University of New South Wales Law Journal
|
|
Home
| Databases
| WorldLII
| Search
| Feedback
University of New South Wales Law Journal |
|
2021 Digital Democracy 959
ANIKA GAUJA*
The legal regulation of politics has, at its core, the aim of preserving three fundamental democratic values: liberty, equality and deliberative political participation. Yet, the reference point for examining these values is rooted in 19th and 20th century practices: where political campaigning and mobilisation follow terrestrial principles of organisation and regulation is undertaken by the state. Using the most recent empirical evidence drawn from political science on the changing nature of political participation and organisation, this article analyses the challenges of regulating digital politics. It argues that while the major focus of current interventions centres on political disinformation, this obscures more fundamental regulatory concerns such as capturing the diversity of new modes of participation and reconceptualising equality. While a model of co-regulation holds promise by institutionalising communities and individuals within technology companies’ policy decision-making processes, firms’ ultimate authority to define and control their user base presents challenges for effective participation.
Advances in digital technologies have profoundly impacted on human society, reshaping the way in which we communicate, socialise and practise politics. The internet is used daily by over four billion people worldwide, a significant proportion of whom access social media sites, such as Facebook, Twitter and Instagram. Australia ranks as one of the leading countries for internet use, with over 93% of adults now using the internet and/or owning a smartphone.1 In addition to facilitating daily activities, such as banking and shopping, digital technologies have enabled new forms of political participation and mobilisation. Social media has been credited with enabling protest movements such as the Arab Spring, the #EuroMaiden revolution in the Ukraine and the global #BlackLivesMatter
* Anika Gauja is Professor of Politics in the School of Social and Political Sciences, University of Sydney.
1 Jacob Poushter, ‘Smartphone Ownership and Usage Continues to Climb in Emerging Economies’, Pew Research Center (Web Page, 22 February 2016) <https://www.pewresearch.org/global/2016/02/22/ internet-access-growing-worldwide-but-remains-higher-in-advanced-economies/>.
960 UNSW Law Journal Volume 44(3)
protests, where citizens have been able to successfully organise through social media channels to achieve political change.2 Hashtags have also been used to mobilise citizens all over the world to advocate for certain causes: for example, the #NeverAgain movement for gun control in the United States (‘US’), the worldwide protest against workplace and domestic violence against women (#MeToo), and more recently in March 2021, rallies in capital cities and regional centres throughout Australia calling for an end to sexual assault and the harassment of women (#March4Justice).
The internet, and social media platforms in particular, have radically altered political participation and changed practices of mobilisation and organisation. The online environment is a space in which like-minded groups and individuals can be mobilised in traditional and non-traditional political activities, for example, by participating in political discussions, convincing others to vote or to take a stand on an issue, displaying political affiliations and showing support for causes, as well as donating and petitioning.3 As political communications scholar Francesco Bailo argues, the internet ‘revolutionises’ exposure to politics, not only by facilitating access to political information and communication, but by ‘bringing political contents through algorithmically determined filtering and recommendation systems to users who are not directly searching for them, something that every Facebook user is very familiar with’.4 The practice of politics on social media is characterised by its immediacy, the expectation that participants are able to see the effects of their actions, and it is driven by an adherence to specific interests or causes rather than movements or ideologies.5 In a highly significant break with traditional forms of political action, contemporary political participation and information provision are mediated by ‘big technology’, which I define in this article as digital technologies owned by corporate entities. Otherwise known as digital platforms, these entities play a fundamental role in modern politics and include Twitter, NationBuilder, Facebook, Instagram, WhatsApp, GoFundMe and Google. In contrast to the power of ‘big media’, digital platforms not only facilitate political debate and communication, but also provide accessible tools for users to organise and mobilise communities, including affordances such as the means to express support for campaigns and donate to causes.
2 See, eg, Ethan Zuckerman, ‘New Media, New Civics?’ (2014) 6(2) Policy and Internet 151, 151; Marcia Mundt, Karen Ross and Charla M Burnett, ‘Scaling Social Movements through Social Media: The Case of Black Lives Matter’ (2018) 4(4) Social Media + Society 1.
