual’s ability to engage in political communication. 69 The High Court unanimously held the provisions were valid and did not infringe the implied freedom.
With respect to executive power specifically, the High Court followed the standard approach in Wotton noted above: where a burden on political communication has its source in statute, the constitutional analysis operates at the level of statute rather than at the level of application of the power by the executive. For example, Edelman J noted in Comcare v Benerji that ‘[t]he constitutional constraint does not operate directly upon the exercise of executive power. It invalidates the executive act only by operating upon the legislation, disapplying the legislative authority for the executive act’. 70 It should be acknowledged that this approach of determining invalidity at the level of the statute rather than at the level of the exercise of power under the statute (which Boughey and Carter have labelled a ‘legislation-centric approach’) 71 may be thought to undermine our argument in this section. It implies it is difficult to extend s 116 invalidity to executive action when, under current authority, invalidity under the implied freedom of political communication operates at the level of statute rather than at the level of executive action. Underscoring the ‘legislation-centric approach’ objection is an even more significant difficulty. Where the explicit text of s 116 limits invalidity to the exercise of legislative power (and not executive power), the implied freedom is not so limited by the text.
This ‘legislation-centric approach’ approach entails separate dimensions of what seems to be the fundamental objection: the implied freedom is not really analogous to s 116 in the way we are claiming. However, the problem is not insurmountable. The textual differences between s 116 and the implied freedom do not detract from the similarities noted above, which are grounded in the fact that both are constitutional freedoms which protect fundamental rights, and both structurally function as protections in the same way (as a limit on the exercise of government power). The fact that both protections operate as limits on government power implies that they should limit both legislative and executive power in the same way, even if this has not been acknowledged by the High Court. Indeed, a recent article by Boughey and Carter powerfully critiques the High Court’s ‘legislation-centric approach’, arguing that it is ‘unclear’, ‘lacks a compelling justification’ and ‘is inconsistent with other constitutional principles and precedent’. 72 They contend that executive power generally is constrained by constitutional structure, which includes the implied freedom and that there is ‘no convincing reason’ to say that constitutional freedoms limit legislative power alone. 73 We, in effect, propose a mutually reinforcing and complementary critique which defuses the ‘legislation-centric approach’ problem by arguing that the High Court has drawn implications beyond the text on the basis of fundamental constitutional principles, precedent and other reasons which we articulate in more detail below. 74
Thus, a ‘legislation-centric approach’ approach undermines constitutional limitations which operate to constrain government power (legislative and executive) as part of an integral instrument establishing a collective polity. It is inconsistent with the very nature of constitutional limitations to allow the executive to do what the Parliament cannot with respect to the same subject matter because it undermines the supremacy of the legislature. Since constitutional limitations such as the implied freedom and s 116 are part of this integral instrument and are related in their operation, the supremacy of the legislature is a relevant interpretive principle for both.
