Abstract
Foreign interference is a growing threat to all liberal democracies, including Australia. To respond to this growing threat, the Department of Home Affairs has developed a complex ‘Counter-Foreign Interference Strategy’ (CFIS). At the heart of the strategy lies a suite of interlocking and overlapping legislation, including the Foreign Influence Transparency Scheme Act 2018 (FITS Act), the National Security Legislation Amendment (Espionage and Foreign Interference) Act 2018 and the Electoral Legislation Amendment (Electoral Funding and Disclosure Reform) Act 2018 (Electoral Funding Act). The aim of this paper is to explain and clarify the legislation and the free speech burdens it imposes, and determine whether the laws are suitably targeted at foreign interference without unduly limiting legitimate communication activity. We argue that the current criminal law regime is ineffective in addressing the problem because foreign interference is a complex and pervasive phenomenon taking many different forms — from espionage on university campuses to anonymous and targeted social media campaigns. The legislative scheme is not properly tailored to tackle foreign interference as it actually occurs.
I Introduction
Foreign interference is a growing threat to all liberal democracies, including Australia. The term refers to coercive, corrupt, deceptive or clandestine activities by or on behalf of a foreign actor, intended to undermine Australia’s national security, democratic processes and sovereignty. 1 In contrast, foreign influence is defined as routine diplomacy and open and transparent discussions which are part and parcel of engagement with international partners. 2
Foreign interference is a multifarious phenomenon. There are multiple vectors for foreign interference including cyber-attacks, social media, traditional media and electoral campaign funding. As the Department of Home Affairs noted in its submission to the 2021 Select Committee on Foreign Interference through Social Media, foreign interference risks often take the form of cultivating and manipulating people to influence their decision-making, including through personal, political, business and diplomatic relations; limiting freedom of expression and shaping the media and communications landscape to spread misinformation; dominating foreign language media or undermining public discourse about issues of national significance; and singling out sections of the community through pressure and manipulation to sow discord, silence dissent or damage social cohesion. 3 Foreign interference in Australia has involved foreign actors, including foreign intelligence services, trying to interfere with Australian decision makers at all levels of government and across a range of sectors, including democratic institutions, education and research organisations, media and communications companies, culturally and linguistically diverse communities, and critical infrastructure. 4
To respond to this growing threat, the Department of Home Affairs developed a complex ‘Counter-Foreign Interference Strategy’ (‘CFIS’). The strategy engages with vulnerable sectors (such as tertiary institutions); seeks to deter perpetrators; defends against foreign interference through a coordinated government response; and enforces CFIS laws by investigating and prosecuting breaches. 5 At the heart of the strategy lies a suite of interlocking and overlapping legislation, including the Foreign Influence Transparency Scheme Act 2018 (Cth) (‘FITS Act’), the National Security Legislation Amendment (Espionage and Foreign Interference) Act 2018 (Cth) (‘Espionage and Foreign Interference Act’) and the Electoral Legislation Amendment (Electoral Funding and Disclosure Reform) Act 2018 (Cth) (‘Electoral Funding Act’). 6 In 2020, the Commonwealth Parliament passed a further piece of legislation, the Australia’s Foreign Relations (State and Territory Arrangements) Act 2020 (Cth) (‘Foreign Relations Act’), 7 designed to protect and manage Australia’s foreign relations more broadly, including arrangements entered into by Australian universities. Those laws are scaffolded by various governance and risk frameworks, and guidelines for Australian universities. In 2019, a University Foreign Interference Taskforce (‘UFIT’) was also established. 8 The complicated legislative response has been criticised for being badly targeted, poorly designed and imposing unjustified burdens on free speech especially for groups such as academics, non-government organisations (‘NGOs’) and journalists.
This article is the result of an interdisciplinary collaboration between legal scholars and a journalist with prolific expertise and experience in media. Our aim is to explain and clarify the legislation and the free speech burdens it potentially imposes, to determine whether the laws are suitably targeted at preventing and deterring foreign interference without unduly limiting legitimate communication activity. Freedom of speech is a central pillar of democratic societies. It is how we hold our governments to account, communicate about and decide on complex policy issues, make informed voting decisions, disseminate knowledge and realise our autonomy as individuals. Foreign interference undermines this democratic function of speech by covertly distorting the deliberative landscape, inhibiting informed decision-making through the dissemination of dis- and mis-information and, eventually, undermining democratic institutions. It is therefore crucial that legislation designed to deter foreign interference and protect democracy and representative government does not have the perverse consequence of undermining democracy, while failing to deter foreign interference. In this paper, we suggest that Australia’s foreign interference legislative package is poorly targeted and risks inadvertently harming the democracy it is designed to protect, by chilling the speech of important civil society actors such as journalists, academics and think-tanks.
We defend three main claims. First, we argue that the legislative requirements impose an unjustified and disproportionate burden on political communication, which is constitutionally protected by the implied freedom of political communication. Second, we argue that the legislation is poorly tailored to address foreign interference where it occurs most, namely, on social media. Third, we suggest that the legislation may have perverse and unintended consequences: by chilling the speech of important civil society actors, the scheme inadvertently undermines the democratic institutions and discourse it is designed to protect; and by creating multiple and overlapping regimes of reporting, the legislation forces institutions to create comprehensive (and arguably unnecessary) online databases of foreign engagement likely to be an enticing target for malicious hackers. Recent Optus and Medibank hacks demonstrate the real risk in keeping this sort of data. In short, the scheme may increase rather than diminish Australia’s vulnerability to foreign interference, while chilling the robustness of our political engagement and communication to the detriment of our political system.
In part 1, we provide overviews of the FITS Act, the Espionage and Foreign Interference Act, the Electoral Funding Act and the more recent Foreign Relations Act, to demonstrate the potential impact on the implied freedom and the free speech of critical civil society actors such as academics, journalists and NGOs. We also draw on theoretical accounts of the importance of academic freedom and public interest journalism to democratic deliberation to demonstrate the high stakes involved in this legislative package. In part 2, we unpack the nature of foreign interference in Australia and the various forms it takes through social media, cyber-attacks, surveillance of students on university campuses and other covert activities intended to manipulate the deliberative landscape. We demonstrate how some aspects of the legislative framework are poorly targeted and may not be fit for purpose. Some aspects of the scheme are duplicative, ineffective and redundant, while important methods of foreign interference are not addressed by the legislation at all. For example, social media are primary vehicles for foreign interference, but the legislation imposes no reporting requirements on social media platforms, even though social media companies have mechanisms to identify when foreign interference is occurring. 9 This is a significant legislative gap.
In the final part, we make recommendations for reform. We propose greater free speech protections for civil society actors and suggest the registration requirements under the FITS Act should be scrapped because they unduly burden the implied freedom and are not fit for purpose.
