Public Servants and the Implied Freedom of Political Communication

A History, Hobbes v Locke and Current Debates

Paradoxically, the freedom of an individual to speech has often been under threat in jurisdictions that would regard themselves as democracies, literally government by and of the people. From earliest Greek times, not all were permitted to participate in the discussions and decisions that a community held and made. In the Middle Ages, it was a serious offence to ‘imagine’ the death of the monarch, and laws relating to sedition and treason operated to temper and deter criticism of those in power. ³ A system of prior restraint was in effect until 1694, pursuant to which only material that had been approved by those in power could be made public. Governments have long viewed with antipathy those who wish to criticise them publicly and often tried to discourage and deter such behaviour. They have often sought to control the release of information that might be considered unfavourable to them.

In the scheme of things, Australia has a reasonably strong record of embracing ‘democracy’. Whilst it was not a feature of early colonial life, by the end of the 19th century Australia’s democratic institutions were well established. ⁴ The Australian Constitution reflects strong principles of democratic government, ⁵ with numerous provisions concerned with Australia’s structure of government, and assurances of the involvement of individuals in choosing who is to sit in Parliament, and ultimate sovereignty residing in the people. The founding fathers were aware of the United States and United Kingdom models of constitutional government. They readily embraced the kind of representative government model that features in the United States Constitution of 1789, including by using phrases from that document in key Australian Constitution provisions concerning voting, including ss 7 and 24. ⁶ Democracy and representative government came later in the evolution of the constitutional history of the United Kingdom, having evolved from a constitutional (hereditary) monarchy advised by a group of councillors to a system of representative government by the 19th century.

The question of accountability, openness and transparency in government is very topical in Australia at present. Constitutional law expert George Williams stated publicly that Australia had rightly been described as ‘the world’s most secretive democracy’ and that democratic freedoms in Australia were under ‘sustained assault’. ⁷ Recent scholarship has noted the increased use of national security to justify ever-increasing incursions on freedom of communication. ⁸ The extent of protection for whistleblowers in Australia has been questioned. ⁹ A right-to-know movement has developed considerable force in this country, generally concerned with government attempts to shut down communication, to stop the media (and through it, the public) from knowing about particular matters of potential concern with how Australia is governed, and government raids on journalists, including at their homes. ¹⁰ It has been argued that this culture of increasing secrecy is at odds with democratic values. ¹¹ Criticism of government has been noted as an essential aspect of democracy. ¹² Mill noted that ‘the interest of the people is that there should be full liberty of censure on every public officer and on every public act or measure’, in contrast with the interest of those in power, which was (at least sometimes) to prohibit censure or criticism. ¹³ Informed criticism can only occur when the public have recourse to relevant sources of information. This culture of secrecy is occurring at a time when the trust that individuals have in government is at record lows. ¹⁴ Secrecy is unlikely to improve that trust.

On the other hand, governments insist upon the need for secrecy, in order to further national security or other pressing governmental interests. This attitude is seen, for example, in relation to the common response to freedom of information requests, and past efforts to condition funding on the recipient not engaging in advocacy and/or lobbying activity. It is also seen in the increased use of secret legal proceedings against those suspected of wrongdoing, ¹⁵ as well as laws specifically prohibiting a person from discussing particular matters. ¹⁶

In broad philosophical terms, this debate probably reflects different conceptions of our structure of government. At the risk of simplicity, at one end we have a Hobbesian view of all-powerful government, with strong powers needed to curb a wayward populace and to ensure a safe society. Here, law and order assumes prime importance; questioning of government, accountability and freedom of speech are very much of secondary importance. At the other end, we have a Lockean-inspired social contract theory, where governments act at the behest of and as representatives of the people. In such a system, practical means of making governments accountable to the people, including permitting people the means of knowing what ‘their’ government is doing and how it is performing, are critical. ¹⁷ English law on freedom of speech generally, including the right to criticise politicians with impunity, evolved as philosophical views on government did:

Our law of libel has, in many respects, only gradually developed itself into anything like a satisfactory and settled form. The full liberty of public writers to comment on the conduct and motives of public men has only in very recent times been recognized. Comments on government, on ministers and officers of state, on members of both Houses of Parliament, on judges and other public functionaries, are now made every day, which half a century ago would have been the subject of actions or ex-officio informations (sic), and would have brought down fine and imprisonment on publishers and authors. Yet who can doubt that the public are the gainers by the change, and that though injustice may often be done, and though public men may often have to smart under the keen sense of wrong inflicted by hostile criticism, the nation profits by public opinion being thus freely brought to bear on the discharge of public duties. ¹⁸

The High Court recently considered the interplay of the implied freedom of political communication discerned from the Australian Constitution with restrictions placed by government on the extent to which public servants could discuss political issues in public. It unanimously found that restrictions on the ability of public servants to contribute to public debate were justified, given the ideals of an apolitical and independent public service, their responsibility to give frank and fearless advice to the government of the day, and implement its legislative and policy agenda. This decision will be discussed later. While in a way the result was not surprising, because the occasions on which the High Court has struck down laws on the basis they infringe the implied freedom have been limited, the result may be troubling to those who accepted the High Court’s vision of the implied freedom in the early 1990s as being a fundamental aspect of our democratic form of government.

B Implied Freedom of Political Communication in Australia

Prior to a discussion of the specific question of free speech and public servants, I must summarise the jurisprudence on the implied freedom, with particular emphasis on the High Court decisions in Australian Capital Television Pty Ltd v Cth ¹⁹ and Nationwide News v Wills, ²⁰ the foundational cases in this area. These judgments contain comments that are considered to have particular application to the current context involving the question of the speech of public servants. While there have been numerous cases in this area since, the original cases continue to provide the essential intellectual foundation for the implied freedom. Prior to those decisions, freedom of speech was protected in Australian law in the orthodox common law way — the right existed to the extent it was not removed by statute. ²¹ It might also be defended through principles of statutory interpretation, including those suggesting that in the absence of clear words, a court might presume that Parliament did not intend to abrogate fundamental common law rights, or that where more than one interpretation of a statutory provision were open, a court might adopt the interpretation that is less invasive of common law rights. ²²

In these cases, the High Court found that the Australian Constitution contemplated a democratic form of government, including a system of representative and responsible government. This form of government required a free flow of information between voters and the government, and among voters. Individuals had to be free to access a range of information in order to effectively cast judgment at election time. This implied freedom was deduced from the text and structure of the Australian Constitution, in particular ss 7 and 24. There are three ideas in particular to be drawn from the High Court’s judgments in the initial free speech cases for current purposes. They are closely related.