Indeed, as Forrester carefully argues, the free exercise clause of s 116 in particular operates in conjunction with the implied freedom of political communication to support religious perspectives influencing political discourse and government policy as part of ‘Australia’s constitutionally prescribed system of representative and responsible government’, including facilitating communication and association. 75 As such, views about religion may inform both the Parliament and the executive ‘when lawmaking and discharging executive functions’. 76 The normal operation of constitutional protections of rights and freedoms function as a denial of legislative and executive power, and since the implied freedom of political communication and s 116 are both constitutional protections of rights and freedoms which function as limits on legislative power, they should also function as limits on executive power. 77
Thus, we suggest the reasoning undergirding the application of the implied freedom to executive power is based on the principle of legislative supremacy, and we argue by analogy that this principle also applies to the interpretation of s 116. The exercise of both statutory and non-statutory executive power has the capacity to undermine the constitutional bedrock of representative and responsible government, by burdening the legal rights and duties of citizens. Similarly, the exercise of executive power has the capacity to undermine the freedom of religion in ways contemplated by s 116 with respect to the legislature. It does not make sense that the legislature be restricted in a way the executive is not. Furthermore, the Australian Constitution establishes an integral polity where legislative and executive powers are limited with respect to the same subject matter. To limit legislative power without correspondingly limiting executive power with respect to s 116 would be deeply incoherent. As Boughey and Carter put it, ‘if Parliament cannot legislate in a manner which impermissibly infringes a constitutional freedom, it would seem axiomatic that the executive should not be able to make a decision which has this effect. To do so would undermine the protections offered by the Constitution’. 78
IV Interpreting Section 116
A Legislative Supremacy and the Executive
This part of the article explains legislative supremacy as a principle in the Australian constitutional system in more detail, including how it is assumed by the Constitution and what it means for the relationship between the legislature and the executive. Parliamentary supremacy over the executive can be understood in a political sense in which Parliament is supreme over the executive. That is, under the conventions of responsible government which are a fundamental aspect of our constitutional system, the executive is a creature of Parliament and is accountable to Parliament for the exercise of its powers. 79 This relationship of supremacy and accountability is reflected in provisions of the Constitution such as s 64, which requires Commonwealth ministers to be members of Parliament, and s 83, which requires expenditure of Commonwealth funds to be appropriated by law. In theory, this principle has the consequence that ‘[i]n the long run the Parliament, comprising the House of Representatives and the Senate, is in a position to control the Executive Government’. 80
Parliamentary supremacy over the executive can also be understood in another sense, namely the supremacy of legislative power over the executive and in relation to executive power. This derives from the English principle of parliamentary sovereignty, which, as classically explained by A V Dicey, means that Parliament has ‘the right to make or unmake any law whatever’, and secondly, ‘that no person or body is recognised by the law of England as having a right to override or set aside the legislation of Parliament’. 81
The incorporation of executive powers into the Constitution has led some to argue that the powers specifically conferred by the Constitution are entrenched and cannot be removed by legislation. Further, s 61 might be thought to constitutionalise the powers of the executive and therefore place it beyond the reach of the legislature. Enid Campbell argued that powers such as those contained in s 72 of the Constitution could not be abolished nor amended by legislation. 82 JE Richardson argued that many of the High Court’s pronouncements relating to the subordination of executive to legislative power concerned the prerogative, rather than the specific powers conferred on the Governor-General or executive power ‘in its extension to the execution and maintenance of the laws of the Commonwealth and the Constitution’. 83 Thus, while the principle of legislative supremacy over the executive clearly permits the Commonwealth Parliament to control the exercise of the prerogative component of the executive power contained in s 61, ‘the executive power is partly inalienable and not within reach of the federal Parliament’. 84 That is, because s 61 expressly states that the executive power ‘extends to the execution and maintenance of this Constitution’, and of the laws of the Commonwealth, then it is beyond the power of the Parliament to remove those powers. Section 61 ‘cannot become a complete sacrificial victim to the legislative powers’. 85
Whatever the validity of these arguments in 1977, they are now much more doubtful in light of the High Court decisions in Pape and Williams (No 1), and the High Court has clearly held that the principle of legislative supremacy applies to Commonwealth executive power. The High Court has held that the Constitution imports ‘the basal assumption of legislative predominance inherited from the United Kingdom’
86
and that ‘it is of the very nature of executive power in a system of responsible government that it is susceptible to control by the exercise of legislative power by Parliament’.
87
Jacobs J said in Victoria v Commonwealth and Hayden: The Parliament is sovereign over the Executive and whatever is within the competence of the Executive under s. 61, including or as well as the exercise of the prerogative within the area of the prerogative attached to the Government of Australia, may be the subject of legislation of the Australian Parliament.