A The Legislative Framework
In 2018, the then Liberal government implemented the CFIS. This included a package of ‘interlocking components’, 10 each intended to address the risk of foreign interference. 11 The legislation included the FITS Act, Espionage and Foreign Interference Act and Electoral Funding Act. Submissions from academics, journalists and lawyers argued that the laws would stifle free speech, deter public interest journalism and inhibit the advocacy of domestic NGOs. Constitutional law scholar Anne Twomey said the legislation was so ‘sweeping and poorly drafted it would force thousands of people — including academics, authors and book publishers — to resister as agents of other countries’. 12 She argued that the Bill would impose a significant and disproportionate burden on the implied freedom of political communication. 13 Journalist Lenore Taylor said while some improvements had been made through iterations of the National Security Legislation Amendment (Espionage and Foreign Interference) Bill 2018 (Cth), it still broadened the circumstances in which journalism can become a crime. 14 Politically diverse groups such as the right-leaning Institute of Public Affairs joined the Greens, charities and activist group GetUp! to criticise the donation laws, which some said amounted to an ‘unprecedented and dangerous’ attack on democratic debate. 15 A common complaint was that the legislation was incomprehensibly complex, and that this in itself would chill free speech. In the following sections, we briefly explain the legislation enacted in 2018 and 2020, to unpack what each Act involves, how they interact and their potential impact on free speech.
1 Foreign Influence Transparency Scheme
The objective of the FITS Act is stated narrowly in s 3, to ‘provide for a scheme for the registration of persons who undertake certain activities on behalf of foreign governments and other foreign principals, in order to improve the transparency of their activities on behalf of those foreign principals’. 16 The Act defines ‘foreign principal’ as either a foreign government, a foreign government-related entity, a foreign political organisation or foreign government-related individuals. 17 It requires individuals or organisations 18 to register details about themselves and their foreign principal with the Secretary, if the person communicates to the Australian public under an arrangement with a foreign principal. 19
‘Communications activity’ is broadly defined in s 13 as circumstances where a person ‘communicates or distributes information or material to the public or a section of the public’ or ‘produces information or material for the purpose of the information or material being communicated or distributed to the public or a section of the public’. Information can take any form, and influence is broadly conceived as to ‘affect in any way’, 20 presumably meaning to ‘affect’ a person or their views, although the precise meaning of the term remains unclear. Part 2 div 3 of the FITS Act sets out what constitutes ‘registrable activities’ as including parliamentary lobbying on behalf of foreign governments 21 and activities in Australia for the purposes of political or governmental influence, such as communications activities. 22 This Division also includes former Cabinet Ministers 23 and recent designated position holders who may have specialised experience, knowledge, skills or contacts gained as in their former capacity. 24
Persons exempt from registration include those engaged in humanitarian aid or assistance, the provision of legal advice or representation, religious activities, registered charities, artistic purposes, diplomatic and consular officials and the members of certain professions, such as tax agents, customs brokers and liquidators and receivers. 25 Deliberately excluded from the list of exemptions are academics, universities and schools, media organisations and think-tanks. Those who are registered under the Act have onerous responsibilities, including the reporting of information or other communications distributed to the public in Australia on behalf of a foreign principal, 26 reporting material changes in circumstances, 27 annually renewing registration 28 and keeping records. 29 There are also notification requirements to cease to be registered. 30 Penalties for failing to register or acquit responsibilities under the Act include imprisonment from 6 months to 5 years. 31
Part 4 of the FITS Act explains the responsibilities of the Secretary administering the register, including broad powers to obtain information from any person within 14 days if the Secretary reasonably believes that person has information relevant to the scheme. 32 The Secretary is responsible for determining what information is to be made publicly available, such as the name of the person and the name of the foreign principal, a description of the kind of registrable activity undertaken and any other information for these purposes. 33 The Secretary can decide to omit information from the public register that is assessed as being commercially sensitive or affects national security. 34 This effectively creates two tiers of registration: a private and public register, with the Secretary having the power to decide what is made public. The Secretary has broad powers over how they communicate and deal with the information provided. That includes giving it to any enforcement body such as Australian police, or to any other Commonwealth Department, agency or other authority of the Commonwealth, State or Territory. 35 These powers are not only extensive but also largely unchecked.
The FITS’s reach is evident in its framing language. While part of a ‘Counter Foreign Interference Strategy’, the title of the FITS suggests its target is ‘foreign influence’ rather than just ‘foreign interference’. Foreign influence suggests a much wider net and a lower bar than foreign interference. Foreign interference is understood in the framing material as the use of ‘covert, deceptive, corrupting or threatening means to damage or destabilise the government or political processes of a country’. 36 This is different to foreign influence, which refers to the benign activities of a foreign principal that seek to influence government and political systems: only influence which is covert, deceptive or corrupting should constitute interference. 37
This difference between foreign influence and interference was highlighted in the debate preceding this reform. In his Second Reading Speech for a related Act in the CFIS, then-Prime Minister Malcolm Turnbull stated that Australia would ‘not tolerate foreign influence activities that are in any way covert, coercive or corrupt’ suggesting that this ‘is the line that separates legitimate influence from unacceptable interference’. Turnbull was emphatic that the reforms being proposed were ‘not concerned with “soft power,”’ which is ‘an attractive force’ because ‘[i]f another nation has cultural or economic gravitational pull then it suggests they are doing something right and we would all benefit from being involved’. 38 Accordingly, influence arising from persuasive argument, lobbying and debate in the free marketplace of ideas should not be restricted, even if it flows across international borders. The government’s factsheet similarly explained that ‘foreign influence is not a bad thing…when conducted in an open, lawful and transparent manner’ because it ‘contributes to our vibrant and robust democracy’. Foreign interference, however, went ‘beyond the routine diplomatic influence’ to include ‘covert, deceptive and coercive activities’ by foreign nationals to ‘advance their interests or objectives’. 39
Yet after making this careful distinction, the government enacted the FITS, which indeed regulates the ‘soft’ and ‘attractive force’ of ‘foreign influence’, 40 presumably to weed out or deter foreign interference, or to make it transparent and therefore legitimate. As Draffen and Ng point out, the aim of the FITS is to increase transparency through the public disclosure of the identity of those acting on behalf of a foreign principal. 41
There is, however, a contradiction at the heart of the FITS. The scheme goes beyond mere transparency, to restrict and arduously regulate foreign influence. The onerous registration requirements, the existence of a private register and the wide powers given to the Secretary, including the discretion to share information with other government agencies, together with criminal sanctions, impose a significant and unwarranted burden on speech more generally and the freedom of political communication in particular. 42 While voters should be informed of the influences on the decision-making of their elected officials, the effects of laws requiring foreign actors to register are generally not aligned with their objectives. 43 As similar schemes enacted in the United States, Russia, Ukraine and Israel demonstrate, such arrangements have stifled dissent through burdensome reporting requirements and the stigmatising label of ‘foreign agent’. 44 Most absurdly, the reporting requirements necessitate disclosure of foreign engagement — often via media or public speaking — which is already totally public and open. The reporting requirements duplicate what would already be readily available in the public sphere.
The Australian High Court has recently examined the validity of the FITS in LibertyWorks Inc v Commonwealth of Australia (‘LibertyWorks’) to determine whether the burdens it imposes on the implied freedom of political communication are justifiable.
45
By way of background, the High Court has long recognised that an implied constitutional freedom of political communication operates as part of a representative and responsible system of government. The implied freedom is not absolute, and nor is it a personal right. It may be limited by laws which are reasonably adapted, and which serve a legitimate end. To determine whether the laws in question are properly adapted, and thus justifiably burden the implied freedom, the Court applies a proportionality test. Initially articulated as a two-part test in Lange v Australian Broadcasting Corporation,
46
the approach was revised in subsequent cases
47
and now includes three limbs
48
: 1. Does the law effectively burden freedom of communication about government or political matters? 2. If ‘yes’, is the purpose of the law legitimate and compatible with the maintenance of the constitutionally prescribed system of representative and responsible government? 3. If ‘yes’, is the law reasonably appropriate and adapted to advance that aim in a manner that is compatible with the maintenance of the constitutionally prescribed system of representative and responsible government?