The first idea is the notion of the sovereignty of the people. ²³ Mason CJ stated clearly that in the Australian Constitution, the people were the sovereign body, exercising sovereign power. The people as the repositories of this power delegated this power to their representatives in Parliament. The Parliament exercised this power on behalf of the Australian people. ²⁴ This was the essence of the system of representative government from which the implied freedom was deduced. ²⁵ As Meiklejohn put it, ‘the people need free speech because they have decided, in adopting, maintaining and interpreting their Constitution, to govern themselves rather than to be governed by others’. ²⁶

The second idea, related to the first, is that of accountability. Given our model of representative government, members of the Parliament are accountable to the Australian people, who are their masters. ²⁷ An open discussion reflecting on our politicians can be a ‘stimulus to performance’. ²⁸ McHugh J tied this in with representative and responsible government:

If [those] institutions…are to operate effectively and as the Constitution intended, the business of government must be examinable and the subject of scrutiny, debate and ultimate accountability at the ballot box. The electors must be able to ascertain and examine the performance of their elected representatives and the capabilities and policies of all candidates for election. ²⁹

Maddison refers to a study of political disengagement in the United Kingdom which showed that

Disengagement create[s] a climate in which a ‘quiet authoritarianism’ can flourish. By this, the study’s authors meant a situation could develop where governments are, effectively, no longer held to account and where policy and law are made ‘in consultation with a small coterie of supporters and with little reference to wider views and interests’. They argue that in such circumstances other democratic processes and institutions, including elections, become ‘empty rituals’. The silencing of dissenting and independent opinion in the various institutions…is a decisive move towards this form of authoritarianism…there are important reasons to value a more open and participatory form of democratic practice that encourages debate and does not shy away from or suppress dissent. ³⁰

The third idea is that discussion is required in order that voters can make an informed decision about the performance of their political representatives at election time. ³¹ Voters need as much information and discussion as possible in order to increase the chances that they will be able to accurately reflect on the performance of incumbents and to assess election platforms and policies of all political parties. ³² This includes hearing criticisms of elected officials. ³³ Meiklejohn said that the ‘primary social fact which blocks and hinders our experiment in self-government is that our citizens are not educated for self-government’. ³⁴

Thus, there is a strong link between our system of representative government and public discussion, including criticism, of those who comprise it. This necessarily involves discussion, including criticism, of Ministers of the Crown. ³⁵ The High Court has notoriously declined to articulate any theory of freedom of political communication, or embrace any of the established theories of free speech, including the search for truth, marketplace of ideas, self-actualisation or self-government. ³⁶ However, the decisions in this area are most consistent with the freedom of speech theory of self-government typically associated with Alexander Meiklejohn. ³⁷

Subsequently, the High Court has been at pains to emphasise it is not a personal right of any individual; rather it is a negative freedom, in the sense that it is a protection against law (or executive action), rather than a source of positive rights. Further, it applies only to what is ‘political’ communication, though this has been difficult to identify with precision. ³⁸ For both reasons, it is thus narrower in scope than freedom of speech provisions elsewhere. It includes non-verbal communication. ³⁹ A law that directly targets political communication will be more difficult to justify as being an acceptable infringement of the freedom, as compared with a law that is targeted at another end, with a merely incidental effect on the freedom. Laws of the former category must be justified in a manner that is ‘compelling’. ⁴⁰

Different tests have been utilised at different times in order to determine whether a law is compatible with the implied freedom. At one time, a two-stage approach was taken. ⁴¹ A majority of the High Court has now settled upon a three-stage process. ⁴² Firstly, the court will consider whether the law effectively burdens political communication, in terms or effect. If so, it considers whether it was passed for a purpose that is compatible with Australia’s system of representative and responsible government (compatibility testing). If so, the court will consider proportionality analysis. Here it will consider whether the law is suitable, necessary and adequate in its balance. Two members of the court, Gageler and Gordon JJ, continue to apply the two-stage test identified in Lange.

I will now consider the application of the freedom in the specific context of the speech of public servants.

A Competing Interests

In the context of the implied freedom of political communication in Australia, and its deep roots in representative government and democracy, the question of the freedom of public servants to speak about political matters is important. Public servants have an important role to play in educating the public about political matters. ⁴³ Given that they work in areas likely to be political, they have a contribution to make to greater public understanding of issues, including issues relevant to the choice that the Australian public make at election time, judgments on our current leaders and accountability of those in power for the decisions they make. They might be aware of information that could be used to criticise the government of the day. Writing in 2007, Hamilton and Maddison note with concern that

Individual citizens have been targeted with the apparent aim of driving them out of the public domain…the targets are most likely to be individual experts who are critical of controversial government policy. ⁴⁴

They might also play a whistleblower role, reflected in legislation such as the Public Interest Disclosure Act 2013 (Cth), since as has been noted, they may be among the first to become aware of possible wrongdoing. ⁴⁵ Section 10 states that a person who makes a ‘public interest disclosure’ cannot be the subject of civil, criminal or administrative liability as a result. Section 26 defines a public interest disclosure as one made by a person, in respect of an external disclosure, where the information shows ‘disclosable conduct’, the discloser has previously internally reported such disclosable conduct, that an internal investigation was or response to that investigation was inadequate, disclosure is not in the public interest, ⁴⁶ no more information is released than is necessary to identify the disclosable conduct, and the information does not include intelligence information or relates to an intelligence agency. Section 29 defines disclosable conduct broadly, to include unlawful activity, activity that perverts the course of justice, corruption, maladministration (including improper motives, unreasonable conduct or negligence), abuse of public trust, fabrication of scientific research, wastage of public money, conduct that unreasonably endangers safety or health of others, or will damage the environment.

This legislation is relevant because it reflects recognition, by the Commonwealth Parliament, of the important role that public servants can play in governance. It is an acceptance of the reality that those within the organisation are well placed to observe wrongdoing, based on their expertise. It recognises the value of public servants’ speech, at least in relation to certain types of conduct. Of course, there are things that public servants observe that fall outside the definition of disclosable conduct but which would still be valuable for the public to hear. ⁴⁷ Thus, the position is that, while the whistleblower legislation is an important part of our system of democratic government, there are behaviours of public servants that might fall outside the parameters of that legislation, yet still be laudable, and which the law might seek to protect, preserve and encourage. ⁴⁸

Parker alluded to this in the Australian context many years ago, reflecting on the

pervasive influence of government in the modern community, and the corresponding need, and difficulty, of effective appraisal of Ministerial policies in gestation, and effective criticism, restraint and improvement of adopted policies in action. So much of the requisite knowledge and expertise for these purposes is now absorbed in the public services that the community cannot afford that these resources should be monopolised…by ‘responsible’ Ministers. New ways and means are needed to put the knowledge and experience of the public service at the service of the public interest, and not merely of the government of the day. ⁴⁹

There is a history of government suppression of public servants’ participation in political activity in Australia. ⁵⁰ It should also be acknowledged that government might in some cases have legitimate reasons for seeking to withhold information, in pursuit of national security interests or good governance. It can be a challenge to separate what are considered legitimate reasons for preventing the release of information from governments sometimes attempting to prevent the release of information, or stopping criticism, which might simply be embarrassing or inconvenient for the public to hear about. ⁵¹

It should be acknowledged that public servants have important obligations around neutrality, independence and loyalty to government:

The civil servant is expected to give…the qualities of loyalty and discretion. [They] are not to obtrude [their] opinion unless it is invited, but when it is needed [they] must give it with complete honesty and candour. If it is not accepted, and a policy is adopted contrary to [their] advice, [they] must…do [their] best to carry it into effect, however much [they] may privately dislike it. ⁵²

Similarly, the Privy Council noted that:

The preservation of the impartiality and neutrality of civil servants has long been recognised in democratic societies as of importance in the preservation of public confidence in the conduct of public affairs…along with these elements of neutrality and impartiality [of the public service] their Lordships would associate an element of loyalty, in particular to the minister whom the civil servant has been appointed to serve. The importance of these characteristics lies in the necessity of preserving public confidence in the conduct of public affairs. That is at least one justification for some restraint on the freedom of civil servants to participate in political matters and is properly to be regarded as an important element in the proper performance of their functions. ⁵³