88
The Court has also held that the Commonwealth Parliament has clear power to modify or abrogate the prerogative powers which have been incorporated into s 61. 89 In cases such as Williams (No 1), the principle of parliamentary supremacy over the executive was employed by the Court in interpreting the Constitution to significantly limit Commonwealth executive power in ways not immediately apparent from the text of the Constitution. 90
As noted by Aroney, parliamentary supremacy is ‘[p]erhaps the most fundamental principle of them all’ but ‘is implied rather than explicitly stated in the Constitution’. 91 This illustrates several points of relevance for this article. First, it highlights the limitations of the text in understanding the Australian constitutional system. Although the High Court has emphasised that its fundamental role in constitutional interpretation is to expound the text of the Constitution, ‘upholding it throughout precisely as framed’, 92 the meaning of the Constitution cannot be gleaned solely from its text. 93 The Constitution assumes various constitutional and common law doctrines which inform its meaning 94 and the High Court has often employed historical materials when interpreting the Constitution. 95 Secondly, it illustrates the point that the High Court is not afraid to draw judicially enforceable implications from the principle of parliamentary supremacy over the executive.
Stellios and Zines referred to the ‘accepted view that executive power extends only to matters within Commonwealth legislative competence’. 96 As already noted, Williams (No 1) held that s 61 of the Constitution does not include a general power to contract and spend money in relation to matters falling within the heads of Commonwealth legislative power. 97 It does not seem to challenge the principle that ‘executive power extends only to matters within Commonwealth legislative competence’. If this is the case, then it suggests that any limitations on legislative power contained within the Constitution correspondingly limit the scope of executive power.
Section 51(xxxix) of the Constitution confers power on the Commonwealth Parliament to legislate in relation to ‘matters incidental to the execution of any power vested by this Constitution … in the Government of the Commonwealth … or in any department or officer of the Commonwealth’. This confers power on the Parliament to legislate in relation to executive power, although it is probably not a plenary power to do so. 98 In summary, as Harrison Moore pithily put it, ‘we are not encouraged to believe that the executive can make good an independent sphere of its own, free from legislative interference and control’. 99 This principle has been ‘constitutionalised’ by cases such as Pape and Williams (No 1).
B Preventing the Commonwealth from Sidestepping a Constitutional Prohibition
In several cases, the High Court has interpreted various provisions of the Constitution in such a way as to prevent the Commonwealth from circumventing constitutional prohibitions or restrictions. These cases are relevant to the subject of this article and shed light on how the High Court deals with such cases.
Section 96 empowers the Commonwealth to ‘grant financial assistance to any State on such terms and conditions as the Parliament thinks fit’. The High Court has held that this enables the Commonwealth to undertake activities beyond the heads of legislative power in s 51 and may impose conditions to induce a State to do something which the Commonwealth does not have power to do. 100 That is, the Commonwealth can use s 96 to do things indirectly via the intermediary of the States which it could not do directly. 101 However, the Commonwealth cannot use s 96 to sidestep a constitutional prohibition such as s 116. As put by Gibbs J, ‘ss 96 and 116 should be read together, the result being that the Commonwealth has power to grant financial assistance to any State on such terms and conditions as the Parliament thinks fit, provided that a law passed for that purpose does not contravene s 116’. 102
Section 51(xxxi) empowers the Commonwealth to legislate with respect to ‘the acquisition of property on just terms from any State or person for any purpose in respect of which the Parliament has power to make laws’. The provision has been described as a ‘a very great constitutional safeguard’ and so should ‘receive a liberal and wide construction’.
103
As such, the High Court has held that the Commonwealth cannot avoid the requirement to afford just terms by effecting acquisitions of property indirectly such as by means of an independent agency or a State government.
104
In P J Magennis Pty Ltd v Commonwealth, Latham CJ held: It is obvious that the constitutional provision could readily be evaded if it did not apply to acquisition by a corporation constituted by the Commonwealth or by an individual person authorised by a Commonwealth statute to acquire property. Further, the present case shows that the constitutional provision would be quite ineffective if by making an agreement with a State for the acquisition of property upon terms which were not just the Commonwealth Parliament could validly provide for the acquisition of property from any person to whom State legislation could be applied upon terms which paid no attention to justice.