As Morris and Stone explain, ‘[i]f the first question is answered “yes” and the second or third “no,” the law is invalid’. 49 These court approaches have evolved to protect communication necessary to facilitate free and informed federal voting choices and to enable effective functioning of responsible government. 50
The facts of the LibertyWorks decision were as follows. LibertyWorks Inc is a private Australian conservative think-tank that promotes individual liberty and rights in public policy. Since its incorporation in 2015, it has organised several conferences in Australia and made submission to parliamentary inquires on freedom of speech. In 2018, the President of LibertyWorks agreed to run an American Conservative Union (ACU) event in Australia in 2019. A Deputy Secretary of the Attorney-General’s Department asked LibertyWorks to provide documentation to determine whether it had registration obligations. LibertyWorks refused to comply and filed an action in the High Court arguing the registration provisions unduly burdened the implied freedom of political communication.
In a 5:2 decision, the High Court found that the policy objective of controlling and minimising foreign interference through increased transparency justified the registration scheme, which was proportionate to achieving the policy ends. Though the majority accepted that the registration requirements might deter a small number of people from engaging in political communication, 51 they did not think the scheme created a ‘chilling effect’ 52 and characterised the burden as modest. Their Honours held that the scheme was rationally connected and therefore suitable to the stated purpose of the Act. 53 They further found that the scheme was reasonably necessary to achieve the stated purpose of promoting transparency in political discourse because ‘[r]egistration enables both the relationship between the person and their foreign principal and a description of the political communication undertaken by the person … to be matters of public record’. 54
This is a conclusion we find unpersuasive. As noted in the two dissenting judgements by Gageler and Gordon JJ, the two tiers of public and private registration under the FITS means the legislation fails to achieve transparency. Gageler J observed that the registration requirements operate as a precondition to engaging in political communication, which is incompatible with the implied freedom because it burdens the registrant much more than is necessary to achieve the legislative intent. 55 It restricts communication at the outset by forcing the relevant entity or person to register, under pain of criminal sanction, before political communications can occur. This is more prohibitive than a system of subsequent punishment. 56 For Gageler J, the registration requirements were therefore not ‘reasonably appropriate and adapted’ to advance the Act’s purpose in a way that is consistent with the maintenance of the constitutionally prescribed system of government. Gordon J similarly emphasised the definition of ‘communications activity’ which regulates political communications of ‘the broadest kind’. 57 The provisions are not limited to circumstances where a person is acting in the service of or at the request of a foreign government or a foreign government entity: they apply to any person who has an arrangement with a foreign organisation whose purpose is to pursue political objectives, irrespective of whether their views align with those of the foreign principal. 58 This creates a substantial and arguably unjustified burden on political communication.
If the purpose of the scheme is to preserve a country’s self-determination and national sovereignty, then as Draffen and Ng convincingly argue, ‘such a law should accordingly be directed at the actors that hold the greatest influence over elected officials’ decisions and thus reduce the power of citizens and residents to influence politics’. 59 Who are these actors that have the greatest influence? Draffen and Ng suggest that democracy is undermined ‘by the influence of the few who wield disproportionate power through networks and resources to which the average person does not have access’. 60 As such, foreign actor registration schemes should be narrowly targeted to those actors who ultimately pose a risk through their significant resources and access to foreign governments, rather than actors who are just in some way connected to a ‘“foreign” element’. 61 Registration requirements, where they exist, should be aimed at the ‘powerful few who in reality influence our leaders the most’. 62 On the basis of Draffen and Ng’s principle, the FITS should cover those who undertake a ‘significant level or type of activity that warrants closer public scrutiny’. 63 This will only include entities like foreign governments and political parties, who have the power, resources and funding to directly access and influence elected officials, but it should not include foreign individuals or private corporations. 64
Contrary to this approach, the FITS imposes registration requirements on a broad range of foreign actors. It includes foreign governments, defined as ‘the government of a foreign country’, and foreign government-related entities such as companies where the foreign principal holds more than 15 per cent of the issued share capital, more than 15 per cent of the voting power or is in a position to appoint at least 20 per cent of the company’s board of directors. Other registrants include foreign government-related individuals, meaning persons who are not Australian citizens or residents and are related to a foreign principal, ‘foreign political organisations’ such as a foreign political party or foreign organisations that exist ‘primarily to pursue political objectives’. 65
According to Draffen and Ng, expanding the scope of registrants beyond agents of foreign governments and foreign political parties, to principals who are foreign individuals and corporations not linked to government, is difficult to justify. Such registration schemes have been used in both Russia and China to silence, eliminate or control civil society organisations by labelling the organisations as foreign agents. Even in the United States, the Foreign Agents Registration Act (FARA) has been used to ‘brand certain films as “political propaganda,” effectively creating a chilling effect by deterring American distributors from importing these films and showing them to the public’. 66
Thus, only entities like foreign governments and political parties should be required to register to make their influence on political decision-making more transparent, not foreign organisations engaging in political influence. On these grounds, LibertyWorks should not have had to register. The American Conservative Union (ACU), who was co-sponsoring the event, is also not a political party. While based in the United States, there is no evidence that either organisation has the power and resources to do anything but influence opinion using ordinary and accepted channels of communication in a democracy: namely, public communication. The fact that the communications activity being regulated is already public and overt adds additional weight to the argument that this type of communication should be classified as foreign influence, rather than covert foreign interference. This activity should not be captured by legislation.
The practical impact of the FITS lends weight to our concerns that it is ineffective in achieving its aims, while over-burdening free speech, especially the speech of think-tanks, journalists, academics and other high-profile individuals, such as former Prime Ministers.
A.i. FITS Act and the Media
News organisations are captured by ‘communications activities’ under section 21 of the FITS Act. On the one hand, these communication activities can be very influential in affecting the opinions of those involved in Australian political processes and decision-making. If there is a foreign principal involved, there is a case that this should be transparent so the public can make an informed decision. However, as Draffen and Ng point out, the US experience with FARA demonstrates that this kind of law creates costs for compliance and uncertainty around who should register. 67 That has a chilling effect on the free flow of international ideas.
This is particularly awkward for foreign news bureaus based in Australia or for freelance reporters who file for foreign news organisations. The ownership structures of many international news organisations such as the BBC (publicly owned but explicitly independent) or the New York Times (a privately owned business) means they fall outside the legal definitions of ‘foreign principal’. But others, such as Al Jazeera English (financed at least partly by the government of Qatar), 68 or the international broadcast network Deutsche Welle (financed entirely by the German government), 69 are more problematic. While the law is unclear, in both cases the FITS Act appears to regard these organisations as ‘foreign principals’ even though they claim editorial independence and are widely seen as authoritative news services. The main purpose of all those organisations is to gather news from within Australia, primarily for audiences in their country of origin. They have a tiny footprint inside Australia and hold little direct influence on domestic politics. But, s 13(1)(a) of the Act says, ‘[a] person undertakes communications activity if: the person communicates or distributes information or material to the public or a section of the public …’. This suggests any foreign news organisation with a website accessible in Australia, and with an Australian office, must register.