To some extent, different expectations on public servants can be difficult to reconcile. For example, it has been noted that it can be difficult to expect both that public servants will be independent and apolitical, on the one hand, but also responsive to the government of the day, on the other. ⁵⁴

The public service, through legislation such as the Public Service Act 1999 (Cth) and Codes of Conduct made under it, seeks to regulate and control the extent to which public servants interact with public debate, for example, by releasing information to which they are privy but of which the public might not be aware, or by expressing opinions about government policy or its implementation. The provisions are relatively general in their requirements. For instance, s 10 of the Public Service Act 1999 (Cth) refers to a public servant’s objectivity and apolitical nature. It does not refer specifically to loyalty. The precise content and meaning of these obligations is unclear. ⁵⁵ Perhaps for this reason, the public service has issued guidelines designed to assist employees. They state that an employee should not make public comment that does or might be perceived to compromise the employee’s ability to fulfil their duties in a professional and unbiased manner, or contains criticism so harsh that it calls into question the extent to which an employee might be able to carry out their tasks professionally and objectively, could disrupt the workforce, or is ‘unreasonably or harshly critical’ of others. Notwithstanding these attempts, a member of the High Court of Australia recently still characterised the line between what was acceptable speech by a public servant and what was not as ‘ill-defined’. ⁵⁶

These opinions are often particularly valuable, given they come from a person with greater than usual experience with their workings and implementation. They might have worked with particular individual politicians, for example, and can offer insights into the performance of that politician. As the Australian Law Reform Commission presciently noted a decade ago, it can be hard to reconcile openness and transparency in government, which most would laud as inherent in good government, with these kinds of secrecy provisions. ⁵⁷

B Critical Voices

The article will now consider a selection of critical voices reflecting on the value of freedom of speech within democratic societies. Each of them is a highly regarded expert in their field. The first is Paul Finn. Finn is a highly respected public law scholar who was appointed as a Federal Court judge and has made a distinguished contribution as a legal intellect in Australia over many years. Cass Sunstein is a high-profile American constitutional law scholar. Renowned economics scholar Joseph Stiglitz provides a valuable interdisciplinary perspective of the value of free speech through an economic lens. On occasion, the High Court has considered constitutional law issues through such a lens. ⁵⁸

C Comparative Case Law

D Fair Work Commission Decision Involving Public Servants, Codes of Conduct and Speech

The Fair Work Commission considered a factual scenario very similar to that in Banerji in the case Starr v Department of Human Services. ¹⁰⁶ Mr Starr was a longstanding Centrelink employee. His role included close contact with Centrelink’s ‘clients’. Mr Starr was terminated due to various social media posts he had made over a 3-year period. It was said that these posts breached the relevant Code of Conduct relating to Mr Starr as a public servant. That Code of Conduct was materially similar to that considered in Banerji, requiring public servants to behave in a way that upholds APS values. These values were stated to include respect for all people, openness and accountability, and impartiality. It was specifically stated that the APS was impartial. The Code required employees to work to uphold the reputation of the employee’s department and the APS generally. A related document entitled Conduct and Behaviour Policy referred to criticism of government. It stated that any such criticism must not lead the audience to suspect that the employee would be unable to undertake their duties impartially, professionally and/or efficiently, would be of such a nature as could disrupt the workplace, or could compromise the employee’s ability to perform their duties in an unbiased manner. All of Mr Starr’s posts were made outside of work hours. He did not make his identity known but answered ‘yes’ when someone online asked him if he worked at Centrelink.

The Commission found that it was not entirely clear on the evidence which of Mr Starr’s posts were being relied upon as amounting to a breach of the Code of Conduct. Notwithstanding this, the Commission found that three of Mr Starr’s posts could have amounted to valid reasons for dismissal, as amounting to breaches of the Code of Conduct. These were posts that used derogatory names in describing some of Centrelink’s clients, posts that suggested that the claims of many clients that they could not work due to depression were in many cases not genuine and posts that lamented the time it took to process some claims which made him ‘embarrassed’ to work where he did (in his words). In the case of the first two kinds of posts, these could be breaches of the Code of Conduct as evincing a lack of respect for others, and in the case of the third kind of post, the Commission found a possible breach in relation to failing to preserve the reputation of the relevant department.

It is also instructive to note those of Mr Starr’s posts that the Commission found could not be the subject of a finding that the Code of Conduct had been breached. Mr Starr posted comments highly critical of government policy, specifically in relation to proposed longer wait times to access welfare, the government’s paid parental leave scheme, and broader welfare policies. The Commission ‘reject(ed) entirely’ the suggestion that these posts amounted to a breach of the Code of Conduct. ¹⁰⁷ The Commission noted the common law right to freedom of expression, and the constitutional freedom of political communication. ¹⁰⁸ It concluded that the Code of Conduct did not ‘requir(e) that all APS employees be apolitical at all times outside of working hours’. ¹⁰⁹

The Commission accepted that some restrictions on the public expression of political views by public servants in their private time might be justified, consistent with the common law and the constitutional requirement. However, it suggested this would be more likely to be so in the case of senior public servants, for example, department secretaries, policy advisers, ministerial staff members of senior diplomats. It was much less likely in respect of the vast majority of public servants who performed administrative tasks. The Commission stated that, in respect of such employees, ‘it is difficult to envisage any circumstance in which the robust expression of political views and criticism of the government outside of work could have an impact on the performance of their duties’. ¹¹⁰ The Commission concluded there was no suggestion that Mr Starr had ever treated a client with a lack of respect, or acted other than impartially in relation to his decision-making functions at work. ¹¹¹ Nor was there any evidence that Mr Starr’s conduct had damaged the department’s reputation, or that anyone had drawn an adverse conclusion about the department’s impartiality. ¹¹² As a result, the Commission found that his dismissal was unfair. ¹¹³

Though the decision was by a single member of the Fair Work Commission, and was primarily focused on the fairness of the dismissal as opposed to the constitutional freedom, it is considered to contain important statements of principle in this area. Respectfully, there is considered to be much to be commended in the way the Tribunal in this case balanced the important, legitimate government interests around independence, impartiality and effectiveness of the public service with freedom of political communication in a representative democracy. These will be revisited during the critique of the High Court decision in Banerji.

Part IV now turns to a consideration of the High Court of Australia’s recent decision on the implied freedom in the context of public servants. Criticism of the decision will be informed by the above international cases, and scholarship reflecting the potential important role public servants can make to political discussion.

A Facts and Tribunal Decision

The case involved a relatively junior federal public servant who worked for the Department of Immigration. She had strong views about the Australian Government’s policies in this area, including questions about refugees, offshore processing and mandatory detention. She had views about Australia’s involvement in wars. Over some years, she had been anonymously tweeting about the Australian Government’s policies in this regard, as well as making negative comments about particular federal politicians. These tweets were generally critical. Importantly, the public servant was critical of politicians in both major political parties. She also made disparaging comments about American foreign policy and American leaders who prosecuted it. Her tweets numbered in the thousands. All but one had been made outside regular working hours.