105
Section 51(xxxi) requires just terms for legislation whose object is to acquire property, even if the Commonwealth is not the entity which directly acquires the property, and it is not possible for the Commonwealth to evade this requirement by effecting an acquisition indirectly. The guarantee of just terms in s 51(xxxi) is not to be avoided by ‘a circuitous device to acquire indirectly the substance of a proprietary interest’. 106 The Constitution does not permit the Parliament to achieve ‘by indirect or devious means’ what is expressly forbidden by s 51(xxxi). 107 Section 51(xxxi) therefore qualifies the Commonwealth’s power under s 96 to make grants to the States such that the Commonwealth has no power to grant financial assistance on the condition that a State acquires property on other than just terms. 108
Further, in Spencer v Commonwealth Perry J noted that s 51(xxxi) is relevant to the scope of Commonwealth executive power: With respect to Commonwealth executive power to enter a funding agreement, as s 96 is relevantly qualified by s 51(xxxi), an agreement to facilitate such a grant which could not be authorised by s 96, would equally not be supported by s 61 of the Constitution conferring executive power.
109
That is, although s 51(xxxi) is a head of legislative power and therefore not directly concerned with executive power, it has been interpreted by the courts so as to limit Commonwealth power to enter into agreements, limiting the scope of executive power in s 61. Notwithstanding the apparently clear text of s 51(xxxi), it has been interpreted to apply to executive power.
C Application to Section 116
This section considers how the principles in the previous sections apply to the interpretation of s 116. As noted, s 116 consists of four clauses. The establishment, religious observance and free exercise clauses prohibit the Commonwealth from making any law ‘for establishing any religion, or for imposing any religious observance, or for prohibiting the free exercise of any religion’. These clauses are directed to the making of laws which do those things, and so these clauses are a restriction on legislative power. 110 These clauses of s 116 do not deal with executive power or the administration of a law in their terms, although, as we saw Barwick CJ note, if the administration of a law ‘is within the ambit of the authority conferred by the statute, and does amount to the establishment of a religion, the statute which supports it will most probably be a statute for establishing a religion and therefore void as offending s 116’. 111 The weight of authority appears to be that the establishment, religious observance and free exercise clauses do not apply to executive power, although McTiernan J held in Jehovah’s Witnesses that s 116 ‘creates a restriction both on legislative and executive power’. 112 In addition, s 116 provides that ‘no religious test shall be required as a qualification for any office or public trust under the Commonwealth’. Thus, as Stephen J noted, s 116 ‘prohibits one avenue of encroachment open to legislature and executive alike – the imposition of religious tests for office holders’. 113
Thus, it must be acknowledged that the express terms of the establishment, religious observance and free exercise clauses apply only to legislative, and not executive action, although considerations relating to the administration of a law cannot be excluded. As Barwick CJ noted in the DOGS Case, in constitutional interpretation ‘the text of our own Constitution is always controlling’. 114 Nevertheless, it is legitimate to draw implications in some cases which are not expressly stated in the Constitution. 115 Although the High Court gives primacy to the text when interpreting the Constitution, Dixon J said in Australian National Airways Pty Ltd v Commonwealth that ‘[w]e should avoid pedantic and narrow constructions in dealing with an instrument of government and I do not see why we should be fearful about making implications’. 116 The High Court will draw implications ‘based on the actual terms of the Constitution, or on its structure’. 117 Dawson J wrote that ‘[t]he only implications which may be drawn from the express terms of the Constitution are those which are necessary or obvious’. 118
Should, then, such an implication be drawn extending the prohibition in the establishment, religious observance and free exercise clauses of s 116 to the exercise of executive as well as legislative power? We argue that there are four reasons why a constitutional implication should be drawn which extends s 116 to the exercise of executive power.