3 Brian D Loader, Ariadne Vromen and Michael A Xenos, ‘Introduction: The Networked Young Citizen’ in Brian D Loader, Ariadne Vromen and Michael A Xenos (eds), The Networked Young Citizen: Social Media, Political Participation and Civic Engagement (Routledge, 2014) 1, 1 (‘Introduction’); Michael A Xenos, Ariadne Vromen and Brian D Loader, ‘The Great Equalizer? Patterns of Social Media Use and Youth Political Engagement in Three Advanced Democracies’ in Brian D Loader, Ariadne Vromen and Michael A Xenos (eds), The Networked Young Citizen: Social Media, Political Participation and Civic Engagement (Routledge, 2014) 17, 21 (‘The Great Equalizer?’); Clay Shirky, Here Comes Everybody: The Power of Organizing without Organizations (Penguin Group, 2008).
4 Francesco Bailo, Online Communities and Crowds in the Rise of the Five Star Movement (Palgrave Macmillan, 2020) 4.
5 Zuckerman (n 2) 156–7.
2021 Digital Democracy 961
Researchers working in the field of political science have highlighted the potential of new communications technologies, not just for citizens, but also for political organisations and movements. They have analysed the emergence of many online groups, often labelled ‘digital natives’, which include both advocacy organisations such as GetUp and Change.org, as well as ‘new’ political parties such as the Pirate Party and Italy’s Five Star Movement. The changes identified by political science researchers also encompass the extent to which ‘traditional’ political parties and interest groups have embraced online platforms and digital technologies to perform campaigning, mobilisation and organising functions. For example, advocacy organisations such as MoveOn, GetUp and 38 Degrees all use a mix of digital media to create organisational infrastructure by organising members online, ascertaining their views and policy priorities and converting this information into the performance of asks or political activities.6 38 Degrees, a progressive not- for-profit online advocacy group based in the United Kingdom (‘UK’), claims to have facilitated over 39 million individual actions across 16,000 campaigns.7
While the nature of political participation, organisation and advocacy has changed profoundly, the laws that regulate these political activities have not. This has exposed significant policy challenges. The Cambridge Analytica scandal, in which personal data from 50 million US Facebook accounts were harvested and used without users’ knowledge to create targeted political advertising,8 has highlighted the need for legislative protections to safeguard against the unauthorised use of personal information for political purposes. The spread of ‘fake news’ and misinformation in political campaigning brings into question the law’s ability to provide for transparency and truth in political advertising in a fragmented media landscape. Social media companies, such as Facebook, have stepped in to regulate content, and Twitter has banned political advertising entirely.9 These examples illustrate the pressing need for the law to keep pace with how politics is practised today.
Governments and policymakers in Australia and internationally have struggled with the practical consequences of outdated legal regimes and the regulatory challenges posed by new forms of political participation and communication. A number of high-profile inquiries have recently been established to investigate these issues – for example, the Australian Parliament’s Senate Select Committee on Foreign Interference through Social Media (2019), the Australian Competition and Consumer Commission’s Inquiry into Digital Platforms (2019) and the Victorian Parliament’s Inquiry into the Impact of Social Media on Victorian Elections
6 See, eg, Ariadne Vromen, Digital Citizenship and Political Engagement: The Challenge from Online Advocacy and Campaigning Organisations (Palgrave Macmillan, 2017); Andrew Chadwick, The Hybrid Media System: Politics and Power (Oxford University Press, 2013) ch 9; Dave Karpf, The MoveOn Effect: The Unexpected Transformation of American Political Advocacy (Oxford University Press, 2012).
7 ‘Home’, 38 Degrees (Web Page) <https://home.38degrees.org.uk>.
8 Carole Cadwalladr and Emma Graham-Harrison, ‘Revealed: 50 Million Facebook Profiles Harvested for Cambridge Analytica in Major Data Breach’, The Guardian (online, 18 March 2018) <https://www. theguardian.com/news/2018/mar/17/cambridge-analytica-facebook-influence-us-election#>.
9 See ‘Ads about Social Issues, Elections or Politics’, Facebook (Web Page) <https://www.facebook.com/ policies/ads/restricted_content/political>; ‘Political Content’, Twitter (Web Page, 2021) <https://business. twitter.com/en/help/ads-policies/ads-content-policies/political-content.html>.
962 UNSW Law Journal Volume 44(3)
(2014).10 Yet, the empirical evidence base needed to guide critical changes to regulation is underdeveloped.
Disciplinary silos work against producing the kind of holistic analysis of contemporary political practice and its relationship with the law that is required to produce evidence to guide reform. The legal regulation of politics is an established field of research, having developed over time as a collection of parallel endeavours spanning numerous sub-fields in political science (party scholarship, interest group and social movement studies), political communication (social media studies) and the law (electoral, constitutional, administrative and charity law). While several key works bring these strands together,11 the developmental trajectory has struggled to combine both an understanding of the law’s reach with empirical insights into the everyday practice of politics and how this is evolving.