The employee’s tweeting activity came to the attention of the Department, which commenced disciplinary proceedings against her. It believed her conduct might breach the Australian Public Service Code of Conduct (Code of Conduct). Section 13 of the Public Service Act 1999 (Cth) sets it out. It relevantly requires public service employees to act in a manner that upholds Australian public service values and the integrity and good reputation of the public service. Section 10 of that Act defined Australian public service values to include that the service is apolitical, performing its functions in an impartial and professional manner and delivering services fairly, effectively and impartially to the public. The Department also issued guidelines around the use of media, suggesting it was not appropriate for a departmental employee to make unofficial public comment that might be seen to compromise the employee’s ability to fulfil their responsibilities professionally and in an impartial and unbiased manner, or to criticise the government, member of Parliament or political party so harshly that it calls into question the ability of the employee to work professionally or impartially, or such that it might disrupt the workplace, or was unreasonably and harshly critical of department stakeholders. Section 15 of the Act set out various possible sanctions the head of the agency could impose if it were determined that an employee had breached the Code of Conduct. They included termination. The agency head terminated the employee’s employment due to her social media activity. The employee had unsuccessfully sought an injunction to prevent her dismissal. ¹¹⁴

The employee then ‘lodged an application’ for compensation for her dismissal, arguing she suffered anxiety and depression as a result. The Department argued in defence that it had taken reasonable administrative action in relation to the employee’s employment, given what it considered to be breaches of the Code of Conduct, thus the employee had not suffered an ‘injury’ under the relevant legislation. This indirectly raised the implied freedom of political communication. The employee argued that what the department did was not ‘reasonable administrative action’, having regard to the effect of its decision on the implied freedom of political communication.

The Administrative Appeals Tribunal found the management action here did infringe the employees’ implied freedom of political communication, and could not be justified. ¹¹⁵ It acknowledged it was legitimate for the government to be concerned about maintaining public confidence in the integrity of the public service, and the Code was tailored to that end. On the other hand, the Tribunal emphasised the tweets were anonymous. ¹¹⁶ This meant their content did not undermine the confidence that the public might have in the independence of their public servants. It could not do so because the public was not aware of the identity of the person tweeting, or that she was a federal public servant working in the policy area about which she tweeted. The Tribunal was heavily influenced by Canadian case law, which had sought to reconcile the legitimate interest in governments employing public servants who were loyal and able to implement government policy with the freedom of public servants to contribute to public debate. ¹¹⁷

B High Court Decision

On appeal, all members of the High Court overturned the decision. Kiefel CJ, Bell, Keane and Nettle JJ (‘the joint reasons’) stated the question whether a law (or measure) infringes the implied freedom of political communication must be answered in relation to the legislation (or measure) as a whole, rather than its application to particular individuals in particular cases. ¹¹⁸ The joint reasons commented on the fact the employee’s tweets were anonymous, concluding there were no special rules regarding anonymous communication. It would be a question of fact and degree in particular cases as to whether a particular anonymous communication infringed the Code of Conduct. The joint reasons found the fact the comments were anonymous was largely irrelevant; a person posting comments online could not assume their identity would remain private. ¹¹⁹ Even if the comments did remain anonymous, if they were cast in highly critical terms, as they were here, they could undermine perceptions of the integrity and good reputation of the public service, seriously disrupt the workplace and create questions around the employee’s ability to work professionally and impartially. ¹²⁰

In terms of the three elements, the court was satisfied the impugned measure burdened the implied freedom of political communication, as conceded by the government. The court noted the measures had a ‘material effect’ on the volume of political communication. ¹²¹ Regarding proportionality testing, the measures were designed to maintain an apolitical and professional public service. This was a legitimate objective, consistent with representative and responsible government. The public service was a constituent part of the system of representative and responsible government enshrined in the Constitution. The joint reasons noted an apolitical public service in the national interest had been a hallmark of public service in Australia. ¹²²

The joint reasons considered thirdly the question of proportionality testing, and whether the measures were suitable, necessary and adequate in their balance. They found a law would be suitable if it were rationally connected to its purpose. This was the case — it was essential to the proper functioning of the system of representative and responsible government that the government retain confidence in the public service to provide high quality, impartial and professional advice, and that the public service would faithfully and professionally implement accepted government policy, irrespective of particular employees’ personal views.

In determining whether or not a law was ‘necessary’, the joint reasons reiterated it was relevant to determine whether an obvious and compelling alternative existed to secure the government’s legitimate objectives in a way that was less invasive of the implied freedom. ¹²³ Here the joint reasons focused on an argument that the laws were not necessary because they extended to anonymous comments. Because the joint reasons rejected the suggested distinction between anonymous and non-anonymous comments, it did not accept an argument that the laws here were not ‘necessary’ because they extended to anonymous comments. ¹²⁴

In determining whether or not the law was ‘adequate in its balance’, again the analysis was somewhat attenuated, focused on the question of the possible penalty to be applied for breach of the Code of Conduct. The joint reasons reiterated that law would be regarded as adequate in its balance unless the benefit sought to be achieved was ‘manifestly outweighed’ by the adverse effect on the implied freedom. ¹²⁵ Applying this through the lens of the penalty, the reasons found the penalties not excessive, having regard to the importance of upholding public service values and its integrity and good reputation. ¹²⁶

Gageler J made extensive use of historical material, explaining the position of the public service in Australia. He explained its ethos throughout its history had been apolitical, focused on the national interest. He referred to the ‘somewhat idealised’ view of Professor Hearn regarding the Australian public service, and his views that public servants were focused on giving effect to the views of the political masters of the day, without regard to their personal view. For this reason, a public servant could not be dismissed solely because of their political views. He referred to arguments for maintaining restrictions on public servants’ political activity, including that if they were withdrawn, the public may no longer believe in an impartial public service, and ministers may lose confidence in departmental advice. He referred to the work of Professor Spann, who similarly spoke in favour of the neutrality of public servants, that they faithfully implemented policies of the federal government, without distortion based on personal views.

Gageler J agreed the impugned measures substantially burdened political communication. He noted they were also targeted directly at political communication, rather than something else. ¹²⁷ This meant a compelling justification for them was required. ¹²⁸ He noted Canadian case law similarly required such justification, and acknowledged the United States precedent had some limited relevance. ¹²⁹ He was satisfied the law passed the compelling justification requirement. It was imperative the government receive independent and impartial advice from public servants. It was necessary to achieve this end that requirements be imposed on public servants to maintain public service values, including during non-work time. ¹³⁰ Gageler J concluded:

the confidence of the government, the parliament and the Australian public in the APS as an apolitical and professional organization would be undermined without more were an APS employee free to engage with impunity in clandestine publication of praise for or criticism of a political policy of the government of the day or of a political party which might then or later be represented in the parliament. ¹³¹

Gordon J apparently disagreed with Gageler J that the law targeted political communication; rather, she said the provisions targeted conduct. She did not apparently reach a conclusion on whether the measures burdened political communication. Regardless, Gordon J found the measures were aimed at a legitimate end — maintenance of an apolitical public service of integrity and good reputation. Gordon J stated within a system of responsible government, public servants work for Ministers, who were accountable to Parliament. Public servants were required to advise upon and implement government decisions, regardless of the party or parties. Ministerial responsibility to Parliament necessarily entailed loyalty of public servants to Ministers, and their anonymity and neutrality. ¹³² It was essential to the constitutional system of government that public servants be apolitical. The measures were designed purely to uphold a fundamental aspect of responsible government. ¹³³

Edelman J noted some development in attitudes towards the speech of public servants over time. He noted the past absolute ban on public political communication by public servants had been tempered over time, though continuing restrictions still made public servants ‘lonely ghosts’. ¹³⁴ He acknowledged the impugned provisions imposed ‘deep and broad’ constraints on freedom of political communication. ¹³⁵ It was deep because it impacted those ‘uniquely qualified to comment’ on such matters. It was broad because it applied to conduct within and outside the workplace, at all times. They affected thousands of people. ¹³⁶ He said the public service values including being apolitical, impartial and professional served to create a trusted relationship between the public service and Parliament, the executive and the public.

