Contracts and the Implied Freedom of Political Communication

Abstract

This article considers the phenomenon of contractually-imposed restraints on political communication. Such restraints often incidentally arise from broad limits on out-of-hours conduct imposed by employment contracts or from confidentiality or non-disparagement clauses in deeds of settlement. It is argued that the implied freedom of political communication has work to do in relation to at least some categories of such restraints. The various objections to that view are examined and it is argued those objections are not compelling. The article analyses the question of how the implied freedom would operate in respect of contracts that impermissibly burden freedom of political communication, and suggests that this may be achieved by developing the common law of contract to accomodate a doctrine similar to the doctrine governing unreasonable restraints of trade.

I Introduction

Contractual terms may significantly restrain political communication. For example, employment contracts commonly restrict employees’ ability to disseminate political opinions or information both at and outside work. Confidentiality or non-disparagement clauses in settlement agreements restrict communication about disputes even when they have political or governmental significance. This article argues such contractual terms burden the implied freedom of political communication. Where those burdens are unjustified, they infringe the freedom.

Lange v Australian Broadcasting Corporation (‘Lange’)¹ established that the common law must not infringe the freedom. Where it does, it must be developed to accommodate the freedom. This reasoning ought to apply to the common law of contract just as it applied to the common law of defamation in Lange. Contract law infringes the freedom. It must be developed by incorporating a requirement that contractual terms that unreasonably restrain political communication are unenforceable, just like the common law already refuses to enforce categories of contractual terms such as unreasonable restraints of trade. Such a development would ensure the enforceability of reasonable contractual restraints on communication is undisturbed, but also ensure contract law no longer facilitates the imposition of significant and unreasonable restrictions on Australians’ freedom to communicate about important political matters.

II How Contractual Terms May Burden the Freedom of Political Communication

The first step in determining whether a law infringes the constitutional freedom of political communication (‘the freedom’) is to determine whether the law ‘burdens’ the freedom.² A ‘burden’ on the freedom is any effective limit on someone’s practical capacity to communicate about political or governmental matters.³ One may easily imagine many examples of contractual terms that limit political communication. For simplicity, this article considers only two possible categories: restrictions on communication in employment contracts, and ‘confidentiality’ and ‘non-disparagement’ clauses in settlement deeds. Other examples (not here considered) might include lease agreements that might enable landlords to forbid tenants from displaying political signage and ‘commercial-in-confidence’ provisions in contracting agreements between the Crown and private contractors, or contractors and sub-contractors.⁴

A Employment Contracts

Employment contracts might burden the freedom in at least four ways.⁵ First, a term of the contract might expressly forbid some type (or all types) of political communication.

Second, the contract might impose some broad obligation, such as not to bring the employer into disrepute, or to keep certain matters confidential, which incidentally forbids some forms of political communication. For example, a 2014 Parliamentary Committee report confirmed that all staff of private contractors working at the immigration detention centre on Manus Island must sign ‘restrictive confidentiality agreements’.⁶ It seems likely that such agreements would have restricted some forms of political communication.

Third, the contract might incorporate documents such as workplace policies into the contract of employment.⁷ Such policies may contain similarly broad restrictions. In particular, ‘social media policies’, which constrain what employees may say even in their private lives on social media, are now commonplace in workplaces.⁸ For instance, in 2015, the social media policy for employees of a major contractor for immigration detention centres was leaked to the press and caused controversy for its draconian bans on expressions of opinions.⁹

Finally, employment contracts may give rise to implied terms of fidelity¹⁰ or confidence,¹¹ which terms may also incidentally restrain employees from certain political communications.¹²

If an employee engages in political communication which allegedly breaches the employment contract, there may be at least two consequences. First, the employee could be sued for breach of contract. The remedies sought might include damages and an injunction restraining further breaches. Second, and perhaps more realistically, the employer may rely on the breach to bring about the termination of the contract of employment.¹³ This will not be relevant for all categories of employees—casual employees, for example, may be lawfully dismissed by their employer without any breach of contract having occurred.¹⁴

In recent years, a number of examples of political communication allegedly breaching a contract of employment have arisen. On Anzac Day 2015, Scott McIntyre, an SBS reporter, tweeted outside of work hours an offensively expressed but political opinion that official war commemorations overlook the wrongdoings of Australian soldiers.¹⁵ SBS decided the tweet brought it into disrepute, thereby breaching a policy incorporated into Mr McIntyre’s employment contract, and consequently dismissed Mr McIntyre.¹⁶ Mr McIntyre’s dispute of the dismissal was settled out of court.¹⁷

More recently, a similar example emerged from the University of Sydney, where a lecturer was dismissed for expression of an allegedly offensive political opinion concerning his perception of similarities between the government of Israel and the German Nazi regime. The lecturer’s union unsuccessfully argued the dismissal breached a guarantee of academic freedom contained in the relevant enterprise agreement.¹⁸ In Rumble v HWL Ebsworth Lawyers (‘Rumble’),¹⁹ a law firm terminated an employee’s contract because he expressed political opinions concerning a government department that was a client of the firm. Because the employee was casual, the employer had the right to terminate the contract for any reason, and the question as to whether the employee had breached the contract did not arise. But if the employee had been permanent or on a fixed-term contract, the employer may well have had to allege a breach of contract.

In Comcare v Banerji (‘Banerji’),²⁰ a government department dismissed a public servant for authoring political tweets critical of government policy. The dismissal was justified under the Australian Public Service Code of Conduct, enshrined in the Public Service Act 1999 (Cth), which relevantly required the employee to ‘behave…in a way that upholds the APS Values and the integrity and good reputation of the APS’.²¹ A sui generis aspect of private sector employment is that most relevant aspects of the employment relationship usually governed by an employment contract are instead provided for by legislation and regulations (in addition to contract).²² The question of whether Banerji has any application to private law must therefore be approached with significant caution. But it seems arguable that were an analogous fact scenario to occur in a private law setting, the employer may have been able to rely upon any contractual obligations to uphold the employer’s ‘good reputation’ (or similar) to justify dismissal.

In James Cook University v Ridd,²³ a professor was dismissed because of his repeated vociferous and public disagreement with colleagues about the science relating to the deterioration of the Great Barrier Reef and climate change generally. Information about both the deterioration of the Reef and climate change (or the alleged failure of such events to transpire) are highly likely to affect voters’ formation of political opinions about environmental issues, and about whether, and if so how, the Commonwealth Parliament should legislate in respect of such issues. Given that fact, it is likely that communication about such topics could reasonably be argued to constitute political communication.²⁴ Again, the worker argued the dismissal contravened the relevant enterprise agreement’s guarantee of academic freedom, but again that argument was rejected by the Full Court of the Federal Court.

B Settlement Agreements

The second example this article examines is settlement agreements containing non-disparagement or confidentiality clauses. For example, to continue the immigration detention centre theme, the Commonwealth or a private contractor might make a contract with an asylum seeker to settle a claim against the Crown concerning injuries suffered by that person while detained. This contract may well include confidentiality and non-disparagement clauses.²⁵ A confidentiality clause in a settlement deed commonly restricts parties from discussing the details of the settlement, while non-disparagement clauses are often drafted very broadly to forbid any disparagement of any party at all.²⁶ This has the necessary consequence that any conduct of any party that reflects unfavourably on it cannot be discussed, even if true.²⁷

To take a second example, a deed between a private employer and a worker settling a wages underpayment claim would commonly include a confidentiality and non-disparagement clause requiring the worker not to disparage the employer, which practically means the worker simply cannot talk about what happened. In this way, systematic underpayments of wages may be hidden from the public. The public are deprived of information highly relevant to the political question of the appropriate legislative response to wage theft.²⁸

III Contract Law’s Source

The nature of the freedom’s interaction with a legal rule depends on whether that rule is imposed by an act of legislative or executive power on the one hand, or the common law on the other.²⁹ The High Court has often characterised the freedom as a ‘constraint on legislative and executive power’ (or very similar language).³⁰ But it never describes the freedom as a ‘constraint’ on the common law. The language it uses to describe the relationship between the freedom and the common law is quite different. It speaks of the common law having to ‘conform with’ the freedom, rather than the freedom being a ‘constraint’ upon it.³¹

This distinction in language necessitates an inquiry into the source of the legal rule we are here concerned with—the rule that contracts are enforceable.