Based on the free speech principles earlier outlined, however, news organisations should arguably not be liable to register. By its very nature, news reporting is a public activity. While journalists may advocate for particular political positions or seek to influence the deliberative landscape, this is precisely what the marketplace of ideas is for and why we value this free exchange of ideas in a democracy. A possible objection to this argument is that while reporting is public (and thus transparent), it is conceivable for news organisations to nevertheless covertly distort the deliberative landscape via disinformation and misinformation campaigns or by promoting espionage. While this is certainly possible, it is not clear that a system of registration is the best way of addressing this issue — a point to which we return in the second section.
The absurdity of the registration requirements is further highlighted by the experiences of two former Australian Prime Ministers, Kevin Rudd and Tony Abbott, both of whom had to seek legal advice on their obligations under the FITS Act. Before he was appointed as Australia’s ambassador to Washington, former Prime Minister Kevin Rudd was one of the most prolific of the FITS Act registrants. He logged his various public speaking engagements, writings and appearances for foreign-owned media to discuss current issues with international public broadcasters, such as the BBC and Radio New Zealand. His registration documents demonstrate the ongoing legal disagreement between him and the Attorney General’s Department about his obligations, with Rudd disputing the claim that he is an ‘agent of foreign influence’. 70 On his registration forms, Rudd also routinely highlighted another weakness of the scheme, pointing to a Memorandum of Advice written by Brett Walker SC for the Senate References Committee on Communications and the Environment. Walker argued that under a wide interpretation of the law, News Corp ought to also be listed as an agent of foreign influence on the register. 71 In his memo, Walker cited a story published by a News Corp publication that appeared to be based on information produced by anonymous ‘Western governments’. Walker argued that the information was passed to the publication for political purposes and intended to sway political decisions. The paper chose not to reveal the source of that story, and News Corp had not registered.
While Walker appeared to be exposing News Corp as an undisclosed agent of foreign influence, his memorandum highlights the very serious implications that the FITS Act has for editorial independence more broadly. Rudd’s attempt to out News Corp (presumably as part of his crusade against that organisation) shows how this legislation can be politically weaponised, but it also demonstrates the breadth of the registration scheme — if Rudd and News Corp need to register their international media engagements, presumably every Australian news outlet needs to register each time they seek comment from a foreign government or agency.
This has serious implications for free speech and journalism. Freedom of speech is crucial for the protection of public interest journalism that keeps governments accountable, provides us with information to enable informed electoral choices and facilitates public deliberation necessary for a vibrant and robust democracy. Sources engage with journalists for a variety of reasons, but they are often motivated by a desire to influence political outcomes or policy debates. For journalists, those motivations are generally secondary to the significance of the information alone, and journalists are also expected to hold to the principle of protecting a source’s identity on request. For example, foreign governments seeking to sway domestic public opinion about international trade negotiations would routinely speak to Australian journalists and pass on critical information — sometimes anonymously. That kind of communication is routine journalists’ work and arguably routine diplomacy but still appears to require registration under the FITS Act.
Similarly, Tony Abbott was asked to register because he was giving an address at the Australian Conservative Political Action Conference, an organisation founded by the American Conservative Union, with links to the US Republican Party. Abbott labelled the decision ‘absurd’, angering the then Attorney-General, Chris Porter. 72 Of note for our purposes is that both Rudd and Abbott were giving public interviews and addresses, which by their very nature are transparent. If there is anything untoward in these engagements, it can be dealt with using the ordinary channels of public deliberation in a democracy, such as through publicly challenging the views being expressed, contestation, and so on. It is not clear what is achieved through the registration scheme that cannot be achieved through ordinary democratic channels. Furthermore, if both high-profile and well-resourced former prime ministers required extensive legal advice to determine their obligations, ordinary people will find the requirements even more onerous, which may chill speech. Given these issues and the principles earlier outlined, we suggest that the FITS Act unduly burdens implied freedom and is not fit for purpose.
A.ii. FITS and Academia
Along with the media, the FITS Act does not include exemptions for academics and universities. As noted by the A-G Department: universities are no different to any other organisation. If [a] university is closely affiliated with a foreign government … then it is appropriate for a person to register if they undertake registrable activities in Australia on behalf of the university for … political or governmental influence.
73
This potentially captures any Australian academic conducting research under an ‘arrangement’ with a foreign university to influence government policy. Such an arrangement may be completely benign, such as an academic conducting research on bushfire responses, to use Draffen and Ng’s example, or any other research designed to influence government policy. 74 A significant amount of academic research is aimed at influencing government policy or making reform recommendations. This is an essential part of research impact and one of the primary functions of the university and its place in a democratic society. As Evans and Stone argue, universities are institutions of civil society, and the freedom of speech that occurs within these institutions is an integral part of democratic government, personal freedom and dignity. 75 This legislative package potentially undermines or inhibits these freedoms, by chilling academic freedom, public interest journalism and free speech more broadly.
2 National Security Legislation Amendment (Espionage Act)
The FITS Act is but one piece of this complicated legislative puzzle. It is designed to sit alongside the Espionage Act and the Electoral Funding Act. All three Acts came into effect at the same time, but the Espionage Act is more comprehensive than the FITS Act. It repealed the four espionage offences that had been legislated in 2002, replacing them with 27 new offences. 76 As Sarah Kendall explains in her interrogation of this Act, ‘[n]ot only are these offences broader than previous offences, but many penalties are also more severe’. 77 Kieran Pender from the Human Rights Law Centre described the Act as, ‘the most significant revisions to Australia’s official secrecy laws in a century’. 78 The 27 new offences fall into four main categories: (i) treason, treachery and sabotage; (ii) espionage offences; (iii) foreign interference offences; and (iv) offences related to secrecy and disclosure of classified information. As Kendall points out, at their core, the foreign interference offences criminalise conduct that is covert, deceptive or threatening on behalf of, in collaboration with, or where directed, funded or supervised by a foreign principal or person acting on its behalf. 79
Category 1 offences involve egregious acts of violence and interference involving armed conflict such as treason, 80 treachery, 81 sabotage, 82 advocating mutiny, 83 assisting prisoners of war to escape, 84 military style training involving foreign government principal 85 and interference with political rights and duties. 86 They fall outside the scope of this paper which is concerned with the impact of the law on freedom of speech.
However, category 2 offences related to espionage are concerning for free speech. They include three forms of ‘dealing with information’: information concerning national security which is or will be communicated or made available to foreign principal 87 ; information which is or will be communicated or made available to foreign principal 88 ; and security classified information. 89 There is also an aggravated espionage offence, 90 espionage on behalf of a foreign principal, 91 the offence of soliciting or procuring an espionage offence or making it easier to do so 92 and the offence of preparing for an espionage offence. 93
Category 3 offences include the offences of intentional foreign interference (s 92.2), reckless foreign interference (s 92.3), preparing for a foreign interference offence (s 92.4), knowingly supporting foreign intelligence agency (s 92.5), recklessly supporting foreign intelligence agency (s 92.6), knowingly funding or being funded by foreign interference agency (s 92.9) and recklessly funding or being funded by foreign interference agency (s 92.10).