This article combines both political science and legal perspectives to examine the key challenges of regulating politics in the digital sphere. It does so by articulating the fundamental normative principles that have traditionally governed politics and elections in Australia and assessing their continued utility with reference to the latest developments in political communication, participation and organisation – drawing on the most salient and recent empirical research in these areas. The focus of the analysis is on digital politics driven by citizens rather than on more formalised or institutionalised modes of political engagement such as elections – although there are, as the analysis will show, important lessons to be learned in regulating digital politics from the formal – or traditional – political sphere.
I argue that despite changes in the technological context, the principles of liberty, equality and deliberation still provide a suitable normative framework for the legal regulation of new and emerging forms of political activity. However, I suggest that any legislative developments need to acknowledge the profound impact of technology – not just on political communication, but also on participation and organisation. To evolve with political practice and maintain its relevance, the law must also recognise contemporary practices of political participation and organisation, which in the modern era are becoming more and more separated from the formal institutions of representative democracy and increasingly mediated by big technology. Part II of this article outlines the principles of liberty, equality and deliberation, which together provide the normative basis for the regulation of politics online. Part III examines the contemporary characteristics of digital
10 Australian Competition and Consumer Commission, ‘Digital Platforms Inquiry’ (Final Report, June 2019) <https://www.accc.gov.au/system/files/Digital%20platforms%20inquiry%20-%20final%20report. pdf>; Electoral Matters Committee, Parliament of Victoria, Inquiry into the Impact of Social Media on Victorian Elections and Victoria’s Electoral Administration (Discussion Paper, August 2014) <https:// www.parliament.vic.gov.au/images/stories/committees/emc/Social_Media_Inquiry/Final_Discussion_ Paper_-_Inquiry_into_the_impact_of_social_media_on_Victorian_elections_and_Victorias_electoral_ administration.pdf>. The Senate Select Committee is due to present its final report by May 2022: see ‘Select Committee on Foreign Interference through Social Media’, Parliament of Australia (Web Page)
<https://www.aph.gov.au/Parliamentary_Business/Committees/Senate/Foreign_Interference_through_ Social_Media>.
11 See, eg, Graeme Orr, The Law of Politics: Elections, Parties and Money in Australia (Federation Press, 2nd ed, 2019) (‘The Law of Politics’); Nicolas P Suzor, Lawless: The Secret Rules That Govern Our Digital Lives (Cambridge University Press, 2019) (‘Lawless’).
2021 Digital Democracy 963
democracy (specifically, developments in political participation and organisation) and analyses their regulatory implications. Part IV evaluates three core models of governance: internal, external and co-governance.
To define what is ‘political’ and, therefore, to delineate the boundaries of appropriate legal regulation is an extremely difficult task. There is no discrete ‘law of politics’, although political concerns are present in constitutional, electoral and administrative law and beyond. Although there is no concrete definition of ‘political’ in the eyes of the law, political activities tend to be referenced to the state, and for the most part, to the institutions of representative democracy and government.12 In a somewhat broader conceptualisation, Ron Levy and Graeme Orr define the laws of politics as those which regulate the sites of political choice in a democracy, with a common focus ‘on how legal norms shape political decision-making by setting the ground rules, conditions for umpiring, and proper scope of debate among democratic participants’.13 The law’s conception of politics is, however, in large part synonymous with the field of election law and aligns with what political scientists call ‘Big P’ politics – that is, a focus on formal representative institutions. What is deemed to be ‘political’ intersects with the broader conceptual debate concerning public and private spheres. The private sphere is one of autonomy, where the law has a legitimate place facilitating decision-making but its primary purpose is protecting individuals and groups from the encroachment of others (including the state). By contrast, the public realm is a ‘world of government institutions, obliged to serve the public interest rather than private aims’.14 This distinction is crucial not only in defining the types of activity that might be seen as political – and therefore a legitimate subject of regulation – but appropriate
regulatory motivations and responses.