He noted the values did not preclude a public servant making public comments, but there was a line crossed when the comments were such as to imperil trust between the public service and other stakeholders. He acknowledged the line could be ill-defined. Edelman J stated six factors were relevant in determining this. ¹³⁷ He noted the legislation was directed towards fundamentally important objectives of an impartial and apolitical public service. Edelman J cited Mason CJ in Australian Capital Television that freedom to criticise government action was an essential component of representative government. Edelman J stated while this may be so:

[i]t is also fundamentally important…that in a system of modern representative government a parliament has freedom to make laws that implement the policy decisions it makes for the welfare of the governed. Where a law impairs freedom of political communication in a reasonably necessary manner in pursuit of another legitimate object, the law should only be held unconstitutional if there is such a gross imbalance between, on the one hand, the importance of that legitimate object to the parliament and, on the other hand, the magnitude of the burden that the law places on the implied freedom of political communication, so as to pose a threat to the integrity of the constitutionally prescribed system of representative and responsible government. ¹³⁸

C Criticisms of the High Court’s Decision

Before critiquing aspects of the High Court’s decision, one important point should first be made. This is to recognise the legitimacy of a government interest in suppressing the speech of some public sector employees. Governments are entitled to rely on the independence of the advice they receive from public servants and are entitled to expect that public servants will faithfully implement government policy, regardless of the employee’s personal views. A government could not function effectively if its employees actively sought to and/or were able to thwart the implementation of government policy, leaked confidential information such as to damage Australia’s interests or actively undermined government processes. Public sector employers must have confidence in the impartial service of their employees.

Thus, it is conceded to be possible that an employee’s criticism of government is such that the employee could not reasonably be expected to perform their functions, including implementing government policy. This article takes the position that a public servant’s freedom of political communication is by no means absolute. Restrictions on such speech may well be consistent with the Australian Constitution.

However, such restrictions would need to be carefully tailored to meet the legitimate government interests involved, whilst taking due heed of the possibly valuable contribution that a public servant could make to political discussion and debate in Australia. Limitations on speech would require careful justification, with supporting evidence, not based on simple assumption or assertion. This careful balancing of potentially competing interests is a feature of the case law in this area in other countries, as explained above.

Given that this section is necessarily of some length, I summarise here the main criticisms I explain further below:

I now explore these criticisms in more detail.

The first criticism of Banerji is the lack of recognition of the specific, valuable role that public servants play in facilitating the kind of open and accountable government that democracy requires in order to function effectively. All of the judgments recognise the potential for the Code of Conduct provisions to impinge on the freedom of individuals to communicate about political matters. These discussions may imply some recognition of the important role that public servants can play in political discussions. However, there is little express recognition of the particularly valuable role that public servants can play in this space, given their knowledge of the internal workings of government.

That having been said, there is limited recognition of this potentially important role. I refer in particular to the judgment of Edelman J. His Honour acknowledges that the challenged provisions burden the implied freedom deeply and broadly. In elaborating upon this point, Edelman J noted that the burden is exacerbated by the fact it targets individuals who are ‘uniquely qualified to comment’ on political matters. ¹³⁹ He takes this quote from the decision of the United States Supreme Court in City of San Diego v Roe. ¹⁴⁰ There the Court stated that a public servant did not relinquish their First Amendment rights simply by becoming a public servant. ¹⁴¹ It recognised that government had special interests, such as would permit restrictions on the speech of public servants that would not be permitted in different contexts, it recognised ‘the right of (public sector) employees to speak on matters of public concern, typically matters concerning government policies that are of interest to the public at large, a subject on which public sector employees are uniquely qualified to comment’. Notwithstanding that the First Amendment ‘right’ is broader and in the nature of a personal right, as opposed to the Australian negative ‘freedom’, it is submitted that the recognition of the particular, special value of the speech of public servants on political matters is important.

There is much less recognition of this particular, special role of public servants and their communication about political matters in the other judgments. It is true that all other judgments acknowledge that the Code of Conduct measures do burden freedom of political communication, and some acknowledge that the burden is substantial. However, in contrast with the judgment of Edelman J, the other judgments do not, expressly at least, reflect on the particular, special role of public servants in political communication, given the specialised knowledge they may have. My own view is that this was an important part of the case and should have been expressly recognised.

It is possible that the High Court may not have emphasised the important contribution that public servants might make to public debate in this particular case because the facts of the particular communication at issue in the case did not appear to reflect a deep-seated, philosophical debate. Obviously, the nature of social media often tends to actually preclude this from occurring. On the other hand, the facts raise in my view an important principle — the potential value of public servants’ speech to public policy debate and discussion. The judgments might have reflected on this potential value in terms of principle, rather than (possibly) be influenced by the nature of the particular tweets here. It is understood that the judges may perceive their role as quelling the controversy before them, rather than consider the bigger picture principles in play. This is admittedly conjecture on my part, however, it is possible that if the facts had involved a well-thought out, deep level contribution to a public debate by a public servant, the High Court may have taken more time to acknowledge the value of a public servant’s contribution to political debate and discussion. On the other hand, of course the High Court enters difficult territory by measuring the constitutional protection to be given to discussion based on its perceived quality, just as it does when it does so by reference to its perceived tone (discussed further below). Thus, this paragraph seeks to explain what might have constrained the framing of the court’s judgment, not to argue that it ought to have.

The High Court’s approach to the implied freedom contemplates a balance between competing interests, as many legal principles do. However, in order to conduct this balancing exercise fairly, it is important to identify what the competing interests and values are. It is in this light that the omission from most judgments of specific recognition of the important contribution that a public servant might make to public debate on contentious issues assumes particular importance. In contrast, decisions in other jurisdictions discussed above, such as Pickering in the United States and Guja in Europe, recognise expressly the importance of scrutiny of government actions and the important role public servants play in this regard.

The second criticism is of the High Court’s application of its ‘necessity’ test in the proportionality testing of a measure challenged on the basis of the implied freedom. Here the discussion was confined to rejecting a suggestion the measure failed the ‘necessity’ test because it did not distinguish anonymous and non-anonymous tweets. For this reason, the court concluded the measure was ‘necessary’. To be fair, it is possible that the High Court focused on the question of anonymity because that principle was pivotal in the decision reached below. ¹⁴² In a sense, the court was evidently reacting to the decision from which the appeal was taken.