First, we argue that s 116 should be given a flexible operation consistent with its rights-protective focus. Stephen J wrote in the DOGS Case that s 116 ‘is a constitutional provision of high importance’. 119 In Church of the New Faith v Commissioner of Pay-roll Tax (Vic), Mason CJ and Brennan J wrote that ‘[f]reedom of religion, the paradigm freedom of conscience, is of the essence of a free society’. 120 However, s 116 has typically been interpreted narrowly by the High Court and has never resulted in the invalidation of a Commonwealth law. 121 Similar to s 51(xxxi), s 116 is ‘a very great constitutional safeguard’ and so should ‘receive a liberal and wide construction’. 122 As Dixon J wrote in relation to s 51(xxxi), ‘consistency with the principles upon which constitutional provisions are interpreted and applied demands that the paragraph should be given as full and flexible an operation as will cover the objects it was designed to effect’. 123 As noted above, s 51(xxxi), which is concerned with legislative power, has been interpreted so as to limit executive Commonwealth power. 124 There is therefore nothing unprecedented about extending the same principles of interpretation in relation to s 116.
Secondly, such an implication is necessary to prevent the Commonwealth from evading a constitutional guarantee. As Dixon J said in Bank of New South Wales v Commonwealth, ‘when a constitution undertakes to forbid or restrain some legislative course, there can be no prohibition to which it is more proper to apply the principle embodied in the maxim quando aliquid prohibetur, prohibetur et omne per quod devenitur ad illud [“when anything is prohibited, every means by which the thing may be accomplished is also prohibited”]’.
125
A similar principle was expressed by Mason CJ, Deane and Gaudron JJ in Georgiadis v Australian and Overseas Telecommunications Corporation: It is often said in relation to constitutional guarantees and prohibitions that ‘you cannot do indirectly what you are forbidden to do directly’. That maxim is, in fact, an important guide to construction, indicating that guarantees and prohibitions are concerned with substance not form.
126
Lord Atkin famously said that ‘[t]he Constitution is not to be mocked by substituting executive for legislative interference with freedom’,
127
which was specifically applied in relation to s 116 by McTiernan J in Jehovah’s Witnesses.
128
Prohibiting the Commonwealth from legislating so as to establish or prohibit the free exercise of religion but allowing the Commonwealth to do so by means of executive power would amount to mocking the Constitution ‘by substituting executive for legislative interference with freedom’. Gaudron J wrote in Kruger: Another matter which points in favour of construing s 116 as extending to laws which prevent the free exercise of religion, not merely those which, in terms, effect a prohibition in that regard, is the need to construe constitutional guarantees liberally, even limited guarantees of the kind effected by s 116. In this respect, it is inconsistent with established principles of constitutional construction to construe constitutional guarantees as concerned with form rather than substance. So too, it is inconsistent with established principle to interpret constitutional guarantees ‘pedantically’ so that they may be circumvented by legislative provisions which purport to do indirectly what cannot be done directly.