The public-private dichotomy is particularly important in terms of the regulation of digital politics as the distinction goes to the fundamental question of whether or not such regulation is desirable, the extent to which the state and public law should intervene in the activities of political actors and platforms, and which of these activities it should regulate. If we categorise social media spaces such as Facebook as ‘public’, regulating their activities may be normatively desirable, perhaps in order to implement particular democratic principles or outcomes (for
12 Politics and ‘the political’ are mentioned most commonly in texts and commentaries referring to elections: see, eg, Matthew J Streb, Law and Election Politics: The Rules of the Game (Routledge, 2nd ed, 2012); Martin Loughlin, The Idea of Public Law (Oxford University Press, 2004) 153. By analogy, this is also the conventional understanding of political activity in the context of anti-discrimination and media law, where protected political beliefs must relate to formal political institutions, public entities and contentious policy debates: see Lange v Australian Broadcasting Corporation [1997] HCA 25; (1997) 189 CLR 520.
13 Ron Levy and Graeme Orr, The Law of Deliberative Democracy (Routledge, 2017) 5.
14 Alan Freeman and Elizabeth Mensch, ‘The Public-Private Distinction in American Law and Life’ (1987) 36(2) Buffalo Law Review 237, 237.
964 UNSW Law Journal Volume 44(3)
example, equality of participation). If these spaces are characterised as ‘private’, state regulation may be seen as an undesirable intrusion upon the autonomy of these entities and an unnecessary interference with the political expression of citizens.
Appropriately mapping the distinction between public and private is problematic in the context of elections, let alone in online deliberative spaces that transcend this distinction. As Caroline Morris has noted with respect to electoral law:
A peculiar feature of the law’s approach to the questions which crop up at the various stages of the electoral process is a reluctance at times to acknowledge their public nature. Certain aspects of the electoral experience are characterised as private events: action over candidate selection is seen as akin to a private club dispute; the challenge to an election outcome, a contest between two private individuals.15
Changes in citizens’ preferences for political participation further complicate this distinction. As Pippa Norris argues, ‘traditional political activities that arose and flourished in industrial societies during the late nineteenth and early twentieth centuries are often thought to have peaked in the postwar era and waned in popularity today’.16 Political activity has become increasingly personalised and individualised, intertwined with life experiences and no longer firmly centred on the institutions of the state.17 This has necessitated a rethinking of the political in social science research:
[T]he political should be defined in such a way as to encompass the entire sphere of the social. The implication of this is that events, processes and practices should not be labelled ‘non-political’ or ‘extra-political’ simply by virtue of the specific setting or context in which they occur. All events, processes and practices which occur within the social sphere have the potential to be political and, hence, to be amenable to political analysis. The realm of government is no more innately political, by this definition, [than] that of culture, law or the domestic sphere.18
With this rethinking of the nature of politics comes an acknowledgement that political actors and frameworks, other than those sanctioned by the state, play a significant role in our everyday political lives and that our understanding of politics may be too focused on the state and the mechanics of elections to be truly meaningful in a 21st century context. It is here that ‘big technology’ and social media are paramount. Digital platform companies such as Twitter, Facebook and NationBuilder have an increasingly important role in politics by providing much of the infrastructure through which contemporary political participation occurs. While companies such as Twitter and Facebook can certainly be characterised as big technology by virtue of their market share, the significance of others – such as the online advocacy platform NationBuilder and crowdsourcing platform GoFundMe – is derived from their uptake as the organisational foundations of a new blueprint for political mobilisation.
15 Caroline Morris, Parliamentary Elections, Representation and the Law (Hart Publishing, 2012) 3.
16 Pippa Norris, Democratic Deficit: Critical Citizens Revisited (Cambridge University Press, 2011) 220–1.
17 See, eg, Zizi A Papacharissi, A Private Sphere: Democracy in a Digital Age (Polity Press, 2010); W Lance Bennett, ‘Changing Citizenship in the Digital Age’ in W Lance Bennett (ed), Civic Life Online: Learning How Digital Media Can Engage Youth (MIT Press, 2008); Russell J Dalton, Citizen Politics: Public Opinion and Political Parties in Advanced Industrial Democracies (CQ Press, 7th ed, 2019).