The final category of offences include theft of trade secrets involving foreign government principal (s 92A.1); secrecy of information: communication and other dealings with inherently harmful information by current and former Commonwealth officers (s 122.1); conduct by current and former Commonwealth officers, etc. causing harm to Australia’s interests (s 122.2); unauthorised disclosure of information by current and former Commonwealth officers, etc. (s 122.4) and communicating and dealing with information by non-Commonwealth officers etc. (s 122.4A).
The espionage and foreign interference offences have specific implications for journalists, academics and NGOs. A particular problem is the broad definition of key terms such as ‘concern’, ‘national security’ and what it means to ‘deal’ with information. As Ananian-Welsh, Kendall and Murry argue, ‘[t]he term “dealing with” is defined with exceptional breadth [and] includes receiving, obtaining, collecting, possessing, making a record, copying, altering, concealing, communicating, publishing, or making available’. 94 Simply placing the information or article somewhere to be accessed by another person risks violating the law. So does giving the information to an intermediary to pass on to someone else, or describing how to obtain access to it, or other methods that could facilitate access (such as disclosing the name of a website, an IP address a URL or the name of a newsgroup). 95 While spies might occasionally use those methods, they are routine for journalists and academics whose main currency is information.
‘Foreign principal’ is also defined broadly to include not only foreign governments but also foreign political organisations, a terrorist group or an entity or organisation owned, directed or controlled by a foreign principal. This would include foreign universities and news entities owned or controlled by foreign governments, in addition to companies where the government holds more than 50 per cent of share capital. 96 ‘National security’ for the purposes of all the offences is defined broadly to include the defence of the country, including its territorial integrity, the protection of the country from activities such as espionage, sabotage, terrorism, political violence and foreign interference, and the country’s political, military or economic relations with another country or countries, effectivity drawing Australia’s international relations into the orbit of national security. 97
Offences in the second category concern dealing with security classified information which is or will be communicated to a foreign principal (note: ‘security classification’ refers to a classification of secret or top secret). The espionage offence of dealing with information concerning national security carries a penalty of life imprisonment, while a penalty of 25 years imprisonment applies to being reckless as to national security. For journalists, this staggeringly broad definition could be interpreted to include almost any story about foreign affairs, trade, strategic or military relations.
While the FITS Act appears to be an ineffectual and bureaucratic annoyance (and potentially a deterrent to free speech), the espionage offences are gravely consequential for news organisations and journalists. When the legislation was debated in 2018, the journalists’ union, the MEAA, argued that the sweep of the offences captured a significant proportion of their otherwise ordinary work. 98 As we have already seen, the extraordinarily broad ‘dealing with information concerning national security’ means that a journalist simply receiving a document or answering a phone call conveying security classified information runs the risk of violating the Act. Similarly, ‘communicating or making (information) available’ to foreign principles means publishing a story that includes ‘secret’ information that harms Australia’s interests runs the risk of putting the journalist in prison. 99
After several media organisations objected to the original wording of the legislation, the government agreed to give journalists a defence for the offence of dealing with protected information where they ‘reasonably believe’ it to be ‘in the public interest’. But the legislation remains deeply problematic. While some media groups welcomed the defence, the MEAA called for a blanket exemption from prosecution because the concept of the ‘public interest’ was vague, and the classification of documents as ‘secret’ or ‘top secret’ was an administrative decision that could trigger a criminal prosecution. At the same time, they said attempts to mount and prove a defence might force journalists to reveal information about their sources. 100 The editor of The Guardian Australia, Lenore Taylor, argued the law effectively expanded the range of reporting that risked criminal sanction and increased potential penalties for journalists to 10 years in jail. ‘That certainly focuses the mind’, she wrote. 101
It is worth considering stories published before the legislation came into force, that might have crossed the line under the Act and therefore might not have been published, or could have put journalists in prison. In 2017, the ABC published allegations that Australia’s Special Forces had committed war crimes in Afghanistan. 102 The story was based on information contained in leaked classified documents and ultimately triggered the Brereton Inquiry which recommended sweeping reform of the military. It is hard to dispute the importance of the story.
In November 2013, The Guardian published details of how Australia’s Department of Defence and the Defence Signals Directorate tried to monitor the phone calls of the then Indonesian Prime Minister and his wife. 103 Again, the story relied on leaked classified information and triggered a brief crisis in relations between the two countries — factors that would have made it extremely dangerous to publish under the current legislation, despite a clear public interest in the story.
Kendall provides other examples of the ways in which the legislation will undermine press freedoms, such as situations when a journalist is reporting but intends or is reckless about influencing a political or government process, such as by advocating for electoral reform or encouraging the public to vote against a corrupt or unethical political party. There are other situations where a journalist or a source might be reckless about prejudicing Australia’s national security, for example, by publishing a story that reveals war crimes by members of the Australian Defence Force, but that is nevertheless, in the public interest. 104
Academics might also be caught by the Espionage Act if they engage in research that is critical of the Australian military or intelligence policies and practices, or which catalogues Australian government misconduct in its dealings with other countries. 105 As Kendall notes, an academic may also commit an offence by simply teaching students about this research or engaging in preliminary research which can be captured by the preparatory offences for espionage. An academic working for a foreign public university (even if the country is an ally of Australia’s) is also risking prison by running an anonymous survey to collect information to advocate for Australian electoral reform. The anonymous nature of the survey could be enough for it to be regarded as ‘covert’ under the legislation. 106 Like the criminalisation of journalists, the kind of academic research that the Espionage and Foreign Interference legislation makes illegal may be in the public interest, or it might produce new knowledge that has important social and political consequences 107 — hardly an ideal outcome for a law intended to protect an open democracy.
Division 92 specifically addresses foreign interference (as opposed to the FITS, which regulates foreign influence). S 92.2 makes it an offence for a person to engage in conduct on behalf of or in collaboration with a foreign principal if the conduct is directed, funded or supervised by a foreign principal; the person intends to influence a political or governmental process or influence the exercise of an Australian democratic or political right or duty, or support the intelligence activities of a foreign principal and prejudice Australia’s national security. Significantly, s 92.2 captures conduct that is covert or deceptive, or involves a person making a threat to cause serious harm, or involves the person making a demand with menaces, raising it above the definition of the more benign ‘influence’ captured by the FITS.
That the espionage legislation appears much more appropriately targeted than the FITS raises two significant questions. First, it brings the relevance and appropriateness of the FITS into sharp relief. What is the point of the FITS if there is specific legislation addressing covert and deceptive foreign interference? The FITS may perversely chill the speech of various actors, including academics, journalists and other civil society figures engaging in lawful foreign influence, while failing to increase transparency.
Second, while the purposes of the two Acts differ (the FITS has a transparency purpose, while the Espionage Act is for the prosecution of offences), having legislation which covers allegedly benign and productive foreign influence creates confusion. In a recent inquiry into national security risks that affect the Australian higher education and research sector, many universities argued that the government’s policies and legislation were too complex, disconnected and confusing. 108 Monash University noted that legitimate foreign engagement and influence activities had been mischaracterised as interference because the differences between the legislation was not clear to end users, irrespective of the government objectives. It argued that the legislation created a climate of fear among academics who engage internationally (and who are encouraged to do so, including for promotions) and risk damaging the sector’s international standing. 109 La Trobe echoed these concerns, complaining of ‘many overlapping and disjointed processes without any apparent central coordination’. 110 Queensland University of Technology said this creates serious compliance challenges for universities and could undermine their collective efforts to resist foreign interference. 111
The authors have had first-hand experience with these schemes in our engagement with international universities, partnerships and consulting for foreign organisations. The administrative burden is immense. When we have emailed the university’s dedicated team to find out if we are required to register our international engagement activities under the legislation, answers are never clear. This administrative burden takes university resources away from education and research, and we end up questioning whether the partnership or consulting is worth the risk and effort. While there needs to be greater collection of data, there is clear anecdotal evidence that some academic speech has been chilled by the blurred legislative boundaries between interference and influence.