A contractual term limits political communication only because of the fact that a contract is enforceable by courts. An unenforceable promise to refrain from political communication is not a limit on political communication in any relevant sense. So, it is the legal rule that provides for the enforcement of contracts that burdens the freedom.

Where a contract is formed between two individuals, it is clearly the common law that provides the rule that those individuals’ promises shall ordinarily be enforceable in courts of law, so long as they are (subject to certain important exceptions)³² mutually³³ and voluntarily exchanged,³⁴ the individuals have capacity to contract,³⁵ and they intended to create legal relations³⁶ (in this article, this common law rule regarding the enforceability of contracts is referred to as ‘the enforceability rule’). Any burden imposed on the freedom by a contractual term agreed between natural persons is a burden imposed by the common law.

This is straightforward. But as soon as contractual parties other than natural persons are introduced, the analysis becomes more complicated. For example, at common law, corporations only have capacity to form contracts if the contract was written and sealed, or in other limited circumstances.³⁷ But s 124 of the Corporations Act 2001 (Cth) confers corporations constituted under that Act the same capacity to enter contracts as an individual.³⁸ The position is similar for statutory corporations like the SBS, mentioned above. Section 44(1)(k) of the Special Broadcasting Service Act 1991 (Cth) empowers the SBS to ‘do all things that are necessary or convenient to be done for or in connection with, or as incidental to, the performance of its functions’, including the ‘power[]…to enter into contracts’.³⁹

So, the rules concerning enforcement of a contract to which a corporation is party remain the province of the common law, but the contract was only able to be formed at all because of a statutory modification to the common law. How is this to be accounted for in the legislative and executive power/common law dichotomy referred to above? As Leeming observed:

‘Common law’ is a deeply attractive, but also a deeply misleading concept. What is commonly thought of as ‘common law’, namely, the various bodies of judge-made law…taught in law schools and written about in law books is and always has for the most part been sourced in statute and is unintelligible without reference to statute. Most of the time, as Windeyer J said, ‘it is misleading to speak glibly of the common law in order to compare and contrast it with a statute’.⁴⁰

But, as noted above, the jurisprudence concerning the freedom does appear to distinguish between, on the one hand, exercises of legislative and executive power (which obviously includes statutes) and on the other hand, the ‘common law’.

Arms of federal and state government are also routinely conferred by statute with wide capacities to contract in respect of certain subject matters. For example, s 8 of the Income Tax Assessment Act 1936 (Cth) permits the Commissioner of Taxation to enter contracts settling disputes between the Commissioner and taxpayers.⁴¹ Section 198AHA of the Migration Act 1958 (Cth) grants the Commonwealth an extremely broad power to take any action, make any payment, or do anything incidental or conducive to any such action or payment, in relation to the arrangement with Nauru or the regional processing functions of Nauru. This provision would likely extend to entering a settlement contract with an asylum seeker who had made a claim against the Commonwealth in relation to wrongs done to them while in immigration detention.

Where a government lacks a statutorily conferred capacity to enter a contract, any contractual obligation upon the Crown to pay money will be unenforceable,⁴² unless it involves an exercise of an identifiable ‘non-statutory executive power’ to contract. In Williams v Commonwealth,⁴³ the High Court appeared to accept that the Commonwealth Crown (and presumably also State Crowns) has capacity to make contracts without legislative authority in the ‘ordinary course of administering a recognised part of the Government’, as an incident of ‘the ordinary and well-recognised functions of that Government’,⁴⁴ and perhaps also, directly pursuant to s 61 of the Constitution, where a contract is ‘necessary or reasonably incidental to the execution and maintenance of a valid law of the Commonwealth once that law has taken effect’.⁴⁵

The latter capacity is clearly sourced directly from the Constitution. But otherwise, the source of these non-statutory capacities is unresolved.⁴⁶ The term ‘non-statutory executive power’ is itself obfuscatory, serving to conceal the confusion about the source of these ‘powers’.⁴⁷ The capacities might be common law prerogative powers (though this is doubted),⁴⁸ ‘common law capacities’,⁴⁹ directly granted by the Constitution,⁵⁰ or a mixture of some or all of these.⁵¹ It is not necessary to resolve these questions. It suffices to say that these non-statutory capacities to contract are either sourced in the common law or the Constitution.

These capacities may plausibly support a settlement agreement entered into by the Crown if it settled a claim brought in respect of an act of a ‘recognised part of the Government’, like a government department, or in respect of an act of a Crown servant purportedly acting pursuant to a law of the Commonwealth. That likely encapsulates a significant proportion of claims brought against the Crown. Many such claims are likely to concern disputes involving facts that, if known to voters, would have the capacity to affect their political opinions concerning whether, and if so how, the Commonwealth Parliament should legislate in respect of a particular subject matter. The communication of those facts would therefore at least arguably constitute political communication.⁵² Similarly, an employment contract may be considered a contract made in the ‘ordinary course of administering a recognised part of the Government’.

Thus it can be concluded that, except where a contract is formed between two individuals, contract law involves a mixture of common law and statute, and in the case of non-statutory Crown capacities, perhaps also the Constitution itself. But the common element of all contracts is, whatever the source of the rule permitting their formation, all contracts rely on the common law of contract (and only the common law of contract) for their enforcement. Common law rules concerning enforcement of contracts continue to apply to all contracts, whatever the source of the parties’ capacity to form the contract. That analysis is powerfully supported by parts of the High Court’s reasoning in Griffith University v Tang,⁵³ where the Court considered whether a decision made pursuant to a contract by a statutory corporation with a statutory power to enter contracts could be characterised as a ‘decision…made under an enactment’ for the purposes of the Judicial Review Act 1991 (Qld). The plurality adopted what Davies and Einfeld JJ had held in General Newspapers Pty Ltd v Telstra Corporation:⁵⁴

A contract entered into by a corporation under a general power to enter into contracts is not given force and effect by the empowering statute. The empowering statute merely confers capacity to contract, whilst the validity and effect of the contract is determined by the ordinary laws of contract.⁵⁵

So the common law is the source of the enforceability rule, the legal rule permitting the enforcement of contracts that burdens the freedom.

IV Does the Enforceability Rule Infringe the Freedom?

A The Purpose of the Enforceability Rule and Its Legitimacy

In order to determine whether the enforceability rule infringes the freedom, it is necessary to apply the ‘Lange test’ as modified by Coleman v Power,⁵⁶ McCloy v New South Wales ⁵⁷ and Brown v Tasmania ⁵⁸ (referred to as the ‘McCloy test’ in this article, following the High Court’s practice).⁵⁹ The McCloy test requires that, once it is established that a law burdens the freedom, we determine whether the law has a legitimate purpose, in the sense it is compatible with the maintenance of the constitutionally prescribed system of representative and responsible government.⁶⁰ This test is applied to ‘common law rules, as they have traditionally been understood’ in just the same way it is applied to provisions of a statute.⁶¹

Applying this test requires analysis of an antecedent question: what is the purpose, or end, of the enforceability rule, and contract law more generally?⁶² While it may seem very odd to inquire, as a step in the analysis, as to the purpose of an entire field of common law, that is precisely and expressly what the Court did in respect of the common law of defamation in Lange.⁶³

The purpose of contract law is a large question, about which economic⁶⁴ and behavioural psychology⁶⁵ analyses have been given. It is, however, very difficult to locate judicial pronouncements on this question. In Australia, at least, ‘[c]ontract law has been developed at the margins. It has rarely been questioned in its fundamentals.’⁶⁶ For present purposes, then, it suffices to look to Atiyah’s Introduction to the Law of Contract (‘Atiyah’), which asks:

‘[W]hat, if anything, is the justification for contract law?…[O]n what basis is it legitimate for the state, acting through the courts, to sanction individuals for breaking contracts? Why lend the state’s support to what is an essentially private complaint?’⁶⁷

Atiyah offers two straightforward responses.⁶⁸ First, the utilitarian justification that contract law facilitates and encourages mutually beneficial exchanges. Each beneficial exchange that occurs in a society better allocates the resources of that society in a way that makes the members of that society happier or wealthier than they were before.⁶⁹ Second, the moral justification that someone who forms a contract, but then fails to perform the contractual obligations they have accepted, has perpetrated a moral wrong, and it is the courts’ proper function to remedy such wrongs.⁷⁰

In Lange, the purpose of the law of defamation was said to be the protection of the right to reputation, but in a way that balanced that right with ‘freedom of speech’. The Court went on to hold that:

The protection of the reputations of those who take part in the government and political life of this country from false and defamatory statements is conducive to the public good. The constitutionally prescribed system of government does not require—to the contrary, it would be adversely affected by—an unqualified freedom to publish defamatory matter damaging the reputations of individuals involved in government or politics.⁷¹

Thus, the Court found that the purpose of defamation law—‘protecting personal reputation’—was a ‘legitimate end’.⁷²

Similarly, the purposes of the law of contract are conducive to the common good. Australia’s constitutionally prescribed system of government would be seriously adversely affected if there was no law of contract, and mutually exchanged promises could simply be disregarded with impunity. It follows, therefore, that the law of contract, including the enforceability rule serve a ‘legitimate’ ends in the relevant sense—that is, they are compatible with the maintenance of the constitutionally prescribed system of representative and responsible government.