However, there is more to discuss here. The court said a measure would only lack ‘necessity’ where there was an obvious and compelling alternative equally practicable and available to achieve the government’s purpose that would result in a significantly lesser burden on the implied freedom. ¹⁴³ Was there such an obvious and compelling alternative? The government’s objective, which the court accepted was legitimate, was that the government have confidence in the ability of public servants to give high quality, impartial, professional advice, and would faithfully implement government policy irrespective of the employee’s political beliefs.

I must respectfully disagree that there was no obvious and compelling alternative to achieving the government’s legitimate ends. The Code of Conduct and/or legislation could be written in more precise terms to focus more clearly on the government’s legitimate objectives. It might say, for instance, that an employee breaches the code of conduct when they engage in behaviour such that a reasonable outside observer might have significant doubt about the ability of the officer to give quality, impartial, professional advice, or that a reasonable outside observer might have significant doubt about the ability of such an officer to faithfully implement government policy. This is somewhat analogous to one of the factors articulated by Edelman J as being relevant in determining whether or not a limitation on the speech of a public servant might be justified. ¹⁴⁴ It is somewhat similar to the position in the United States, where courts might accept restrictions on public servants’ speech to the extent that the speech at issue is reasonably likely to impair their ability to do that which they were employed to do. A test such as ‘reasonable outside observer’ is desirable, because it places the evidentiary bar at a higher level than would be met by a mere government assertion that the comments should be punished because they will damage confidence in the independence of the public service, or that ‘it might be perceived’ that the person could not implement government policy. It imposes some requirement of objectivity, as opposed to mere assertion.

In addition or in the alternative, it might have been confined to more senior public servants, or those whose activity could significantly impact on the workings of government, as the Canadian Supreme Court did in Osborne. Some members of the High Court expressly referred to the possible importance of the seniority and position of the relevant public servant as being relevant to an assessment of whether particular communications might be effectively protected as manifestations of an implied freedom to communicate about political matters. ¹⁴⁵ This matter will be discussed presently in the context of compelling justification.

To be clear, it is not suggested that the High Court should have rewritten the Code of Conduct, or that the government should have rewritten it once the Banerji proceedings had commenced. The point is that there are legitimate interests surrounding public sector employee behaviour, such that the need for some restrictions on the speech of public servants is accepted. The problem with the version considered in Banerji, in the author’s view, is its breadth and depth. The government, not the courts, should be invited to rewrite the Code of Conduct in more limited terms. A revised version could well be compatible with how this article believes the implied freedom should be applied to this context.

The government’s clear difficulty with what Ms Banerji did was that it was embarrassing to them. It was critical of their policies. However, of course public servants will have a range of views on many matters. At times, they might be in strong disagreement with the policies of the government of the day. It is submitted that most outsiders would expect that a public servant would be professional enough to separate their own personal views with that of the policy they are being asked to implement. That is an aspect of their professionalism. Most of us, from time to time, are required to implement policy with which we personally disagree. Most of us do so quietly. Respectfully, the High Court was too quick to reach a conclusion based on an assumption that Ms Banerji would not be able to fulfil her duties, or at least agree that a perception would arise that she would not be able to fulfil her duties, because of her strong political views. The European Court cautioned in Spycatcher that limitations on the speech of public servants, or ex-public servants, had to be thoroughly justified in order to be legally acceptable. The Supreme Court of Canada in Fraser also indicated evidence would usually be required of actual impairment, to justify a restriction on the speech of public servants. This high bar was not met in the Banerji case, in my respectful view.

Of course, if there was in fact evidence that Ms Banerji, because of her strongly held views, was not able to fulfil her duties — for instance, she had refused reasonable directions to implement government policy, or she had sought to actively thwart implementation of the government’s policies, then there would be little concern with her dismissal. Demonstrably in such a case, her views would have precluded her from doing that which she was employed to do. Recall that this was an important consideration in the decision of the Fair Work Commission in Starr. ¹⁴⁶ Vice-President Hatcher noted that there was no evidence, notwithstanding Mr Starr’s criticisms of his employer’s ‘clients’, that he showed a lack of respect for clients in his dealings with them, or that he failed to act impartially or appropriately in the exercise of his decision-making powers. ¹⁴⁷ Further, there was no evidence that Mr Starr’s criticisms of government welfare policy ‘could possibly compromise his capacity to carry out his front-line Centrelink functions’. ¹⁴⁸ In other words, there was no evidence that the employee’s personal views in Starr compromised his ability to do the job that he was employed to do, or was reasonably likely to do so. The same observation could be made with respect to Ms Banerji. Similarly, if there were evidence that her conduct had actually undermined public confidence in government, again the government would be justified in taking the course of action it did.

However, in relation to balancing the fundamental notion of freedom of speech in a democracy versus government concerns about impartial and neutral public servants, evidence would (should) be necessary that a particular employee could not in fact fulfil their functions or were actually damaging public confidence (or, at the very least, that a reasonable outside observer would so conclude), in order to conclude that an action to dismiss them because of their social media tweeting was a response that was ‘necessary’. The Canadian Supreme Court emphasised this in Fraser. A court should not too readily assume a conclusion in this regard. There is no suggestion in the evidence that her (strongly held views) precluded her from carrying out her professional responsibilities. Employees sometimes work for employers who make decisions, including policy decisions, with which the employee disagrees. The employee might choose to leave the organisation or remain within it. But mere disagreement, even strong, with an employer’s policies should not lead to an apparent assumption they are unable to implement government policy, or that public confidence had been damaged.

Again, the High Court’s discussion of the concept of ‘adequate in its balance’ was highly attenuated. It has been noted above that the decision, apart from one line in the judgment of Edelman J, did not note the special value that a public servant’s contribution to public debate might have, given their expertise. The only lens through which the joint reasons considered ‘adequate in its balance’ was the very narrow one of context. The court repeated that under this principle, the court had to consider the benefit sought to be achieved by the law as compared with the adverse effect on the implied freedom. ¹⁴⁹ Respectfully, the court did not consider the adverse effect on the implied freedom. It merely focused on the extent of penalty.

It is submitted that a fuller discussion of the question of ‘adequate in its balance’ was desirable. Respectfully, it is difficult to discuss whether measures are balanced when the effect of the measure on the speech of those affected is not discussed. It is also difficult to do so when the possible value of that speech is not discussed. A fuller discussion would have considered more issues, as discussed in some of the comparative case law considered above. Specifically, it might have considered questions like whether the material released was confidential in nature, whether its release would likely damage government interests and/or national security, the seniority of the public servant involved, whether the subject matter was of significant public interest, whether its disclosure would likely affect the ability of the employee to continue to function in their position and/or work effectively with others and what motivation the employee had for expressing their views. These suggestions are informed by my consideration of the case law elsewhere, discussed above, as well as the writings of the critics discussed there. It might focus on whether the information released would lead a reasonable outside observer to conclude the employee could not continue to perform their functions in the impartial and professional way expected in the public service, or work to implement government policy. It might have considered the chilling effect of these kinds of regulations. ¹⁵⁰

