Electronic elections are associated with improved accessibility, efficacy, and cost. The digitisation of elections in Australia induces new vulnerabilities that malign foreign entities can exploit to subvert the nation’s democratic sovereignty.

Australia employs stringent hybrid digital-analogue mechanisms which safeguard voting processes from the digital problems of inauthenticity and data (in)security. While this offers protection from hard cyber problems, cyber risks to the integrity of the electoral system stem from factors other than direct interference in the procedural aspects of voting. For example, in 2019, Australia’s parliamentary network was hacked. Intelligence agencies reportedly attributed the hacking to the Chinese government. The hackers gained access to MPs’ emails and party databases. Such interference has no direct bearing on the vote, insofar as it does not manipulate votes to distort the result. Nonetheless it has the potential to indirectly affect the process through disinformation.

The Joint Standing Committee on Electoral Matters concluded disinformation poses a threat to democratic processes in Australia. Ivana Troselj also draws attention to the risk disinformation poses to Australia’s electoral process. She believes it is imperative to increase understanding of the “non-technical vulnerabilities” in the way in which Australia’s  “own culture and conventions facilitate the spread of disinformation.” Elections provide mechanisms for the articulation of preferences formed by the public between elections. Australian Strategic Policy Institute researcher Zoe Hawkins encapsulates the dangers of this, explaining that votes are the product of the “information ecosystem.” Elections may be where the results of interference take root, but the seeds have been planted outside the formal process in the public sphere.

The public sphere is an informal arena in which deliberation over political issues ensues and the dominant perspectives flow into the decision-making process. Some refer to this as “transmission” or “linkage” from the “wild” public sphere to “empowered” spaces, which not only transfers content from the public sphere, but also operates as a filtering process. Transmission affords malign foreign entities (MFEs) power in the context of shaping policy outputs—if the ideas circulating in the public sphere remain there, and do not move into more formal modes of preference articulation, then they do not directly function as inputs into public policy.

MFEs can therefore influence preferences and the political agenda before formal articulation processes begin. The electoral cycle can generate discourse in the public sphere. If MFEs manipulate the communicative exchanges in the public space and the information which sustains them, then the entire decision-making process becomes corrupted in a manner—and to an extent—that is difficult to detect and counter. The technical problems of detecting disinformation are well documented by computer scientists.

The 2016 US presidential election is an example of indirect “soft cyber” electoral interference using disinformation tactics. As the Mueller Report finds, Russia’s Internet Research Agency (IRA) conducted information warfare operations targeting the 2016 US presidential election. The campaign centred on social media to “provoke and amplify political and social discord in the United States.” The IRA spread information that favoured Donald Trump and criticised Hilary Clinton. The IRA purchased political advertisements in the names of US persons and entities. IRA-controlled Facebook groups, Twitter accounts, and Instagram accounts attracted hundreds and thousands of US participants. Facebook data revealed there were 470 IRA-controlled accounts that reached approximately 126 million users. The Mueller Report notes that Twitter confirmed 3,814 IRA-controlled accounts. The report projects that approximately 1.4 million Twitter users encountered IRA material.

The Mueller Report also finds that the Russian government’s Main Intelligence Directorate of the General Staff of the Russian Army (GRU) implemented widespread hacking to diminish support for Clinton. Evidence emerged of hacked email accounts and networks. Hackers released information through “fictitious online personas” and via WikiLeaks that was damaging to Clinton’s presidential campaign.

In France, Emmanuel Macron’s emails were hacked by Russian state-sponsored actors and US malign entities (alt-right groups) prior to the 2017 presidential election. The incident, popularly referred to as “#MacronLeaks,” involved hackers inserting false material into largely authentic email content.

As these examples illustrate, disinformation can be used to shape preferences (influence election outcomes), but also to sow discord and cast doubt on democratic processes. It can be perpetrated by states and their agents, as in the case of the IRA and GRU, or by foreign non-government groups or individuals. Clearly, digitisation has destroyed any geographic barriers to interference. As such, it has lowered the transaction costs so that interference can be conducted by anyone, anywhere.

Influencing elections through disinformation is problematic because it can reshape organic public inputs into the decision-making process. This indicates that while the preference articulation phase of participatory processes is largely secure from direct tampering or “hard” cyber threats, the preference formation and agenda-setting phases are vulnerable to MFE interference and can distort election outcomes by compromising input legitimacy.

Interference that targets people before they cast a ballot is very difficult to detect, let alone counter. It affects preferences, which in turn will affect what voters contribute to the participatory process—their input will reflect their interests, and their interests might be influenced by MFEs’ disinformation campaigns. Indeed, this remains a divisive issue with respect to the UK’s Brexit referendum—to what extent did Russian disinformation affect the result? Interference that corrupts organic preference-shaping mechanisms guarantees a distorted election result.

However, MFEs’ goals are not always directed at achieving a specific election result. Sometimes, the apparent aim is to destabilise a nation to create further vulnerabilities. Disinformation therefore also manifests as an acute concern for Australian democracy for its potential to diminish social cohesion. Discourse associated with elections and other national voting events such as referendums create opportunities for polarising discourse that has the capacity to exacerbate tension within society. Take for example, the Marriage Equality Survey of 2017: its very structure presented the issue in a binary manner with a “yes” or “no” vote required and it generated considerable polarised debate in the public sphere. MFEs can capitalise on existing and endemic polarisation to increase societal fissures and weaken pillars of liberal democracy such as tolerance, inclusion, equality, and trust. Such dynamics were present in the US where Russian disinformation operations reportedly fuelled protest activity without an allegiance to a particular side.

Unlike the US, Australia has mandated compulsory voting. This means the entire population can constitute an MFE target. MFEs have vast attack entry-points—every individual voter. However, compulsory voting might also reduce the consequences of foreign interference (e.g. polarisation). Richard Valelly argues that “when people are obliged to leave their homes and enter the public sphere, as they do when they vote, they tend to become more public minded.” The processes associated with voting, such as attending a polling booth and interacting with other people, have an egalitarian aspect which tends to unite citizens. The performance of citizenship might, in fact, translate to a sense of solidarity despite political, economic, and cultural differences. As well as analogue voter turnout at polling booths, compulsory voting might promote a sense of community, which may offset efforts at externally driven division. Though, elections are in some respects inherently divisive.

Dr Melissa-Ellen Dowling is a Research Fellow at the University of Adelaide. Her research interests centre around democratic resilience, foreign interference, digital information operations and national security policy. Currently, Dr Dowling is engaged in projects exploring cyber information operations and their impact on Australian democracy.

This is an edited extract from Dowling’s article in the Australian Journal of International Affairs titled “Foreign interference and Australian electoral security in the digital era.” It is republished with permission.