B Nature and Extent of the Burden

The next stage of the McCloy test is to ask if the relevant law is reasonably appropriate and adapted to its purpose—here, the maximisation of societal utility and the remedying of opprobrious conduct—having regard to the constitutional requirement of freedom of communication about government and political matters.⁷³

To answer this question, it is necessary to determine the nature and extent of the burden the relevant law imposes on the freedom.⁷⁴ Common law rules concerning contracts are different to other common law rules because contracts are expressions of the free will of the parties to them, rather than rules binding us regardless of our wishes. Joseph posits this difference might suggest contract law really does not ‘burden’ the freedom at all.⁷⁵ But Joseph answers this argument, noting it has force only if one conceptualises the freedom as a ‘personal right’, whose rationale is bound up in respect for the individual’s autonomy and dignity.⁷⁶ In other jurisdictions, this is a commonly cited rationale for protections of free speech.⁷⁷ But as courts have consistently emphasised,⁷⁸ the freedom is not a ‘personal right’ in this sense. In Unions NSW v New South Wales,⁷⁹ the plurality explained:

[W]hat the Constitution protects is not a personal right. A legislative prohibition or restriction on the freedom is not to be understood as affecting a person’s right or freedom to engage in political communication, but as affecting communication on those subjects more generally.…[T]he question is not whether a person is limited in the way that he or she can express himself or herself, although identification of that limiting effect may be necessary to an understanding of the operation of a statutory provision upon the freedom more generally. The central question is: how does the impugned law affect the freedom?⁸⁰

The freedom’s purpose is to ‘ensure that the prescribed system of representative and responsible government is preserved’.⁸¹ Indeed, the operation of the freedom is expressly ‘limited to what is necessary to the effective working of the Constitution’s system of representative and responsible government’.⁸² Thus, the freedom is concerned with protecting a societal interest, not an individual, personal interest.⁸³

Once this is kept firmly in mind, it is plain that one’s consent to a burden does not mean that the freedom’s application has been ‘waived’. An individual may not ‘waive’ their ‘right’ to the freedom, just as the High Court has repeatedly held it is impossible for an individual to ‘waive’ the ‘right’ to a trial by jury under s 80 of the Constitution.⁸⁴ The reasoning is similar. As Stellios argues, the constitutional guarantee of trial by jury is not a ‘personal right’. Its justification (like the freedom’s) is not about individuals but about institutions. ‘[I]t facilitates the exercise of Commonwealth judicial power and strikes a balance between the needs of the federal polity and the concerns and continued involvement of the States in the exercise of criminal justice.’⁸⁵ Thus, a person’s consent to a trial without jury is irrelevant to whether or not the guarantee applies to that person.

But although one’s consent to a burden upon the freedom does not ‘waive’ the freedom, the consent may still be relevant to an assessment of the nature and extent of that burden.

An assessment of the nature and extent of a burden on the freedom requires one to look to the practical operation of the burden. In a contractual context, the practical operation of a term burdening the freedom is that if one engages in political communication, then one may be sued for breach of contract, or even (in appropriate circumstances) made the subject of an injunction, rendering the political communication a contempt of court (and so a criminal offence). Furthermore, in an employment contract context, the person may also be dismissed from their employment. Banerji suggested that (at least in a Public Service setting) a power to dismiss someone because of political communication places a burden on the freedom,⁸⁶ and Justice Edelman said it was a ‘deep’ burden, because an individual’s loss of employment could be ‘catastrophic’ for that individual.⁸⁷

But where a burden is imposed by a contractual term, might the burden be considered less significant because it was consented to? Again, it is important to focus on the practical operation of the burden. If, practically speaking, the relevant contractual term was negotiated by and freely entered into by the relevant person, then it would follow that the nature of the burden is shallower than what it would otherwise be. Yet conversely, in many cases, a person’s consent to a contractual term is more theoretical than real, especially in an employment context where the contract is generally pro forma, formed before an occasion for political communication is contemplated, and where refusal might leave the employee destitute. Where an examination of the practical circumstances reveals there was no meaningful negotiation of the term or consent to it (which will often be the case in both employment and settlement contexts), the burden should be considered a significant one.

Finally, in assessing the nature and extent of the burden, it is relevant to note that most of the examples of contractual restrictions given earlier in this article are not merely ‘content-neutral time, place and manner’ restrictions on political communication,⁸⁸ but actually prohibit communication altogether, or are ‘content-based’ restrictions. A ‘content-based’ restriction on political communication generally ‘demands closer scrutiny’ than a ‘content-neutral’ restriction.⁸⁹

C Is the Enforceability Rule Reasonably Appropriate and Adapted to Its Purpose?

Having regard to the depth of the burden, is the enforceability rule reasonably appropriate and adapted to its legitimate purposes? In Lange, it was necessary to look at the defences to defamation to answer this question.⁹⁰ Similarly, in respect of the enforceability rule, it is necessary to look at the exceptions to that general rule. The above simplified account of contract law’s purposes might suggest all mutual promises should be enforceable. That is not so. Many contractual terms are unenforceable on the basis of ‘public policy’.⁹¹ These include terms deemed immoral, like those concerning prostitution, and terms not conducive to certain beneficial societal ends, like terms that unreasonably restrain trade.⁹² The term ‘public policy’ has been defined very broadly. Jordan CJ explained that ‘[a]s a general rule…any type of contract is treated as opposed to public policy if the practical result of enforcing [it] would generally be regarded as injurious to the public interest’.⁹³ The contract must offend

the ideas which for the time being prevail in a community as to the conditions necessary to ensure its welfare…Public policy is not, however, fixed and stable. From generation to generation ideas change as to what is necessary or injurious, so that “public policy is a variable thing…”…New heads of public policy come into being, and old heads undergo modification.⁹⁴

These ‘public policy’ exceptions further the purposes of contract law, because they are aimed at preventing immorality and the diminution of societal utility generally. Thus the enforceability rule strikes a balance between ensuring parties perform contracts they make, but discouraging contracts inconsistent with contract law’s underlying purposes. So far, that balance has not involved discouraging contracts that restrain political communication—such contracts have not been considered contrary to public policy. But, given the open-ended definition of contracts against ‘public policy’ that encompasses anything offensive to ‘the ideas which for the time being prevail in a community as to the conditions necessary to ensure its welfare’, it certainly seems arguable that contracts that restrain political communication could well be characterised as being against ‘public policy’ in the necessary sense.

In determining whether a law is ‘reasonably appropriate and adapted’ to achieve its legitimate purpose without unduly burdening the freedom, one must ask whether the law is ‘suitable’, ‘necessary’ and ‘adequate in its balance’.⁹⁵ A law is ‘suitable’ if it has a rational connection to its purpose.⁹⁶ The enforceability rule is certainly suitable. A law is ‘necessary’ if there is no obvious, compelling and reasonably practical alternative means of achieving the same purpose with a less burdensome effect on the freedom.⁹⁷

It is questionable whether the enforceability rule is ‘necessary’ in this sense. An obvious, compelling and reasonably practical means by which the enforceability rule could achieve a less burdensome effect on the freedom would be ‘the striking of a different balance’⁹⁸ in drawing the line between enforceable terms and terms unenforceable because contrary to public policy. This different balance would include free political communication as one of the principles, like freedom of trade, that enlivens the ‘public policy’ exception to contracts’ enforceability. Such a shift would be eminently practical and would not substantially inhibit the achievement of the enforceability rule’s legitimate purposes. Like in Lange, it would not require the creation of any elaborate new doctrine but merely the recalibration of an existing doctrine to accommodate the freedom.