18 Colin Hay, Political Analysis: A Critical Introduction (Palgrave Macmillan, 2002) 3.
2021 Digital Democracy 965
Irrespective of whether politics takes places in the public realm, the private realm, or both, the laws that regulate political activity do not exist in a social vacuum. To be viable, any regulation must reflect the normative or cultural expectations of democracy that are shared by members of a society in such a way as to provide ‘good reasons for all participants ... to accept and abide by the outcome’.19 There is a consensus among legal and constitutional scholars that three basic, yet distinct, values underpin the laws of politics: liberty, equality and deliberation. There are, of course, tensions between each of these values insofar as it is difficult to maximise all three simultaneously, and because of disagreement as to which should be privileged, but together they provide a ‘democratic threshold’ from which the principles of regulation might be derived.20
Andrew Geddis has provided an illuminating discussion of these values, which he sees as distinct ‘conceptions of the electoral moment’.21 While this article broadens the discussion of politics beyond the electoral realm, these conceptions nonetheless potentially provide a fruitful basis for the creation of rules, institutions and practices that can produce legitimate political outcomes.22 Liberty’s central concern is ‘allowing all members of society to participate in the resolution of the tensions, disputes, and conflicts generated by the spontaneous nature of social life’.23 Freedom of participation necessitates the provision of rights, but these – as Geddis notes – are of a negative rather than positive nature: preventing exclusion from the political process rather than ‘a substantive ... assurance of some measure of effective participation’.24 The notion of equality provides the foundation for more proactive regulation in requiring that each participant’s interests be given equal consideration, and that each has ‘an effective opportunity to take part in, and attempt to influence’, the political process.25 Together, the notions of liberty and equality provide a set of normative principles that ‘express a foundational aspiration about human dignity and worth in which people have a voice, government is responsive to (competing) collective needs and interests, and society is not a mere marketplace dominated by the wealthy or a plaything of the powerful’.26
The value of deliberative participation builds on both liberty and equality, and emphasises ‘a republican ideal of inclusive, informed and perhaps even respectful discourse’,27 which is characterised by values such as cooperation and reflection to secure democratic inclusion, informed consent, an accurate reflection
19 Andrew Geddis, Electoral Law in New Zealand: Practice and Policy (LexisNexis, 2007) 19.
20 See, eg, Bruce E Cain, ‘More or Less: Searching for Regulatory Balance’ in Guy-Uriel E Charles, Heather K Gerken and Michael S Kang (eds), Race, Reform, and Regulation of the Electoral Process: Recurring Puzzles in American Democracy (Cambridge University Press, 2011) 263, 264.
21 Andrew Geddis, ‘Three Conceptions of the Electoral Moment’ (2003) 28 Australian Journal of Legal Philosophy 53.
22 Ibid 62.
23 Ibid.
24 Ibid 64.
25 Ibid 67 (emphasis in original). Note that the emphasis on participation is a notable feature of the majority’s understanding of political equality in McCloy v New South Wales [2015] HCA 34; (2015) 257 CLR 178.
26 Orr, The Law of Politics (n 11) 14.
27 Ibid.
966 UNSW Law Journal Volume 44(3)
of preferences and the avoidance of majority tyranny through agreement.28 Politics should be practised through ‘reasoned deliberation amongst participants committed to justifying their favoured solution to some common problem through public argument’.29 The public nature of debate is core to this ideal as it exposes participants to a range of arguments and perspectives that are integral to the collective nature of establishing consensus and which provide a justification for regulation that aims to preserve the integrity of this public discourse (in terms of quality, tone and content).30
In electoral law and cognate fields, these policies have come to be expressed in a number of concrete public policy responses and can be grounded in various debates over regulation. Liberty, for example, entails safeguarding freedoms of political communication and association while cautioning against the onerous over- regulation of civil society actors. As Graeme Orr argues, it ‘suggests a voluntary franchise, or at least a clear option to vote informally’, but also conflicts with the principles of political equality, which have led to the introduction of compulsory voting and expenditure and donation limits in political campaigns.31 The values of political deliberation find expression in transparency and authorisation requirements that seek to produce fully informed debate, enrolment and voter convenience initiatives (such as Saturday and online voting) that facilitate inclusion and participation, and regulations against false and misleading speech. Together, equality and deliberation demand policy responses and processes that are procedurally sound and fair, with decision-makers being able to be held to account for their decisions.
It is important to once again note the inevitable tensions that arise between these normative values and the difficulty of achieving agreement on the principles that should underlie political activity.32 While these tensions are discussed in the context of online politics, the primary focus of this article is to examine whether these broad principles are still relevant to the practice of 21st century politics, which necessitates their application to a broader understanding of the political. As abstract concepts, there is scholarly agreement – particularly among those who ascribe to a digital constitutionalism framework – that they can usefully be adapted from electoral and constitutional law to guide regulatory practices that are legitimate and grounded in publicly accepted principles.33 How they align with everyday politics in practice is examined in the following section.