129
To adapt words of Aickin J, ‘[i]t would be a serious gap in the constitutional safeguard’ if the Commonwealth could effect by means of executive power that which is prohibited by legislation. 130 In the DOGS Case, the High Court held that s 96 was subject to s 116, such that the Commonwealth could not evade the constitutional prohibition in s 116 by means of grants made under s 96. 131
Thirdly, our proposed interpretation is necessary to ensure consistency with fundamental constitutional values. Parliamentary sovereignty over the executive is a fundamental constitutional principle which is integral to the Australian system of government, but which is not expressly stated in the Constitution. 132 As noted earlier, the High Court has employed the concept of parliamentary supremacy to draw implications which are not expressly found in the text of the Constitution, namely previously unforeseen limitations on executive power. 133 If the Commonwealth had power to do by means of executive action that which is denied to the legislature, this would mean that there would be an area of executive action not subject to regulation by Parliament. Not drawing the implication we propose would, to adapt the words of French CJ, Gummow, Hayne, Crennan, Kiefel and Bell JJ in Kirk v Industrial Court (NSW), create the potential for ‘islands of [executive] power immune from [legislative] supervision and restraint’. 134 The implication proposed in this article is therefore necessary to safeguard Parliament’s supremacy in relation to the executive. The limitation on legislative power contained in s 116 entails a corresponding limitation on executive power. 135
Finally, our proposed interpretation is necessary to avoid inconsistency in the Constitution. As noted, s 51(xxxix) confers power on the Commonwealth Parliament to legislate in relation to matters incidental to the execution of any power conferred on the Commonwealth government. Not drawing the implication we propose would create a clash between ss 51(xxxix) and 116. That is, where the Commonwealth executive engaged in activity which infringes the establishment, religious observance or free exercise clause of s 116, s 51(xxxix) would suggest that the Commonwealth Parliament possesses some power to legislate in relation to that exercise of power. Section 116 would, however, operate to deny the Parliament power to legislate to regulate aspects of that exercise of power. In order to ensure coherence in the Constitution, it is necessary to draw an implication extending the prohibitions on legislative power contained in s 116 to executive power.
The obvious hurdle to our argument is that, on its terms, s 116 applies only to Commonwealth laws and not executive power. The text of the Constitution therefore appears to clearly preclude our proposed interpretation. However, as noted above, the text of s 51(xxxi) is also clear, being a head of legislative power and not directly concerned with executive power. Notwithstanding this, the High Court has held that s 51(xxxi) should be interpreted so as to prohibit the exercise of executive power which attempts to evade the requirements of just terms. The considerations noted in this section suggest that it is consistent with constitutional principle for s 116 to be interpreted in an analogous fashion. Further, if it is true that ‘the precise language of section 116 was not the result of a careful drafting choice’, 136 then it could be said that the implication proposed in this article is simply making explicit what is implicit in the terms of s 116.
The argument outlined here is intended to apply specifically to s 116, and it is not suggested that it would have the consequence that, wherever the term ‘law’ appears in the Constitution, that should be interpreted to include executive action. This is because s 116 is a restriction on power and our argument, analogous with the case law on s 51(xxxi), is intended to prevent the Commonwealth doing through executive action what it cannot accomplish through legislation. Those considerations do not apply in relation to other provisions of the Constitution which refer to ‘law’ such as ss 61 and 109.
V Conclusion
In this article, we have argued that s 116 should not be interpreted in such a way as to establish a clash between fundamental constitutional values and a literal reading of the text. There has been limited judicial consideration of the application of s 116 to executive power. 137 As such, while this article proposes a novel interpretation of s 116, it cannot be said that our argument departs from a well-established line of High Court precedent. 138 A literal interpretation would allow the Commonwealth executive to do what the Commonwealth Parliament cannot do regarding the same subject matter, creating a disjunct in the constitutionally prescribed exercise of government power. To ensure the coherence of the constitutional instrument and consistency in constitutional interpretation, we therefore suggest an implication that s 116 limits executive power in the same way as it limits legislative power is appropriate. This will ensure the constantly expanding executive arm of government retains the oversight of the limits and balances provided by the Australian Constitution.
As noted above, on the current narrow interpretation of s 116 adopted by the High Court, a law will only infringe s 116 if it has the purpose of establishing religion or prohibiting the free exercise of religion. 139 Given the narrowness of this interpretation, it may be that extending the protection afforded by s 116 to executive power may achieve little in practice, applying only to executive action that has the purpose of inhibiting the free exercise of religion. A recent article by Saunders and Meagher argues that the purpose test applied in relation to s 116 needs to be reconsidered given that it is drawn from a rationale that relates to the establishment clause and does not adequately account for the ways in which law may interact with religion. 140 That article and the argument advanced in this article suggest that the time is ripe for reconsideration of the interpretation of s 116.