Thus, in not presently identifying such a category of term as unenforceable as contrary to ‘public policy’, the enforceability rule burdens the freedom in a way that is unjustified. The enforceability rule therefore infringes the freedom.

V The Freedom’s Application to the Common Law

What consequences flow when a common law rule infringes the freedom? In Lange, the Court held the common law of defamation must not be inconsistent with the freedom, and because it was, the common law defence of qualified privilege had to be enlarged to encompass communications made to the public on government or political matters, so long as the publisher’s conduct was reasonable and was not actuated by malice.⁹⁹ The Court said:

Of necessity, the common law must conform with the Constitution. The development of the common law in Australia cannot run counter to constitutional imperatives. The common law and the requirements of the Constitution cannot be at odds. The common law of libel and slander could not be developed inconsistently with the Constitution, for the common law’s protection of personal reputation must admit as an exception that qualified freedom to discuss government and politics which is required by the Constitution.¹⁰⁰

This passage seems straightforward. Where the common law is inconsistent with the freedom, the common law must be altered. Yet the Court was careful in other parts of Lange ¹⁰¹ and in subsequent decisions¹⁰² always to say the freedom places limits on ‘the exercise of legislative or executive power’,¹⁰³ but to omit reference to the common law (or judicial power). This is the bifurcation between, on the one hand, legislative and executive power, and on the other hand, the common law, to which this article has already referred.

Stellios calls this apparent conflict ‘puzzling’.¹⁰⁴ Stone notes that, despite the Court’s apparent suggestion that only legislative and executive power are directly subject to the Constitution,

the Court’s…analysis [in Lange]…actually treats the common law as if it were subject to the Constitution by requiring that the common law conform to the Constitution.…[T]he High Court is already acting as if the common law is subject to constitutional requirements…¹⁰⁵

Mason seems to agree:

The effect of [Lange] was that both statute and common law may provide for a greater degree of protection of political discourse [than that required by the Constitution] but they cannot provide for less. At first glance this view may appear to be inconsistent with the Court’s earlier statement [in Lange] that the implied freedom of political communication is a restriction on legislative and executive power. The response to this argument is that, as the purpose of the implied freedom is to protect the constitutional system of representative and responsible government, the common law cannot be inconsistent with the freedom.¹⁰⁶

If Stone and Mason are right, and statute and common law are each subject to the freedom, why didn’t the High Court just say so?¹⁰⁷ Hill and Stone¹⁰⁸ have speculated that the Court’s apparent opacity on this issue may be explicable by the fact that Lange was a unanimous judgment of a Court that was formerly sharply divided on the question of the interaction of the common law and the Constitution in Theophanous v Herald & Weekly Times Ltd (‘Theophanous’)¹⁰⁹ and Stephens v West Australian Newspapers Pty Ltd.¹¹⁰ The judgment therefore represents the outcome of a difficult compromise, which might mean different parts have different emphases and are awkward to reconcile.

There is now an extensive body of literature on the question of whether the common law must conform with the Constitution. A review of the literature makes clear that the overwhelming majority of commentators have accepted that Lange establishes the proposition that the common law must conform with the Constitution, including implications arising from the Constitution, such as the freedom.¹¹¹ But at least one scholar, Greg Taylor, has been interpreted¹¹² to argue that Lange only determined that the Constitution was an influence informing the development of the common law, not that it actually requires that inconsistent common law be discarded.¹¹³ Taylor relies on the High Court decision of John Pfeiffer Pty Ltd v Rogerson (‘Pfeiffer’),¹¹⁴ in which principles concerning the Federation arising from the Constitution were taken into account in developing the common law concerning choice-of-law rules between states.¹¹⁵ In fact, the more convincing argument made by (at least) Foley,¹¹⁶ Hill and Stone,¹¹⁷ Goldsworthy¹¹⁸ and Selway¹¹⁹ is that there are two ‘modes’ by which the Constitution may interact with the common law. First, where the common law and Constitution are inconsistent (as in Lange), the common law must develop to become consistent. Second, where there is no inconsistency, but simply values arising from the Constitution relevant to a common law doctrine (as in Pfeiffer), those values may influence or inform the common law’s development as the Court sees fit. Indeed, in a more recent article, Taylor has apparently clarified that he too endorses a similar view about the interaction of the Constitution and the common law.¹²⁰

In Roberts v Bass, ¹²¹ Gaudron, McHugh and Gummow JJ reiterated that the freedom

requires that the rules of the common law conform with the Constitution…In determining whether a rule of the common law is consistent with the [freedom], two questions have to be answered. First, does the rule effectively burden the freedom? Second, if so, is the rule reasonably appropriate and adapted to serve a legitimate end compatible with the constitutionally prescribed system of representative and responsible government? If the answer to the second question is ‘no’, the common law rule must yield to the constitutional norm, for the common law’s impact on the freedom cannot be greater than that permitted by the constitutional norm. In Lange, the Court held that the law of defamation…[was inconsistent] with the freedom…That necessitated the development of the common law…¹²²

Since Roberts, the question of the interaction of the freedom and the common law has seldom arisen.¹²³ One exception was Aid/Watch Incorporated v Commissioner of Taxation.¹²⁴ That case determined that the common law definition of a charity should alter so as to include charities with political purposes.¹²⁵ The majority reaffirmed that:

Any burden which the common law places upon communication respecting matters of government and politics must be reasonably appropriate and adapted to serve a legitimate end in a manner which is compatible with the maintenance of that system of government.¹²⁶

Joseph has remarked that:

Lange established that…at least the common law of defamation…must conform to the freedom. The freedom also informs the common law of trusts. So it [would be] controversial to assume that it has no impact on contract law.¹²⁷

Indeed, the freedom’s effect on the common law of contract has already been recognised. In Bennett v Human Rights & Equal Opportunity Commission (‘Bennett’),¹²⁸ Finn J considered in obiter dicta whether a public servant had breached the duty of fidelity implied at common law into all contracts of employment by disclosing confidential information.¹²⁹ Finn J observed that

the common law duty must only be developed in a way that ensures that the protection given to the various interests which sustain the duty in a public service setting “does not unnecessarily or unreasonably impair the freedom of communication about government and political matters which the Constitution requires”: Lange, at 568.¹³⁰

If, as Finn J held, implied contractual terms must be consistent with the freedom, it surely follows that the rest of the corpus of the common law of contract must also be consistent with the freedom. Indeed, Finn J earlier alluded to these matters in Hughes Aircraft Systems International v Air Services Australia (‘Hughes’),¹³¹ handed down merely eight days before Lange. Hughes concerned a ‘tender process contract’ between the Civil Aviation Authority and a tenderer that relevantly required the Authority to keep the details of the tender confidential. The Authority voluntarily revealed sensitive price information about the tender to the Minister for Transport. Justice Finn found the Authority breached the contract because the disclosure was voluntary, but held that if the Minister had required the information, the contract would have been unenforceable in respect of such a disclosure because:

The [Authority]…operated in the constitutional environment of responsible government. This necessarily entails that it was accountable in some measure to the public…One manifestation of that accountability…[is its] accountability first to the Executive government through [its] respective minister, and then to Parliament via that minister. It is the minister to whom questions in Parliament are directed;…who, within the Government is given portfolio responsibility for the corporation and its legislation; [and] who…is given both specific [statutory] oversight powers and a general and specific direction powers.

In such a setting—statutory and constitutional—the Minister should be taken as having a general right to obtain information from the [Authority]…Parties who contract with government agencies must, in matters of confidentiality, be taken to have done so subject to such lawful rights of access to information in the agency’s hands as our laws and systems of government confer on others.¹³²

Tom Brennan contemporaneously observed that Hughes ‘provides an example of what…is compelled by the reasoning of the Court in Lange’.¹³³ Hughes is limited to its unique context, but this early decision underscores the point that there is no reason why constitutional requirements cannot affect the common law of contract.

VI Objections to the Freedom’s Application to Contract Law

Bennett dealt with implied contractual terms. Where express terms are concerned, judges have been less receptive to the idea the freedom has work to do.