28 Levy and Orr (n 13) 22–4, 26–7. See also Geddis, ‘Three Conceptions of the Electoral Moment’ (n 21) 70–4.
29 Geddis, ‘Three Conceptions of the Electoral Moment’ (n 21) 71.
30 Ibid. See also Graeme Orr, ‘The Law of Electoral Democracy: Theory and Purpose’ in Alan Bogg, Jacob Rowbottom and Alison L Young (eds), The Constitution of Social Democracy: Essays in Honour of Keith Ewing (Hart Publishing, 2020) 161, 173–4. The notion of integrity can be incorporated in the overarching principle of deliberation in the context of a broader participatory process beyond elections (as I do here), but it may also be used in a somewhat narrower sense to provide minimum standards for the conduct of elite-driven elections in realist accounts of democracy.
31 Orr, The Law of Politics (n 11) 15.
32 Orr, ‘The Law of Electoral Democracy: Theory and Purpose’ (n 30) 168, 175–6.
33 Suzor, Lawless (n 11) 112–13; Nicolas Suzor, ‘Digital Constitutionalism: Using the Rule of Law to Evaluate the Legitimacy of Governance by Platforms’ (2018) 4(3) Social Media + Society 1 (‘Digital Constitutionalism’).
2021 Digital Democracy 967
As noted in Part I, technological advances have profoundly affected the ways in which citizens practise politics. Digital politics embraces contemporary trends of political participation, whereby citizens seek ‘more flexible association with causes, ideas, and political organizations’ and more ‘personalized brands of politics organized around individual lifestyles and social networks’.34 Political scientists W Lance Bennett and Alexandra Segerberg conceptualise new digital media as an organisational driver that combines inclusiveness (through the ability to personalise actions) with technological openness, which allows citizens to determine the extent of their participation and facilitates connections between themselves and organisational actors.35 This enables individuals to get involved in campaigns with less effort, more quickly, and in accordance with their interests and abilities, and provides for an expansion of new democratic opportunities.36
In addition to democratic reinvigoration, key questions for internet and social media scholars that directly address the more proactive democratic values of deliberation and equality are the extent to which digital technologies facilitate participation and expand political efficacy, and whether or not they mitigate or replicate the political inequalities that exist offline.37 There is also varied opinion as to the impact of these changes on the quality of democracy and how it is practised. Michael Xenos and colleagues, publishing the outcomes of their cross-national project ‘The Civic Network’, note that the literature on the relationship between social media and political engagement has shown mixed results: from a strong and positive association, to the possibility that engagement is heightened among those who are already active, to studies that show no relationship at all.38 While opportunities have been identified by scholars linking new technologies to the renovation and rejuvenation of politics and democracy, others point to the potential of new technologies to develop echo chambers and perpetuate a growing ‘digital divide’ between citizens. Comparing online political participation amongst young people in Australia, the United Kingdom and the United States, the Civic Network
34 W Lance Bennett and Alexandra Segerberg, The Logic of Connective Action: Digital Media and the Personalization of Contentious Politics (Cambridge University Press, 2013) 5–6.
35 Ibid.
36 See Larisa Doroshenko et al, ‘Ukrainian Nationalist Parties and Connective Action: An Analysis of Electoral Campaigning and Social Media Sentiments’ (2019) 22(10) Information, Communication & Society 1376. Also note that personalised politics has brought about greater emphasis on political leaders, who also use social media and the affordances provided by digital technologies to appeal directly to citizens: see, eg, Gideon Rahat and Ofer Kenig, From Party Politics to Personalized Politics? Party Change and Political Personalization in Democracies (Oxford University Press, 2018).
37 Ariadne Vromen, Digital Citizenship and Political Engagement (Palgrave Macmillan, 2017) 52–3; Loader, Vromen and Xenos, ‘Introduction’ (n 3).
38 Xenos, Vromen and Loader, ‘The Great Equalizer?’ (n 3) 17–20.
968 UNSW Law Journal Volume 44(3)
Project found that social media use generated political engagement and reduced political inequalities over time, through a process of generational replacement.39
It is in the promise of greater political equality that a significant regulatory challenge for digital democracy arises. As previously discussed, the normative value of equality and the collective notion of democracy and politics that it encapsulates has shaped the regulation of politics and elections for decades – for example, in the principles of a universal franchise and the right to political participation. It has also played a central guiding role in the regulation of political finance, where donations and political expenditure in many Australian states and territories are restricted in order to provide a level playing field for actors in the electoral contest. In these instances, the notion of political equality is conceptualised through the prisms of political citizenship, which, more often than not, overlaps with national citizenship and economic resources.