Finally, one university administrator mentioned that the scheme forced them to create a detailed database of information on foreign engagement that would be an attractive target for hackers and cyber-criminals seeking access to that kind of information. Once again, the scheme has perversely appeared to increase the risk to national security.
B Electoral Reform Act
The 2018 reforms also addressed foreign political donations as an avenue for potential manipulation of Australian politicians. 112 The 2016 Sam Dastyari saga (which compelled the Senator to resign) had re-focussed attention on foreign donations. 113 Following initiatives in several States to ban foreign donations, 114 the Joint Standing Committee on Electoral Matters (JSCEM) published an interim report into the 2016 federal election. It recommended prohibiting donations from foreign citizens and foreign entities to Australian registered political parties, associated entities and third parties. This ban would not apply to dual Australian citizens either in Australia or overseas, or to non-Australian permanent residents in Australia. 115 The government supported the proposal, noting that ‘only Australians and Australian entities’ should be able to ‘participate in our elections’. 116 As the Bill was being drafted as part of the CFIS, there was a debate about whether third parties like charities and NGOs should be banned from receiving foreign donations, because of concerns about the administrative burden stifling civil society advocacy. 117 The 2018 Bill ultimately included carve-outs for charities which are not prevented from using foreign donations to advocate for non-partisan issues but cannot use foreign money for political spending. 118
Foreignness is defined in the Act to mean any overseas government or public enterprise, any entity not incorporated (or lacking a principal place of activity) in Australia, or a person who is neither a citizen nor a permanent resident of the country. 119 The legislation prohibits a party, candidate or registrable ‘political campaigner’ from accepting a gift of more than $1000 from a foreign donor 120 and reciprocally prohibits a foreign donor from donating to an Australian party, candidate or third party for federal election purposes. 121 The legislation also restricts parties, candidates or third parties from accepting gifts of $100 or more from foreign donors, where that money is intended or used for electoral purposes. 122 It further prevents third-party campaigners from accepting foreign donations equal to or over the ‘disclosure threshold’ of $13,800, for purposes related to ‘incurring electoral expenditure’ or ‘creating or communicating electoral matter’. 123 However, the legislation specifies that these sections can be avoided if the gift recipient corrects the violation within 6 weeks of receiving it. This was intended to allow time to investigate the source of the money and return it or relinquish it to the Australian government. 124
There are several problems with these amendments. First, enforcement may be difficult. It is unclear how the prohibitions relating to foreign donations can be enforced in international jurisdictions. As Orr and Geddes point out, ‘the practicable means’ to enforce offences targeted at foreign donors offshore are ‘questionable, unless the donor is a regular visitor to the country’. 125 Second, the reform does not address the problem it sought to fix because money flows are fluid. As various experts have noted, foreign companies can still funnel donations through Australian subsidiaries, domestic trust arrangements or local Australian individuals. 126 As the Panama Papers showed, big corporations regularly shift money around to avoid Australian tax obligations and could easily use similar techniques to avoid these donation laws. Further, as Orr and Geddis observe, ‘a party could still accept more than A$1000 (via staggered payments) from a foreign source, as long as there was a mutual insistence that the money does not go towards electioneering’. 127 Even within domestic entities, the legislation is easily avoidable through normal accounting. For example, if a company puts $1000 of foreign donated money towards benign administration costs, that frees up $1000 of domestic money that can be spent on electioneering. According to Twomey, the only way to limit the influence of foreign money is to impose a cap on all donations. 128
Perhaps most glaringly, the two donors central to the Dastyari controversy and other furores at the time remain beyond the ambit of the legislation because Chau Chak Wing is an Australian citizen and Huang Xiangmo is a permanent resident. 129 These individuals do not meet the legislation’s definition of ‘foreign’, so the reforms would not have prevented the Dastyari saga. Rather, that controversy showed that sunlight and political sanction alone were powerful ways to root out improper dealings with foreign entities. The scheme is also complex and difficult to understand, and its reach is broader than necessary. Both these problems make compliance and enforcement even more complicated. Given the legislation is ineffective in achieving its aim, it is debateable whether the law is proportionate under the implied freedom of political communication. Viewed with the other limbs of this scheme, this legislation imposes an onerous burden on free speech, bolstered by criminal penalties, without preventing foreign interference through donations.
Applying the prohibitions to third party campaigners may be particularly questionable, given the High Court’s 2013 decision on donation caps in NSW legislation which chilled the free speech of third party campaigners. 130 This element of the federal donations law was characterised as ‘overkill’ — particularly because the prohibited use of funds is not time limited to an election period but construed broadly as communicating electoral matter. 131 In the first Unions NSW case, 132 the Court said political communication ‘is not simply a two-way affair between electors and government or candidates’. Rather, ‘[t]here are many in the community who are not electors but who are governed and are affected by decisions of government’, and these entities ‘have a legitimate interest in governmental action and the direction of policy’ which means ‘that they, as well as electors, may seek to influence the ultimate choice of the people as to who should govern’. 133 The Court found that s 96D of the Election Funding and Disclosures Amendment Act 2010 (NSW) impinged upon the implied freedom but did not serve a legitimate end — hence, it failed step 2 of the Lange test. This was because the requirement that a donor must be enrolled to vote was not connected to the anti-corruption purposes of the Act, 134 a finding that could be relevant to a future assessment of the validity of the federal ban on foreign donations. Similarly, s 95G(6) created a burden on political communication by capping the amount a political party may spend within the relevant period, but this was also not sufficiently connected to the legitimate anti-corruption purposes of the Act.
The 2013 decision demonstrated that legislated burdens on the implied freedom must be logically connected to the ends they serve. This lends weight to an argument that any legislated burden on the implied freedom must be effective in achieving the policy ends; otherwise, the law is more likely to be unnecessary and disproportionate. However, those NSW laws could be distinguishable from foreign interference laws federally because, as Twomey highlights, the NSW laws were ‘politically unbalanced’. They ‘attacked the relationship between unions and the ALP’, demonstrating a ‘flagrant’ political purpose, which perhaps prompted the High Court to take the ‘unprecedented step’ of ‘pointing out that there was no legitimate end for the laws at all’. 135 By contrast, the aim of preventing nefarious or corrupt foreign interference via foreign donations would seem a legitimate purpose compatible with Australian democracy. However, questions about political bias were raised in debate around the 2018 scheme, and the foreign donation laws were criticised for targeting at left-leaning organisation Get Up!, 136 potentially raising comparisons with the 2013 unions case. Nonetheless, it is constitutionally possible to ban donations from a subsection of the population, which lends weight to an argument for validity of the federal ban on foreign donations. In McCloy, the High Court upheld a NSW law preventing property developers from making donations, accepting that a legitimate purpose of capping donations could be to facilitate ‘a level playing field’ in electoral participation 137 as well as preventing corruption.