In Banerji v Bowles,¹³⁴ an injunction was sought to prevent the then-foreshadowed termination of Ms Banerji’s employment. Unrepresented at the time, Ms Banerji argued the termination would contravene her ‘right’ to political communication. Judge Neville held:

The unbridled right [to political communication]…does not exist.…[E]ven if there be [such] a constitutional right…it does not provide a licence (allegedly) to breach a contract…[T]he [freedom of political communication] is not a personal right. It operates as a restriction on legislative power and does so to support the constitutional imperative of the maintenance of representative government.¹³⁵

In Dickson v Downer EDI Works Pty Ltd,¹³⁶ an unrepresented litigant suggested (among other arguments) that he was suspended and dismissed because he sent some work emails containing political communications. The argument proved unnecessary to decide, but Ross J remarked in obiter dicta that the freedom did not assist the applicant’s argument because ‘[t]he implied freedom of communication is not a personal right but rather operates as a restriction on legislative power’.¹³⁷

Similarly, in Mbuzi v Griffith University,¹³⁸ an unrepresented litigant argued the termination of his PhD candidacy was a termination of a contract that infringed the freedom. Justice Collier dismissed the constitutional argument because the freedom ‘does not allow an absolute and uncontrolled licence’¹³⁹ to communicate politically and did not apply to a ‘private matter’ involving no exercise of legislative or executive power.¹⁴⁰

Some academics, writing in the popular press, have suggested the freedom does not affect contract law. Triggs wrote: ‘While we may say what we please, subject to defined prohibitions, a practical, chilling outcome of freedom of speech is that we must suffer the consequences if that speech is also a breach of an employment contract.’¹⁴¹

Forsyth similarly wrote:

[The freedom] does not…give an employee licence to breach his or her employment contract…Employees in much of the private sector have their political views restricted by company codes and policies that require them not to damage the reputation of the business. These employees cannot invoke the [freedom] to support their right to speak out.¹⁴²

Much of these analyses focus on rejecting the suggestion the freedom operates like a ‘trump card’ that, once pleaded, automatically renders lawful any act that would otherwise be unlawful.¹⁴³ This notion of the freedom as ‘trump card’ is variously expressed as its being a ‘licence’ or a ‘personal right’. But the underlying idea is the same. It is, respectfully, certainly correct to say that the freedom is not a personal right in the sense that it acts as a ‘trump’ that a person can ‘play’ to absolve themselves of any legal wrong.

But it is wrong to suggest that any assertion that the freedom applies to the common law is really an assertion that the freedom operates as a personal right. In some of the cases mentioned above, it is clear that the litigant did indeed suggest that the freedom operates as a personal right, and it was, respectfully, correct to reject any such argument. However, in some of the other quotations above, it appears that the authors have simply interpolated that the assertion that the freedom is a personal right is a necessary step in reasoning to the conclusion that the freedom applies to the common law. That analysis is, respectfully, misconceived. The much-repeated assertion that the freedom does not operate as a personal right originates from Lange itself,¹⁴⁴ which was of course a case in which an inconsistency between the common law and the freedom necessitated the development of the common law so as to remedy that inconsistency. It cannot be, therefore, that the fact the freedom does not operate as a personal right means that it has no effect on the common law. If this were so, then not only the Court in Lange, but also the Court in Aid/Watch, and Finn J in Bennett, all reached a misconceived view of the freedom’s operation.

What does it mean, then, to say that the freedom is not a ‘personal right’? It means that, firstly, in Hohfeldian terms, the freedom is an ‘immunity’ rather than a ‘claim-right’ (or any of the other Hohfeldian classifications).¹⁴⁵ That this is so can be seen from Brennan J’s description of the freedom in Cunliffe v Commonwealth, ¹⁴⁶ endorsed by the Court in Lange, as being not a positive right, but ‘negative in nature: it invalidates laws and consequently creates an area of immunity from legal control, particularly from legislative control’.¹⁴⁷ The freedom does not afford individuals a right of action when they have been, or may be, affected by an alleged infringement of the freedom.¹⁴⁸ Nor does it require governments (or anyone else) to take positive steps to maximise opportunities for, or facilitate, political communication. As Brennan J explained in Theophanous, ‘[i]f the freedom implied in the Constitution were a personal right or immunity, it would extend to what is needed to facilitate or permit its full enjoyment.’¹⁴⁹

Another sense in which the freedom is not a ‘personal right’ is that its rationale is not connected with any notion of the individual’s personal liberty or dignity. Its rationale is institutional—the protection of representative and responsible government, which can only meaningfully exist if there is a degree of freedom to communicate about political matters.¹⁵⁰

Yet, while it is insufficient to simply say that the freedom does not apply to the common law of contract because it is not a personal right, there remains a deeper question requiring further analysis: is there anything about the common law of contract that means that the freedom does not interact with it in the same way that the freedom interacted with the common law of defamation in Lange?

Both Joseph¹⁵¹ and Stone¹⁵² have (briefly) grappled with this question and raised a possible concern: Contract law is concerned with private relations. The Constitution (it is suggested) is concerned with constraining governments, not private parties. It is inappropriate for constitutional guarantees to interfere with relations between private entities.

Joseph also alludes to a second potential objection: the common law may only be developed incrementally and by reference to existing principles. To develop contract law to permit the undoing of bargains freely agreed to would (it might be argued) be a radical departure from the fundamental principles of contract law, and inconsistent with the common law method.

Both of these objections are examined below.

A Inappropriate Interference with Private Relations

The first point to be made against the objection that the freedom cannot interfere with private relations is that that is precisely what did occur in Lange. Lange did not concern any exercise of legislative or executive power. It solely concerned relations between two private parties.¹⁵³

But is there a sense in which it may be said that contract law is ‘more private’ than defamation law? To address this objection, it is instructive to look to the United States. In the United States, there is a well-established ‘state action’ doctrine which demarcates the application of constitutional guarantees. Constitutional guarantees only apply to actions in which the state is ‘involved’ (the state here meaning any government, state or federal).¹⁵⁴ What constitutes ‘state involvement’ in an action, and whether the enforcement of common law rules about private relations counts as ‘state action’, is a difficult question.¹⁵⁵ It is clear from cases such as New York Times Co v Sullivan ¹⁵⁶ that the enforcement of some common law rules, such as defamation, is ‘state action’, and subject to constitutional guarantees.¹⁵⁷ In Shelley v Kraemer (‘Shelley’),¹⁵⁸ the US Supreme Court considered a restrictive covenant (that is, essentially a contractual obligation) that forbade non-white people from occupying a particular property. The Court held that a court’s enforcement of that covenant pursuant to the common law violated the Fourteenth Amendment to the US Constitution (which provides for equal protection of the law) because the court’s enforcement of the covenant constituted ‘state action’:

We have no doubt that there has been state action in these cases in the full and complete sense of the phrase.…It is clear that but for the active intervention of the state courts, supported by the full panoply of state power, petitioners would have been free to occupy the properties in question without restraint.¹⁵⁹

But Shelley is now largely considered anomalous by scholars,¹⁶⁰ and subsequent decisions have sought to avoid applying it.¹⁶¹ The received opinion is that enforcement of private contracts will not ordinarily constitute ‘state action’ and is thus not subject to constitutional guarantees.¹⁶² For scholars like Sullivan and Gunther,¹⁶³ the idea that enforcement of private arrangements should be subject to constitutional guarantees is objectionable and leads to absurdity: ‘Given the entanglement of private choices with law, a broad application of Shelley might in effect have left no private choices immune from constitutional restraints.’¹⁶⁴

Stone argues that, like the American doctrine of ‘state action’, a broad application of the principle that the common law must conform to the Constitution presents similar dangers of troubling results.¹⁶⁵ ‘[T]he owner of private property who wished to limit the discussion of federal politics, or limit advocacy of particular views about federal politics, on that property could be subject to the freedom’ because ‘[t]he owner’s private choice to restrict free political discussion is ultimately backed by the law of trespass’.¹⁶⁶

Stone sees this as objectionable:

[W]e have a strong intuition that such action should be regarded as private, rather than governmental. The state seems really to be acquiescing in the private choices of another. However, it is difficult to find a satisfactory means of distinguishing between circumstances where the state’s role in providing the legal framework for private action is mere inaction and circumstances where it is constitutionally problematic.¹⁶⁷

Stone suggests a means of distinguishing when the common law’s role is ‘mere inaction’ and when it is ‘constitutionally problematic’ would be that a common law rule must conform to a constitutional principle only where ‘the content of [the common law rule]…violates the relevant constitutional principle’.¹⁶⁸ Stone notes that in Lange

the content of rules of the common law offended [the freedom]. The common law of defamation simply did not give sufficient protection to political discussion.…[T]he unconstitutional feature of the arrangement is evident in the face of the rule, rather than in the private ordering that it is used to support. ¹⁶⁹

This may be contrasted with Shelley, where ‘the law relied upon (the law of restrictive covenants) was, on its face, apparently neutral, and the unconstitutional racial discrimination lay in the private arrangement (the particular restrictive covenant at issue)’.¹⁷⁰ So Stone suggests the common law must only conform to the Constitution where the relevant common law rule’s infringement of the freedom is evident in the face of the rule.