By lowering the economic cost and the time required to participate, digital democracy and platform politics create a political environment where the cost of participation is lower for many citizens. However, it produces a new series of inequalities that the law – through its current focus on ameliorating inequality based on economic resources and citizenship – is not yet in a position to ameliorate. Certainly, economic inequality also affects political participation and access online, and as Katherine Dommett and Sam Power have noted in their analysis of UK campaign spending, current regulatory frameworks promoting financial transparency fail to adequately capture the true volume, source and target of money spent on digital political activities online.40 However, digital technologies also produce more fundamental political inequalities based on users’ digital political literacy, which are often tied to employment and socio-economic status, and affect an individual’s basic ability to engage socially and economically, and to participate in online political discussion and decision-making.41 To pursue the goal of political equality in the realm of digital politics, law reformers will need to engage with the various sources of power imbalance online.
The internet, facilitated by ‘big technology’ companies and platforms, has created a truly global marketplace for political ideas and discussion. There is great concern among many nation states as to the possibility of foreign interference in elections through social media, but another, less recognised challenge is to secure deliberative and participatory arenas for political debate and decision-making where one’s legitimacy as a participant does not rest on political citizenship alone. This is not just an abstract or theoretical challenge – it is a particularly acute issue at a time when many policy problems are global in their scale and transcend the boundaries of nation states (for example, climate change). Boycotts and buycotts as a form of political participation are now being organised online, and petitions have been reinvigorated in the digital landscape. Through social media and online
39 Ibid 18, 34.
40 Katharine Dommett and Sam Power, ‘The Political Economy of Facebook Advertising: Election Spending, Regulation and Targeting Online’ (2019) 90(2) The Political Quarterly 257, 258–9.
41 Peter Walton et al, ‘A Digital Inclusion: Empowering All Australians’ (2013) 1(1) Australian Journal of Telecommunications and the Digital Economy 9.1.
2021 Digital Democracy 969
petitioning platforms, for example Change.org, global citizens are expressing their voice and attempting to change policies and politics on a transnational scale.42 The target of this political action can be local, state or federal government, but also other ‘decision-makers’ such as corporations, public institutions (for example, hospitals and schools) and civic institutions (sporting clubs and non-government organisations), individuals and foreign governments.43 In their study of petitioning platform Change.org in Australia, Darren Halpin et al found that around half of the 17,045 petitions created between February 2012 and February 2017 were targeted at domestic governments, 18% at corporations, 9% at public and civic institutions, 5% at individuals, and 3% at foreign governments.44 Politics in the digital age is undertaken by global citizens, seeking to express concerns on global issues. If equality and the right to participation is based on state-centric notions of citizenship that privilege domestic electoral processes, this form of political expression may not be adequately protected. Indeed, changes to the Competition and Consumer Act 2010 (Cth) considered by the Australian Government to prohibit secondary boycotts for environmental protection in the aftermath of the success of the ‘Stop Adani’ movement45 provide an example of the fragility of political action when it is not directed at the state and/or channelled through the mechanism of elections.
One of the broader challenges that the law faces in protecting political participation in the digital era is to recognise not just the full scale of political activity and who it is directed at, but also what it means to participate in politics today. Politics can be a ‘meme’ or a video shared through social media, or it can be as simple as ‘liking’ a Twitter comment or Facebook page.46 Political participation scholars have debated the efficacy of these actions, based on the fact that these forms of engagement – often termed ‘clicktivism’ – are low-intensity, low-cost and relatively minor, particularly when compared to traditional political activities such as joining a political group, canvassing support or writing to a politician.47 However, perhaps what matters more than the form of these activities is their scale
– these are actions undertaken by billions – and that they are the preferred form of political participation for citizens otherwise disillusioned and disaffected with the traditional institutions of representative democracy.
42 David Karpf, Analytic Activism: Digital Listening and the New Political Strategy (Oxford University Press, 2016) (‘Analytic Activism’); Darren Halpin et al, ‘Online Petitioning and Politics: The Development of Change.org in Australia’ (2018) 53(4) Australian Journal of Political Science 428.
43 Karpf, Analytic Activism (n 42) 89; Halpin et al (n 42) 437.
44 Halpin et al (n 42) 437.
45 Paul Karp, ‘Government May Use Government Contracts to Crack Down on Environmental Protests’, The Guardian (online, 18 February 2020) <https://www.theguardian.com/australia-news/2020/feb/18/ coalition-may-use-government- contracts-to-crack-down-on-environment-protests>. See Competition and Consumer Act 2010 (Cth) s 45D.