Whether the burden on third-party campaigners would be valid under the implied freedom may be further illuminated by the 2019 case of Unions NSW v NSW (No 2). 138 This examined s 29(10) of the Electoral Funding Act 2018, which capped electoral spending by third-party campaigners to $500,000 in the 6 months leading up to a State election. Citing McCloy, the High Court noted that donation caps aimed to ‘ensure that wealth does not create an obstacle to equal participation in the electoral process by allowing the drowning out of the voices of others’, 139 but held that s 29(10) breached the implied freedom. The Court rejected the argument that ‘candidates and political parties occupy a constitutionally distinct position’ which legitimises their ‘preferential treatment’ under legislation burdening the implied freedom. 140 Kiefel CJ and Bell and Keane JJ held that ‘ss 7 and 24 of the Constitution guarantee the political sovereignty of the people of the Commonwealth by ensuring that their choice of elected representatives is a real choice, that is, a choice that is free and well-informed’. 141 The High Court thus reiterated the importance of the speech of third-party actors for the democratic system. However, the focus on the ‘political sovereignty of the people’ could also suggest a different approach might apply to restrictions on foreign participation in Australian politics via foreign donations, as they apply to third parties. On the other hand, the plurality noted that caps on expenditure presented a more direct burden on political communication than caps on political donations, 142 perhaps suggesting that restrictions on foreign donations (although the laws also regulate how such donations can be spent) might attract more leeway under the implied freedom. 143 Either way, the chilling effect of the added administrative burden should also be taken into account.
On balance, we think the foreign donation laws could be vulnerable to invalidity under the implied freedom. While the federal donations legislation imposes rules which would encourage greater vigilance from entities accepting donations to inquire into the source of the money, it also imposes administrative burdens which may chill advocacy, but it does not effectively prevent foreign political donations or foreign interference. Like the FITS registration requirements, the legislation adds to red tape and bureaucracy which might hinder NGOs and think-tanks with limited resources from legitimate public engagement. The foreign donation laws also do not address foreign financial manipulation of Australian politics occurring through other pervasive avenues. For example, the legislation has nothing to say about foreign funding of Australian media, subsidiaries of foreign-owned companies that donate to Australian politicians or the impact of foreign-owned social media companies that have immense potential impact on democratic debate, and which are powerful channels of potential foreign interference.
C Australia’s Foreign Relations Act
For completeness, we briefly consider the Foreign Relations Act of 2020. Its purpose, according to the Department of Foreign Affairs and Trade, is to ensure that arrangements between state or territory governments and foreign entities do not adversely affect Australia’s foreign relations and are consistent with Australia’s foreign policy. This scheme includes universities as ‘entities’ of state and territory governments. It requires notification to, and approval by, the Minister for any arrangement with a ‘core foreign entity’. A foreign entity is defined as a foreign country, its national government and a department or agency of that national government. It also includes foreign universities that do not have institutional autonomy. The Minister will approve an arrangement if they are satisfied that the proposed negotiation will not adversely affect or would be unlikely to adversely affect Australia’s foreign relations and that it would not be inconsistent with Australia’s foreign policies. While this Act is more concerned with foreign relations rather than foreign interference, like the FITS, it imposes onerous reporting requirements on universities and creates confusion about what arrangements need approval, given the vast international networks and arrangements in which universities typically engage. It adds layers of burdensome and bureaucratic legislation potentially inhibiting foreign engagement and free speech.
Is the Legislation Fit for Purpose?
In the previous section, we outlined the complex, overlapping and punitive legislative regime regulating foreign influence and foreign interference in Australia. We have demonstrated the ways in which the legislation imposes significant burdens on the speech of civil society actors, including journalists, academics and think-tanks. This raises the question of whether the legislation is functioning to prevent foreign interference. Is it in fact working to achieve its aims? Given the complexity of foreign interference, is it appropriately targeted and thus proportionate?
We have described the ways in which the severe penalties may deter public interest journalism, academic research and international engagement, and stifle the speech of other civil society entities. The risk here is that the legislation brings about the very outcomes it was designed to prevent, contributing to democratic decline by chilling free speech and distorting the media landscape. But to fully appreciate the extent to which this legislation is poorly targeted, we need to consider how foreign interference occurs, its covert nature and why legislation might not be the best way of managing it.
Cyber-attacks using social media platforms have been a primary vehicle for foreign interference. For example, the US Senate Intelligence Committee reports confirm the cyber tools used by Russia to interfere with the 2016 US Presidential election. These included cyber-intrusion into voter rolls and electoral systems, and algorithmically targeted propaganda and disinformation campaigns disseminated through social media, designed to create divisions in civil society. Similar interference has been reported in the United Kingdom, France and Germany. 144 These disinformation campaigns have been described by a former Soviet disinformation officer as ‘carefully constructed false message[s] leaked to an opponent’s communication system in order to deceive the decision-making elite or the public’. More recently, Yevgeny Prigozhin, the head of the Kremlin-linked mercenary unit, the Wagner Group, publicly boasted, ‘we have interfered, are interfering and will interfere (in US elections). Carefully, precisely, surgically and in our own way, as we know how to do’. 145 As Hunt notes, the purpose is to create doubt and confusion over the facts and the sources of those facts. These campaigns propagated conspiracy theories around issues of the day, such as election integrity and COVID-19, which were then disseminated though social media platforms, in some cases, taken up by public officials and those seeking public office.
A 2021 report from the US Treasury explained how the Kremlin sought and was able to obtain, polling data from the Trump campaign to microtarget voters. 146 It then leveraged social media to spread myths about voter fraud, designed to erode trust in electoral infrastructure and democratic processes. 147 This kind of foreign interference is thus a serious threat not only to electoral processes and political decision-making but to democracy itself.