Such a distinction seems problematic. First, nothing expressed in Lange or any other judgment supports it. Second, the nature of common law rules makes it difficult to define what is and is not on ‘the face of the rule’.¹⁷¹ But the most powerful objection is it would impose a substantively different test on the common law to that applied to statutes. It is well-established that scrutinising legislation for infringement of the freedom requires examining both its legal effect and practical operation, not just its ‘face’.¹⁷² The apparent rationale for the proposed differential treatment of common law is that the common law deals with private relations. But there is nothing peculiarly private about the common law. Much common law deals with public matters, and much statute deals with private relations.¹⁷³ Indeed, in 2012, the Attorney-General seriously proposed codifying the rules of contract in legislation.¹⁷⁴ It is difficult to understand why the freedom’s operation upon a law should depend solely upon its source.

It is respectfully suggested that the trespass example seems intuitively objectionable because: first, in a capitalist liberal society such as ours, property and privacy rights are of fundamental importance. Laws which protect those rights (the laws of trespass) form an important part of the very foundations upon which our social order is built. Second, any conceivable burden placed on political communication by the law of trespass will necessarily be a ‘time, place and manner’ restriction. Such restrictions are generally ‘insubstantial and indirect’ compared to a ‘content-based’ restriction.¹⁷⁵ Thus, it seems wrong that an integral building block in our social order, the law of trespass, should be undermined merely so that people may be free of a prohibition on engaging in political communication in a particular location.

But the McCloy test provides an answer to this concern. It provides that a law that burdens the freedom may be justified if it is reasonably appropriate and adapted to achieve a legitimate purpose compatible with the system of representative and responsible government. The law of trespass is such a law—as mentioned, few ideas are more important to our social order than the notion of property rights.¹⁷⁶ The law of trespass recognises this by making it relatively easy for a plaintiff to establish prima facie liability, and then casting the ‘justificatory onus’ onto the defendant, who must rely on a limited range of defences which are ‘narrowly construed’.¹⁷⁷ The defences are constrained to a number of limited categories, and there is certainly no broad ‘public policy’ exception of the kind that exists in contract law.¹⁷⁸ For example, it is not any defence to the tort of trespass that one was only on the land to protest peacefully,¹⁷⁹ or to retrieve information about moral or legal wrongdoing on the land.¹⁸⁰ The exceptions to trespass are narrow—they concern such matters as self-defence, defence of property, arrest, execution of process, necessity, consent and acts authorised by statute or prerogative.¹⁸¹

While the law of trespass will in some situations burden the freedom, the burden will generally be insignificant. Thus, given the overwhelmingly legitimate purpose achieved by the law of trespass on almost all, or all, occasions, the freedom will not be infringed.

It may be that there is some conceivable set of facts where the law of trespass might be thought to infringe the freedom in its operation—perhaps in respect of government-owned land that is ordinarily open to the public, where there was (for some reason) some overwhelming need that a political communication occur there and nowhere else. In those sorts of circumstances, Eldridge and Matthews have tentatively suggested that the common law of trespass might be developed to incorporate a limited exception in respect of entries upon land for the purposes of making political communications, looking to international precedents to that effect.¹⁸²

But in the overwhelming majority of (and perhaps all) instances, private property rights would be unaffected by the freedom. Thus, it is suggested that the McCloy test is robust enough to rebut the claim that the freedom might ‘leave no private choices immune from constitutional restraint’.¹⁸³ It is true many private arrangements which ultimately rely on their enforceability by a court may technically be subject to the freedom. But in cases where that seems objectionable, the balancing exercise required by the McCloy test will mostly leave those private arrangements unaffected by the freedom.

B Undermining Common Law Method

Gordon J said in Clubb v Edwards (‘Clubb’):

The development of the common law occurs in a unique and restricted way. The common law can only be developed logically and analogically from existing legal principles. This analogical quality of common law reasoning differentiates it from other kinds of legal reasoning.¹⁸⁴

The common law’s development must occur ‘incrementally’.¹⁸⁵ That is what occurred in both Lange and Aid/Watch. The developments of the common law did not radically uproot the common law or offend its underlying rationales. They were developments consistent with the time-honoured common law method of evolution to deal with new circumstances. They involved a conservative reformulation of the relevant common law.

It might be argued that the Court in Lange never intended to suggest the Constitution might require violent rewritings of common law doctrine that radically undermine the doctrine’s underlying rationales or require the undermining or abolition of some fundamental common law principle.

Failing to enforce contractual terms that infringe the freedom, even where such a term might truly represent the free will of all parties, certainly contradicts the important principle of freedom of contract and the notion of caveat emptor. Requiring contract law to conform to the Constitution by undermining such principles might go beyond what the Court in Lange intended.

But such an argument’s fallacy is its overstatement of the importance of the principle of freedom of contract to contract law; individuals have never possessed an unqualified entitlement to obtain curial enforcement of their agreements irrespective of the agreement’s contents. The principle of ‘freedom of contract’ is not as foundational to contract law as might be assumed. The reverence for ‘freedom of contract’ was limited to a brief, albeit influential, period in the 19th century.¹⁸⁶ In his famous legal-historical work, Atiyah records the ‘decline’ of freedom of contract as an inviolable principle informing contract law from 1870 onwards.¹⁸⁷ And long before 1870,¹⁸⁸ perhaps as early as the 15th century,¹⁸⁹ the English courts already recognised that terms contrary to ‘public policy’ were unenforceable. This rule is open-ended—the categories of terms contrary to public policy are not closed.

So the notion of ‘public policy’ in contract law is broad, and thus inimical to any uncompromised principle of freedom of contract. One category of contract contrary to public policy is contracts promoting sexual immorality, a head of public policy relied on as recently as 2012 by Brereton J in obiter dicta in Ashton v Pratt (No 2).¹⁹⁰ A particularly significant modern head of contracts against public policy is contracts in restraint of trade. This head developed over the 19th century and earlier and crystallised around the turn of the 20th century.¹⁹¹

Admitting a further category of unenforceable contracts—contracts contrary to the freedom of political communication—would, far from representing a radical departure from the principles underlying contract law, sit very comfortably alongside the open-ended categories of contracts unenforceable for being against public policy. Such a development to contract law would be an incremental one entirely in keeping with the common law method. Indeed, there appears to be at least one isolated case, Neville v Dominion of Canada News Co Ltd (‘Neville’),¹⁹² where the English Court of Appeal found that an agreement to withhold information from the public was unenforceable on public policy grounds. Neville concerned a contract between a newspaper and a company accused of wrongdoing. The company paid the newspaper in consideration for a promise that it would not publish any comment upon the company, its directors, business or dealings. This contract was unenforceable because ‘it was not consistent with the proper conduct of the newspaper in the public interest’.¹⁹³

VII Developing the Enforceability Rule to Accommodate the Freedom

Unreasonable restraints of trade are a notable category of contractual obligation contrary to public policy.¹⁹⁴ The well-established principles for determining whether a contractual term is an unreasonable restraint of trade may be simply summarised as follows:

Does the contract impose a restraint on the covenantor’s capacity to trade? This question is to be determined by looking to the ‘practical working’ and ‘truth and substance’ of the covenant, ‘irrespective of the legal form’, and avoiding ‘legal niceties or theoretical possibilities’.¹⁹⁵

Does the covenantee have a legitimate or proprietary interest to be protected?¹⁹⁶

Is the restraint of trade no wider than reasonably necessary to protect that legitimate or proprietary interest?¹⁹⁷ The covenantee bears the burden of proof in relation to this issue.¹⁹⁸

Is the restraint not otherwise unreasonable in the public interest?¹⁹⁹ The covenantor bears the burden of proof in relation to proving the restraint is contrary to the public interest.²⁰⁰

The parallels with the McCloy test are obvious: stage 1 of the doctrine corresponds with the need for an effective burden on political communication; stage 2 corresponds with both the ‘compatibility testing’ stage and the ‘suitable’ stage; stage 3 corresponds with the ‘necessary’ stage; stage 4 loosely corresponds with the ‘adequate in its balance’ stage.