A further criticism is that, as Gageler J noted, these restrictions on public servant speech were content-based, or in other words, restrictions that directly targeted political communication. ¹⁵¹ As a result, as he again correctly noted, the orthodox High Court position is that such measures are subject to a test of ‘compelling justification’. ¹⁵² Gageler J concluded the impugned measures passed such a test. In contrast, Gordon J disagreed that the measures targeted political communication. The others did not discuss the matter. With respect, Gageler J was correct, and members of the Court, recognising that the measure directly impacted political communication, should have considered whether there was a compelling justification for the measure. My own view is that while there is a legitimate government interest in terms of the impartiality, neutrality and professionalism of the public service, there are very significant implications for democracy if a large percentage of the population is to be effectively silenced from debate. ¹⁵³

I do not agree that there is a compelling justification for the measures, cast as they are in such broad terms. More narrowly targeted measures, around the release of confidential information, acting in ways likely to seriously damage Australia’s security and defence, or behaviour that is likely to seriously raise concerns about the ability of an employee to implement government policy, might meet a requirement of compelling justification. A rule applicable to very senior public servants, whose political activity could seriously undermine public confidence in the workings of government, or whose activity could realistically derail government policy implementation, might be defensible. All of these suggestions reflect that whilst public servant freedom of communication is not absolute, and restrictions might be legally permissible, these restrictions must be narrowly framed and thoroughly justified. Mere assertion is not sufficient.

Again, support for this appears in the Starr decision, where Vice-President Hatcher distinguished between a restriction on a departmental secretary, ministerial staff member, policy adviser or senior diplomat and a restriction on most public service employees. Vice-President Hatcher stated that while restrictions on the speech of the former might be defensible, restrictions on the latter would not. This was because in the case of the latter category, it was practically extremely unlikely that robust communication of the views of such employees could impact on the performance of their duties. ¹⁵⁴ The Administrative Appeals Tribunal referred to Ms Banerji’s position as one of ‘relatively low rank’, ¹⁵⁵ a finding not challenged on appeal. This replicates the approach taken by the Canadian Supreme Court in Osborne.

Thus, there are strong analogies between the factual scenario in Starr and that in Banerji. Both involved relatively junior employees espousing views critical of government policy. As junior employees, their actual ability to thwart the implementation of government policy, or actually reduce public confidence in the government, was limited. In both cases, there was no finding that the employee’s actions or beliefs actually precluded their being able to do their job that they were employed to do. In these circumstances, it becomes difficult to justify the interference with the public servant’s freedom to communicate.

Some will argue that the cases are quite distinguishable. It might be argued that in the case of Starr, he was primarily a customer service worker with little influence over the formulation or implementation of government policy. It might be argued that, in contrast, Banerji was different. It is somewhat frustrating that none of the court documents I have observed in relation to this case go into much detail about the precise nature of Ms Banerji’s duties. However, it can safely be stated that she was a policy adviser, of relatively junior rank. Arguably, this distinguishes her case from Starr, because the possibility that her strongly held views could practically colour her advice to government is a stronger, more important factor than was the case in Starr. It is difficult to advance this argument much in the absence of more detailed information about the nature of Ms Banerji’s role. However, of most importance is the lack of any actual finding that Ms Banerji’s strongly held views did in fact preclude her from providing independent and objective advice to government. This seems to have been assumed. My own view is that, given the very serious restrictions on political communication wrought by the Code of Conduct, the government would need more evidence that this had actually occurred, as opposed to a possibility or assumption. Again, I am fortified in this conclusion by the findings in Starr:

Although some of the online comments made by Mr Starr might theoretically raise a perception that there was an employee in the Department who might not take decisions concerning claimants for benefits impartially, the actual position was that there was no evidence to suggest that Mr Starr was anything other than completely impartial when he carried out his decision-making functions…similarly, while some of the comments conveyed a lack of respect to some benefits claimants, the evidence was that when at work Mr Starr always treated the Department’s customers appropriately and with respect. There was therefore never any basis to form the conclusion that Mr Starr’s comments online ever had reflected or ever would reflect his actual work performance…and there was no evidence that the Department’s perception of the Department’s impartiality. ¹⁵⁶

While these comments concern a front-line customer service worker, as opposed to a policy adviser, the point about lack of evidence of actual impairment is readily transferable to the Banerji context and should have been applied in that case, in my view.

It is acknowledged that, in the Starr decision, the tribunal itself distinguished between the position of Mr Starr and that of other employees. The tribunal included within this category of other employees, departmental secretaries, ministerial advisers, policy advisers and diplomats. Ms Banerji was indeed a ‘policy adviser’. In isolation, it seems as if the tribunal might have taken a different view on the question of Ms Banerji’s freedom of political communication than that of Mr Starr, given the different roles. More broadly, however, the tribunal in Starr was differentiating a category of employees where the expression of views publicly would or could impact on their ability to do their work. It might be readily seen how a departmental secretary’s strongly worded criticism of government policy which they are charged with implementing might affect their ability to do their work. Similarly, it might be readily seen how a diplomat might have difficulties where they had expressed strongly held views contrary to government policy. These roles are considered to have a level of seniority where concerns about a public sector employee’s strongly expressed dissent from government decision-making and policy have more legitimacy. It is submitted that, at lower levels of the public service, including policy adviser level, these concerns are less real.

There is some analogy here with principles developed elsewhere in constitutional law. It must first be acknowledged that these principles operate in a different context to the one being presently considered. The context of these principles was an attempt by the Commonwealth to have its award applied to state government employees pursuant to s 51(35) of the Constitution. That is clearly a different context than that of the implied freedom of political communication. That said, in that case the High Court considered an argument that a constitutional limitation should prevent the Commonwealth from regulating the employment of state government employees, on the basis that it would impair the ability of state government to conduct their functions. ¹⁵⁷ A majority of the High Court in Re Australian Education Union; Ex parte Victoria ¹⁵⁸ accepted only a very limited version of this argument. The joint reasons stated:

Also critical to a state’s ability to function as a government is its ability not only to determine the number and identity of those whom it wishes to engage at the higher levels of government, but also to determine the terms and conditions on which those persons shall be engaged. Hence, Ministerial assistants and advisers, heads of department and high-level statutory office holders, parliamentary officers and judges would clearly fall within this group. The implied limitation would protect the states from the exercise by the (federal industrial relations) commission of power to fix minimum wages and working conditions in respect of such persons and possibly others. ¹⁵⁹

Although the context is different, a test of whether given measures ‘impair a state’s ability to function as a government’ might bear some comparison to a test of whether the speech of a public servant could damage or undermine public confidence in the public service, or practically impede it in the faithful implementation of government policy. Just as the High Court sensibly distinguished the practical ability of senior public servants to do the former, in comparison with the vast majority of more junior public servants, it might do so in the context of the implied freedom of political communication, rather than apparently treat the fact of the public servant’s seniority or otherwise as irrelevant (excepting the judgments of Gageler and Edelman JJ). ¹⁶⁰

The relevant provisions of the Code of Conduct in this case are too sweeping. They apply to hundreds of thousands of public servants, where the political opinions and activities of the vast majority would be most unlikely to realistically thwart government policy development or implementation, or undermine public confidence in the public service. This is an illustration of the value of the American First Amendment doctrine of ‘overbreadth’, though the Australian High Court has (to date) declined to apply that doctrine; indeed, some hostility to American First Amendment jurisprudence has become evident. ¹⁶¹ It should in any event be relevant to any ‘adequate in the balance’ analysis. A law that is overbroad, in that it interferes ‘too much’ with freedom of political communication in a way that is not necessary in order to achieve the legitimate objective(s) it seeks to achieve, will not be adequate in its balance, having regard to the fundamental importance of freedom of political communication in a democracy.