While this level of interference and its consequences have not yet been observed in Australia, a recent report by the Senate Select Committee on Foreign Interference through Social Media (‘the Committee’) cautions that it would be naïve to think Australian elections and public debates have not and will not be the subject of similar attempts. 148 In its submission to the Committee, the Department of Home Affairs noted that it ‘regularly observes campaigns unfolding on social media that involve disinformation’, and some have been linked to foreign actors. 149 Foreign interference is not just about a foreign entity trying to boost one candidate over another in an election. Instead, it includes a range of actors trying to do many different things, such as actively sowing misinformation, trying to inflame existing social tensions and divisions or just creating a general environment of distrust to destabilise the democracy. 150
There are also many different online mechanisms used by foreign actors. Foreign interference can occur via direct means, such as brute-force hacking to gain access to Australian systems (as demonstrated with the recent Medibank attack by Russian actors), attempts to gain access by deceiving users and using distributed denial-of-service attacks. It can also occur through indirect means, primarily through commonly used and trusted social media platforms. This form of interference is much more difficult for end users to detect and it includes coordinated inauthentic behaviour (‘CIB’), algorithmic curation, microtargeting, the use of bots, human-driven interference and automated moderation. 151
The Department noted that foreign actors regularly use social media to promote narratives and spread disinformation, serving the strategic interests of the actors, while undermining democratic processes and institutions and stifling dissenting voices. 152 The Department also advised that ‘if left unchecked, foreign interference can exploit Australia’s way of life and open system of government to erode our sovereignty’. It stated that acts of foreign interference can ‘limit the Australian polity’s ability to make independent judgements and can corrupt the integrity of Australia’s systems’. Such acts can erode public confidence in Australia’s political and government institutions and can interfere with private sector decision-making, hampering both national security and economic prosperity. 153
Despite these risks, the Committee heard that there was significant confusion among the various government departments about who is responsible for what in relation to this complex issue and noted that there was a lack of coordination among government departments. For example, the First Assistant Secretary of the Department of the Prime Minister and Cabinet’s National Security Division was unaware that the COVID-19 taskforce was also working to combat online misinformation and disinformation and officials from the Department of Home Affairs (who are supposedly the policy leads) were not aware which platforms were supposed to report attempts at foreign interference to them. 154
There was similar confusion on the part of social media platforms. Representatives from TikTok did not know if they were supposed to report attempted foreign interference detected on their platforms, or who they should report them to. In fact, at the time of their appearance before the Committee, the government had never contacted the platform about their expectations. 155 This is despite the fact that social media platforms have strategies for identifying and countering disinformation and misinformation campaigns. For example, Google, which also owns YouTube, described some of its security features, such as Gmail protections against phishing and Safe Browsing in Chrome. It also tracks more than ‘270 targeted or government backed attacker groups from more than 50 countries’. 156 Facebook noted that it had found four examples of CIB in Australia and that domestic actors are also engaging in CIB, which in turn helps foreign powers use the same behaviour. 157
Despite the scale of the problem and some attempts at countering foreign interference, there is no legislative requirement that social media platforms report to government. This is especially concerning given that much disinformation and misinformation is difficult for end users to detect, especially younger audiences or less discerning users, many of whom use social media as their primary way of accessing news and information. It also underscores questions about the effectiveness of Australia’s foreign interference legislation, lending weight to the concerns raised in this paper. It is therefore surprising that the legislation examined here includes no reporting requirements for social media platforms, but quite onerous reporting requirements for academics and journalists, among other public figures.
While the legislative framework — as confusing as it is — may deal with some threats, it is not clear that the criminal law in its current form is the most effective and targeted way of addressing foreign interference in its modern forms. First, much of this foreign interference and espionage occurs from outside Australia, and those responsible would need to be extradited to face prosecution in Australia. As Kendall notes, this could be a significant impediment to prosecutions, especially where the person is in a country that does not have an extradition treaty with Australia or the treaty is not in force, such as the case for China and Pakistan, to name a few. 158
Second, identifying the person or groups responsible for the espionage or interference and so determining who to charge can be difficult. Given that inference can occur using bots, and how easy it is for people to hide their online identities with proxy servers, it is not clear that those responsible for the most egregious inference and espionage will be legally accountable. Since 2018 when this legislation came into force, only two people have been charged. Duong Di Sanh was accused of preparing to commit foreign interference; however, the exact nature of the alleged offences remains unknown. Mr Duong was linked with the China Council for the Promotion of Peaceful National Reunification, an organisation affiliated with the Chinese Government’s overseas influence arm, the United Work Front Department. 159
In June 2020, the Australian Federal Police and intelligence agencies raised the home and parliamentary office of NSW Upper House MP Shaoquette Moselmane as part of another foreign interference investigation into his part-time staffer John Zang. Mr Zang has denied any wrongdoing and subsequently filed a High Court challenge to the legislation, especially s 92.3(1) and (2) of the Criminal Code on the grounds that they infringed the implied freedom of political communication. However, the High Court held that the warrants were not invalid on any of the identified grounds and that the remaining substantive questions were unnecessary and inappropriate to answer. 160 At the time of writing, a Sydney man had been charged in NSW with a foreign interference offence, after he allegedly accepted money in exchange for information about national security issues. His legal counsel argue that the information was from publicly available open sources as part of a consulting arrangement during China’s COVID lockdown. 161
Third, it is not clear that the registration requirements under the FITS are improving transparency. As noted by these authors elsewhere, a review of the publicly accessible register demonstrates the wide range of actors captured by the legislation and the kinds of communicative activity in which they are engaged. 162 It captures many activities and individuals for engaging in routine communications activities that are part and parcel of living in a democratic and globalised society. Most of these entries do not appear to be engaging in untoward foreign interference — and if they were this would not be evident from the information presented on the register, undermining the policy purpose of the Act. This does not seem compatible with the implied freedom. In short, the FITS is not fit for purpose.
Reform Proposals
Given the problems with the legislative regime and the complexities of foreign interference identified in this paper, we make some recommendations for reform. The first is to abolish the FITS. Even Malcolm Turnbull, whose government oversaw implementation of the FITS, admits it is not working, arguing that the laws are cumbersome and that the information and relationships being reported are so innocuous they are barely worth reporting. He noted that the register ‘seems to be hoovering up a lot of information of marginal utility while missing the more disturbing forms of foreign influence it was designed to detect’. 163 Former PM Kevin Rudd has also highlighted the absurdity of the scheme, commenting that ‘National security officials have better things to do than chase former cabinet ministers to register TV appearances that, by their nature, are already on the public record’. 164 If there is widespread consensus that foreign influence is benign, then it is unclear why we need an expensive, burdensome and counter-productive piece of legislation to regulate and deter foreign influence. 165 If the proper target of this legislation is foreign interference, then the offences outlined in the Espionage Act should suffice. Second, we recommend stronger defences for academics, journalists and other civil society members such as think-tanks or organisations such as GetUp! should be incorporated throughout the legislation.
Third, we recommend that government develop a more targeted and coordinated strategy to work more collaboratively with social media companies to prevent foreign interference, which may include reporting requirements. The final report of the Select Committee on Foreign Interference through Social Media noted that social media companies have various interactions with government departments and bodies, but this was usually on an ad hoc and uncoordinated basis. For example, Facebook reported that it had been working with Australian electoral authorities and state and territory electoral commissions, while TikTok noted that it worked with the Australian Communications and Media Authority and that it gave the Department of Home Affairs some assistance.
166
WeChat described ongoing relationships with the Department of Home Affairs, the AG’s Department and the Australian Electoral Commission.
167
It was also noted that the relationship between government and social media companies is sometimes fraught, with non-compliance on the part of social media in some cases, and significant confusion about their reporting requirements.
168
At times, social media platforms have requested co-operation with government, with Facebook submitting that ‘there are greater steps the Australian Government could take to engage in information sharing with digital platforms and industry more broadly about foreign interference or influence operations’.
169
Similarly, Twitter noted that the foreign interference: threat we face requires extensive partnership and collaboration with government entities, civil society, experts and industry peers. We each possess information that others do not have, and our combined efforts are more powerful together in combating these threats.
170
This collaborative information sharing should be pursued with urgency, to co-design a comprehensive strategy for tackling online foreign interference. The advantage of imposing reporting requirements is to enable knowledge sharing and give government departments a greater awareness of when foreign interference is occurring and the scale of the inference, so that it might develop better targeted campaigns for at-risk groups and persons. It would also provide social media companies with clear legal direction, which currently does not exist.
Finally, we suggest that better public information campaigns about foreign interference and how it occurs, and greater collaboration with at risk sectors and communities, will be a far better approach to countering foreign interference than the punitive measures imposed by the legislative regime. A collaborative approach, in contrast to criminalisation, is more likely to build trust in government and democratic institutions, protect free speech and preserve the democratic values that are at stake.