The restraint of trade doctrine could be readily adapted to develop, by way of traditional common law analogical reasoning, a common law rule governing the enforceability of contractual restraints on political communication. Such a common law test might well simply be the above four numbered points, but with the first numbered point referring to restraints on the covenantor’s capacity to communicate about political or government matters, rather than to trade.

This approach has numerous advantages. First, closely following the restraint of trade doctrine ensures contract law is developed consistently with the common law method, emphasising its incremental development by close analogy with existing principles. Second, the restraint of trade doctrine’s similarities with the McCloy test means that the common law can be developed in a way that faithfully accommodates the freedom with relative ease. Third, the restraint of trade doctrine has already grappled with some otherwise potentially difficult questions about how a rule against unreasonable contractual restraints on political communication might operate.

For example, one objection against developing a doctrine resembling the McCloy test as part of contract law might be that the McCloy test requires ascertainment of a law’s purpose. Ascertaining a contract’s purpose is practically difficult—there is no second reading speech, no explanatory memorandum and often no ‘objects’ section. Sometimes a contract contains a ‘recitals’ section, which may be loosely analogous to an ‘objects’ section, but often it may not.²⁰¹ The restraint of trade doctrine accommodates this potential difficulty. Heydon explains that

[T]he court is entitled to look both at the contractual words and the surrounding circumstances in order to ascertain [the interest protected by the restraint] by reference to what objectively would appear to be the interests of the parties. However, if the [covenantee] chooses specifically to state in the contract the interest…which the covenant is intended to protect, the [covenantee] is not entitled thereafter to seek to justify the covenant by reference to some separate and additional interest…The reason given is that [a covenantor] invited to enter a covenant of that kind may wish to take legal advice as to its validity and effect before accepting it, and the legal advisers are entitled to advise on the basis of the stated purpose only.²⁰²

So, the potentially difficult question of ascertainment of the relevant interest is resolved by looking only to the parties’ objectively apparent interests, by recourse only to the contractual words and the surrounding circumstances. This approach may well be appropriate for importation into a common law doctrine concerning contractual restraints on political communication.

A second objection to adapting the restraint of trade doctrine to restraints on political communication is that it may create problems concerning the enforceability of the rest of the contract when a contractual term is unenforceable because it infringes the freedom. The restraint of trade doctrine answers this question by reference to the concept of ‘severance’. Severance involves an application of what is called the ‘blue pencil test’. As Heydon explains:

[S]everance can only apply where the only changes are those made by running a blue pencil through the offending parts. That is, severance is only possible in “cases where the two parts of a covenant are expressed in such a way as to amount to a clear severance by the parties themselves, and as to be substantially equivalent to two separate covenants”.²⁰³

So (to adapt an example of Heydon’s), if a covenant requires that a covenantor not run a competing business anywhere ‘in Adelaide or 300 kilometres thereof’, and the court finds the limitation is reasonable in respect of Adelaide, but not otherwise, the covenant is severable. The words ‘or 300 kilometres thereof’ can be ‘blue pencilled’, and the remainder of the covenant still operates grammatically. But if the restraint is found only to be reasonable in respect of the suburb of Athelstone in Adelaide, then the whole covenant is invalid, because it is impossible to ‘blue pencil’ the covenant so it only operates in respect of Athelstone. ‘Reading down’ the word ‘Adelaide’ so it only applies to Athelstone is strictly impermissible. Interestingly, Edelman J discussed in Clubb this contractual doctrine of severance in the context of determining what happens to legislation when a provision partially but not wholly infringes the freedom.²⁰⁴

The drawbacks of the ‘blue pencil test’ are immediately apparent, as Edelman J noted.²⁰⁵ The test ‘operates unfairly as between substantively similar but formally different cases’.²⁰⁶ If the above example covenant had named every suburb in Adelaide rather than stated ‘Adelaide’, the covenant would be substantively identical, but it would be severable because it would be possible to ‘blue pencil’ every suburb except Athelstone.²⁰⁷ This seems absurd. As Heydon notes, it punishes those who ‘clumsily’ draft covenants too wide by rendering them entirely unenforceable. But it rewards those who ‘maliciously’ draft a covenant that is too wide but expressed in severable terms. Such drafters are rewarded because they know that if a court finds the covenant is impermissibly wide, the covenant will be severable anyway.²⁰⁸ The justification for the ‘blue pencil’ test is that contract law’s purpose is to give effect to the bargain struck between the parties, not a ‘bargain’ the court considers reasonable but which no-one ever agreed to. It is more preferable that a court simply refuses to enforce a bargain contrary to public policy, rather than enforcing a ‘bargain’ the parties never actually agreed, but was invented by the court.

The severance doctrine’s problems are magnified when applied to contractual terms unreasonably restraining political communication. That is because, as we have seen, contractual terms infringing the freedom are often likely to do so only incidentally. For example, a contractual term requiring the employee not to bring the employer into disrepute forbids a very wide range of potential actions, most of which have nothing to do with political communication. Similarly, a confidentiality clause may, in all sorts of settings, have much legitimate work to do, but then also incidentally infringe the freedom in some circumstances. Such clauses are not severable—it will be impossible to ‘blue pencil’ the clause so it does not apply to political communication, but otherwise applies. So, the whole clause will be unenforceable. In some cases, that might have highly unjust results. For example, a person might breach a confidentiality clause, where no question of political communication arises, and mischievously escape liability simply by pointing out that the clause is unenforceable because in some unrelated but conceivable circumstance, the clause infringes the freedom, and it is not severable.

There are at least two possible answers to this dilemma. First, it might be said that, while severance might sometimes lead to injustice, it would inevitably lead to tighter drafting of clauses that even incidentally limit communication, to ensure they are severable or clearly avoid political communication entirely. This would be no bad thing. It might help limit the growth of very vague, broad contractual prohibitions that can (and are often intended to) have a substantial ‘chilling effect’.²⁰⁹

Second, there is arguably scope to develop a different approach to the ‘blue pencil test’. Edelman J has mentioned the ‘controversial’ possibility of applying to a contractual term the technique he calls ‘partial disapplication’²¹⁰ (and which Heydon calls ‘indirect severance’²¹¹). Heydon has noted some 19th-century English cases applied this technique,²¹² and that the United States,²¹³ South Africa,²¹⁴ and to an extent Canada,²¹⁵ have adopted variants of the ‘partial disapplication’ approach in contract law.²¹⁶

This would mean that, to take the example of the ‘bringing the employer into disrepute’ clause, the clause would apply to all situations where it did not infringe the freedom, but would be ‘disapplied’ to any situation where it did infringe the freedom.

The advantage of this approach is it avoids the potentially enormous inconvenience and possible injustice that might result from clauses incidentally infringing the freedom being entirely unenforceable. The disadvantage is it involves altering the bargain struck between the parties to something quite different. This is arguably contrary to underlying principles of contract law and may incentivise drafting of very broad restrictions on communication, since the covenantee knows the worst that can happen is the courts will simply ‘disapply’ the covenant to the extent it infringes the freedom, and otherwise enforce it. However, it might be suggested that the rationale underpinning the ‘blue pencil test’—the inviolability of the compact actually struck between the parties—is less compelling in light of the constitutional considerations underlying any doctrine of unreasonable restraint of political communication.