In conclusion, the government has legitimate interests in regulating, to some extent, political communication engaged in by public servants. Public servants do not have untrammelled freedom to say whatever they wish. However, the government’s legitimate interests in terms of an independent, impartial and functional public service must be tempered by acknowledgement of the very valuable contribution that public servants can make to important public policy debates, from their specialised knowledge and expertise. Factors such as the seniority of the relevant employee are important, as is the actual practical likelihood that the employee’s views will or may hamper formulation of government policy, its implementation, or public confidence in the government, or evidence that this has actually occurred. Government claims about the potentiality of this ought not to be too readily accepted, in the absence of actual evidence. In earlier implied freedom cases, the Court has spoken of the scepticism with which claims of the need for regulation of political communication should be viewed. ¹⁶² Sunstein noted claims of the need for governments to censor views were often exaggerated. These sentiments must remain paramount in our thinking.

D Comment about the Relevance of Tone

There is an apparent trend in implied freedom case law towards taking into account the tone of comments in determining whether regulation of them is consistent with the implied freedom. Evidence appears in the judgment of the Full Federal Court in Chief of the Defence Force v Gaynor ¹⁶³ where Perram, Mortimer and Gleeson JJ overturned a finding of a single judge of the Federal Court that dismissal of an army reserve officer for offensive posts on social media was contrary to the implied freedom. The Full Court noted that

It is not so much the subject matter of the communication…which is likely to trigger the exercise of power (to dismiss or discipline the outspoken employee) but rather the tone and attributes of the communication…the risk of harm to the freedom is commensurately lower. ¹⁶⁴

Likewise, Edelman J in Comcare v Banerji noted that in determining whether a public servant’s speech had ‘crossed the boundary’ between acceptable and unacceptable expression of political opinions, one of six relevant factors was ‘the location of the content of the communication upon a spectrum that ranges from vitriolic criticism to objective and informative policy discussion’. ¹⁶⁵

In contrast with these suggestions that the way in which a view is expressed determines whether or not it enjoys constitutional protection is the classic statement of Kirby J in Coleman v Power that

[o]ne might wish for more rationality, less superficiality, diminished invective and increased logic and persuasion in political discourse. But those of that view must find another homeland. From its earliest history, Australian politics has regularly included insult and emotion, calumny and invective, in its armoury of persuasion. They are part and parcel of the struggle of ideas…by protecting from legislative burdens governmental and political communications in Australia, the Constitution addresses the nation’s representative government as it is practised. It does not protect only the whispered civilities of intellectual discourse. ¹⁶⁶

Several judgments in Levy v Victoria express similar sentiments. ¹⁶⁷ John Stuart Mill also rejected any suggestion the ‘free expression of all opinions should be permitted, on condition that the manner be temperate, and do not pass the bounds of fair discussion’. He said (correctly) it was impossible, even if it were desirable, to properly draw such boundaries, it would tend to be applied to the expression of (currently) unorthodox views rather than orthodox ones. He said law obviously had ‘no business’ in restraining vituperative language. ¹⁶⁸

Respectfully, Mill and Justice Kirby are correct. The form in which ideas or thoughts are framed should not determine the extent to which they effectively receive legal protection.

E Argument from Discrimination Law

An employee dismissed from their employment for expressing (unwelcome) political views might also have a claim based on the law of discrimination. ¹⁶⁹ The Fair Work Act 2009 (Cth) and state anti-discrimination legislation protect employees from discrimination. It is true that, in respect of a federal employee, their employment is regulated by the Public Service Act 1999 (Cth). Section 29 of that Act gives power to department heads to dismiss employees, including for alleged breaches of the code of conduct. However, s 8(1) of the Public Service Act 1999 (Cth) specifically states that the provisions of that legislation take effect subject to the Fair Work Act 2009 (Cth). ¹⁷⁰ It is clear from the terms of the Fair Work Act 2009 (Cth) that it is intended to bind the federal government. Section 11 states that the terms ‘employer’ and ‘employee’ are to be given their ordinary meanings. Section 14 specifically states that the Commonwealth is to be included within the definition of the phrase ‘national system employer’. This is significant because that Act applies, among others, to national system employers.

Having shown that the Fair Work Act 2009 (Cth) applies to the federal government as employer, its provisions regarding discrimination and unlawful termination are of interest. Section 351(1) prevents discrimination against an employee on the basis, among others, of an employee’s ‘political opinion’. Section 772(1)(f) states that it is an unlawful termination if an employee is dismissed for the reason, or reasons including the reason, of the employee’s political opinion. Thus, it is quite possible for an employee who was dismissed because of their social media activity, which was clearly informed and motivated by their political belief, to argue that they were discriminated against, and unlawfully terminated, because of their political belief. Of course, the employer will argue that they were not terminated because of their political belief but because the employee breached a relevant code of conduct. The employee will respond that, in substance, because the alleged breach of the code of conduct related to their expression of political belief and view, they were in effect dismissed/discriminated against on that basis. It is beyond the scope of this article to consider this matter in further detail, but there are a range of cases that might provide support to both sides of this argument. ¹⁷¹

It should also be noted that even if there is a technical breach of either or both of s 351(1) and s 772(1), in both cases the prima facie prohibition is subject to an exception relating to ‘genuine occupational requirements’. It might be argued that in relation to the public service, it is a genuine occupational requirement that employees be, and be seen to be, independent, neutral and impartial and loyal to the government. This receives limited support from the Public Service Act 1999 (Cth). Section 10 sets out APS values to which all employees are expected to adhere. It does mention that the public service be apolitical. The government might thus argue that even if it technically might have been in breach of discrimination and unlawful termination provisions of the Fair Work Act 2009 (Cth), its decisions related to a genuine occupational requirement, that of the apolitical nature of the public service.

On the other hand, where an employee does not identify herself as a member of the public service, this, coupled with her anonymity, might suggest that the apolitical nature of the public service was not in jeopardy in the present case. There was never any suggestion in the case that the relevant public servant at issue was demonstrably unable or unwilling to implement the federal government’s relevant policies. If there were, there might have been a stronger argument that she was not able to meet an inherent requirement of her position — to implement the policies of the government of the day. The government’s argument might instead be that the perception arose that the public servant would be unable to implement the government’s policies, because of her clear disagreement with them. However, the argument around perception struggles with the fact that she did not identify herself as a public servant, let alone a public servant working in the area about which she tweeted.

It might also be argued that it is a ‘genuine occupational requirement’ that governments be able to trust and have confidence in their employees, such that where there is evidence that the public service employee has personal views at odds with the philosophy of the government, the dismissal of the employee is valid. ¹⁷² On the other hand, the High Court has found, in the context of the employment relationship generally, that no implied obligation of mutual trust and confidence arises. ¹⁷³ In essence, Australian law risks incoherence where, on the one hand, it protects employees (including public servants) from dismissal because of their political views and activities, and encourages them to speak out about possible wrongdoing within their areas (the whistleblower provisions discussed above), but then in the same breadth, indicates to public servant employees that if they dare to speak their personal opinions about political actors, policies or decisions, they risk dismissal.