A third objection to the proposed approach is it arguably involves scrutinising every contract for infringement of the freedom, rather than the rules of contract law. The narrower the focus of the test, the more likely that the test will not be satisfied.²¹⁷ This is arguably inappropriate when it is considered the freedom is meant to have a limited operation.²¹⁸ Against this argument three points may be made. First, the proposed test is similar, but different, to the McCloy test. There may therefore be scope to accommodate such concerns within the test. Second, the common law of contract must be developed to ensure consistency with the Constitution. There is no way to do this without applying some sort of test on a contract-by-contract basis. Third, in other contexts, the High Court has not refused to apply the McCloy test at a ‘narrower’ level. For example, in City of Adelaide,²¹⁹ the High Court directly scrutinised a by-law for infringement of the freedom, rather than the legislative provision authorising the by-law.²²⁰

VIII Applying a ‘Restraint of Communication’ Doctrine

If the above test were adopted, would the examples given at the beginning of this article fall foul of it? This is a complicated question, dependent on the precise wording of particular contractual terms, and the precise context and surrounding circumstances. It also depends on what meaning the term ‘legitimate interest’ is given. In the context of restraints of trade, there is little analysis of this term. Rather, the concept has been developed, in accordance with the common law method,²²¹ by the identification of categories of ‘legitimate interests’—the protection of customer relationships,²²² confidential information²²³ and ‘know-how’²²⁴ are all examples. Illegitimate interests include bare freedom from competition for its own sake,²²⁵ and the interests of businesses associated with the covenantee, rather than the business of the covenantee itself.²²⁶

In the present context, it is suggested that an employer-imposed burden on freedom of political communication will protect a ‘legitimate interest’, at least where the burden is reasonably necessary to ensure the employee can fully and properly perform their duties as an employee. Whether that is the case may depend on a range of factors such as the scope of the restraint, the employee’s seniority and the ‘public’ nature (or otherwise) of the employee’s role (ie, are they a spokesperson or similar). Where the restraint concerns communications about an employer’s clients, additional factors may include the extent of the employee’s interactions with such clients and the type of work the employer does for any such clients.

So contractual obligations of confidentiality imposed on employees, if limited to information obtained in the course of one’s employment, surely protect a legitimate interest of the employer. These obligations (depending on the precise terms) would also generally be a reasonable restraint no wider than necessary to protect that legitimate interest. Often, it is necessary for an employer to disclose certain of its affairs or its client’s affairs to an employee in order for that employee to do his or her job. As a general rule, it must surely be legitimate for a business to restrain such an employee from publicly broadcasting the information he or she has thus obtained, even if those affairs have a political dimension. There may (or may not) be some very limited exceptions to that rule where such disclosures amount to ‘whistle-blowing’, but that question is a complicated one beyond the scope of this article.²²⁷

But restrictions on political criticism by an employee of their employer or its clients, involving no disclosure of confidential information, might be on shakier ground. Sometimes such a restraint will protect a legitimate interest. Banerji confirmed it is proper to restrain such criticism by public servants, but Banerji relied on the public service’s unique need to be (and be perceived as) apolitical.²²⁸ Thus, Banerji turned on its public service context, underscoring the importance of context in considering the appropriateness of contractual restrictions on political communication. That being said, it is instructive to note the matters of context in Banerji that may also have played some part in the result: Banerji was a mid-level public servant—not junior, but not a leader.²²⁹ Her communications were not just about the Commonwealth Government at large but were (mostly) criticisms of the policies of the Department for which she worked, the (then) Department of Immigration and Citizenship.²³⁰ Moreover, her role involved (at least according to her employer) ‘the management of external communication activities (including media liaison)’.²³¹

It is possible to imagine circumstances where restraints on employee political advocacy may also protect legitimate interests of private sector employers. The facts in Rumble present an example.²³² The employee was a senior legal advisor, who publicly disparaged a client of the firm. A client must have a minimum level of trust in a legal advisor for legal advice to be useful. That trust might well be lost if the legal advisor publicly criticises the client. Thus, restraining such criticism in these circumstances ensures the employee’s effectiveness in performing their work duties.

Yet in other circumstances, there may be no plausible connection between such a restraint and the employer’s legitimate interests. To take a slightly frivolous example, it would be difficult to argue that a legitimate interest was protected by a restraint imposed by Subway on all its ‘sandwich artists’, wherever employed, from publicly criticising Government immigration policy, just because Subway sometimes caters lunches for the Department of Home Affairs. A Subway sandwich artist’s criticism of a client will not plausibly interfere with their sandwich-making duties. On the other hand, a restraint imposed on the particular employee who liaises with the Department about such lunches might more arguably protect a legitimate interest.

Restraints forbidding public expression of political matters unconnected with the employer or its clients seem less justifiable still. At the extreme end, a restraint on employees making favourable public communications about the Liberal Party at any time, imposed merely because the employer prefers the Labor Party, does not protect any legitimate interest. It has nothing to do with an employee’s capacity to perform his or her employment. A coal mining company might argue a restraint on employees publicly supporting pro-environment policies does protect an interest of the employer beyond its mere preference. But again, it is unlikely to be a legitimate interest, because it has nothing to do with the employee’s capacity to perform his or her employment duties. If, however, the relevant employee was a public spokesperson or lobbyist for the company, the position may be more arguable.

In Scott McIntyre’s case, the restraint on making any communication bringing the SBS into disrepute protects SBS’s interest in ensuring its good reputation is upheld. This may well be a legitimate interest, especially in respect of an employee with a public profile such as a journalist. But real problems may emerge in the third step of the test—the restraint is arguably excessively vague and broad.

Turning to settlement deed confidentiality and non-disparagement restraints on political communication, two legitimate interests are arguably being protected. First, restraints on revealing the terms of settlement, it has been argued,²³³ protect defendants’ legitimate interest in discouraging vexatious ‘copy-cat’ claims being brought by opportunistic potential plaintiffs who learn of the terms of settlement. This seems a legitimate interest. However, these restraints also have the effect of depriving genuine plaintiffs with similar claims of information that would greatly assist them in negotiating a fair and favourable settlement against the defendant.²³⁴ This might mean these restraints fall foul of the fourth step of the test—the public interest balancing. Whether that would be so would depend on whether the public interest balancing stage permits public interest considerations unrelated to the freedom, such as the consideration raised above, to be weighed against the legitimate interest being protected. It is suggested that, consistent with the breadth of the ‘public policy’ exceptions to contract law, such considerations may well be permitted to be weighed in the balance. However, if that were so, other public interest considerations, such as confidentiality clauses’ (arguably) beneficial effects in encouraging settlements,²³⁵ must also be weighed in the balance. The outcome of such an analysis is difficult to predict at this abstract level, but it may well be that many restraints on revealing the terms of settlement remain enforceable.

Second, restraints on revealing the underlying facts of the dispute (which is commonly the practical effect of non-disparagement clauses²³⁶) are aimed at protecting the defendant from undeserved reputational harm.²³⁷ That must be a legitimate interest. But that legitimate interest is already adequately protected within constitutional limits by the law of defamation. A contractual restraint on revealing anything about the underlying dispute is far wider than necessary to protect the legitimate interest, and therefore must fail the third step of the test.

Against this argument it might be said that parties are surely free to agree upon restraints protecting reputational harm that go further than the laws of defamation. In some circumstances, defamation law defences permit false, defamatory things to be said, so long as they are said in good faith.²³⁸ It must be admitted that parties surely may legitimately agree at least that they will not say any false and defamatory thing about one another. But to forbid any ‘disparaging’ communication about the facts of the relevant dispute, no matter how true, seems extremely difficult to justify under the third step of the test.

While it is difficult to reach concluded views at this level of abstraction, it can be seen the proposed test could have significant and beneficial work to do in respect of employment contracts and settlement deeds, perhaps especially in relation to non-disparagement clauses in settlement deeds.

IX Conclusion

I have argued that contract law as it presently stands is inconsistent with the Constitution, and that it must be developed to conform with the Constitution. That development can occur in a way that remains faithful to the common law method, through the development of a doctrine closely analogous to the restraint of trade doctrine. Such an approach would strike an appropriate balance between enforcing contractual terms that burden the freedom in a justifiable manner, while not enforcing those terms that are unjustifiable. A problem with this approach would be that contractual terms found to incidentally infringe the freedom are unlikely to be severable, and so will be unenforceable in toto. This may have undesirable consequences. However, it is not an insuperable difficulty. The suggested approach would have a substantial and beneficial effect on restraints on political communication commonly found in employment contracts and settlement deeds, perhaps especially settlement deeds.