46 See, eg, visual social media activity during the 2019 Australian federal election campaign: Lucien Leon, ‘Cartoons, Memes and Videos’ in Anika Gauja, Marian Sawer and Marian Simms (eds), Morrison’s Miracle: The 2019 Australian Federal Election (ANU Press, 2020) 473.
47 See, eg, James Dennis, Beyond Slacktivism: Political Participation on Social Media (Palgrave Macmillan, 2019); Max Halupka, ‘The Legitimisation of Clicktivism’ (2018) 53(1) Australian Journal of Political Science 130; Helen Margetts et al, Political Turbulence: How Social Media Shape Collective Action (Princeton University Press, 2015).
970 UNSW Law Journal Volume 44(3)
The implications for the law and the regulation of political activity are substantial. A rights-based approach to political expression and communication, which, for example, recognises and protects activities such as voting, protesting and undertaking political speech, must also now accommodate more ephemeral, ad hoc and fluid forms of political participation. In diffusing and decentralising political activity, the internet and digital technologies create infinite arenas (or ‘enclaves’ as Cass Sunstein would term them) for political participation and deliberation.48 The challenge for both political scientists and lawyers is to identify which of these fora impact on the political process and are vehicles for political expression, and therefore merit recognition, protection and regulation. Empirical studies have examined the importance of platforms such as Twitter for facilitating online debate during election campaigns and on salient political issues, organised and curated by hashtags such as #auspol and #ausvotes.49 Interestingly, research has also highlighted various ‘third spaces’ such as internet forums, where citizens can engage in ‘everyday political talk’ in addition to conversations related to the forum’s specific topic.50
Analysing parenting forums for political speech during the 2016 Australian federal election campaign, Scott Wright, Verity Trott and William Lukamto found that users regularly engaged in political discussion and that although disagreements occurred, they were successfully self-moderated by participants and the quality of deliberation was high.51 This research is particularly notable and interesting from a regulatory perspective because it suggests that the pressures for regulatory intervention are not equal across all deliberative enclaves on the internet. The same factors that account for civil and deliberate debate on parenting forums – for example, the creation of community through shared interests52 – may suppress disagreement, create political polarisation and arguably fuel extremism in other arenas.
Indeed, the provision of political information, and disinformation, online has been a key concern for regulators in Australia and overseas in recent years. A major issue that has arisen is whether we can characterise these new forums for the dissemination of information in a similar way to traditional media, and thus apply or extend existing regulations that seek to uphold the principles of transparency and informed debate. This has largely been the approach in Australia, where existing laws requiring an authorisation statement have been amended to capture electoral matter on websites and disseminated through social media.53 These provisions,
48 Cass Sunstein, Republic.com (Princeton University Press, 2001).
49 See, eg, Axel Bruns and Jean Burgess, ‘#ausvotes: How Twitter Covered the 2010 Australian Federal Election’ (2011) 44(2) Communication, Politics & Culture 37.
50 See Tim Highfield, Social Media and Everyday Politics (Polity Press, 2016); Todd Graham, Daniel Jackson and Scott Wright, ‘“We Need to Get Together and Make Ourselves Heard”: Everyday Online Spaces as Incubators of Political Action’ (2016) 19(10) Information, Communication & Society 1373.
51 Scott Wright, Verity Trott and William Lukamto, ‘The Election Online: Debate, Support, Community’ in Anika Gauja et al (eds), Double Disillusion: The 2016 Australian Federal Election (ANU Press, 2018) 475, 484–6, 490–1.
52 Ibid 490–1.
53 Electoral and Other Legislation Amendment Act 2017 (Cth); Electoral Legislation Amendment (Electoral Funding and Disclosure Reform) Act 2018 (Cth).
2021 Digital Democracy 971
however, capture only a relatively narrow range of political communications – those made by, or on behalf of an electoral actor, such as a political party or a candidate, or paid political advertisements. Social media content that is communicated for personal purposes, or to personal friends, does not fall within the scope of the regulations,54 nor is there any legislative requirement at the federal level for truth in electoral advertising, let alone political speech.55 While the Commonwealth Electoral Act 1918 (Cth) (‘Commonwealth Electoral Act’) makes it an offence to ‘print, publish or distribute, or cause, permit or authorize to be printed, published or distributed, any matter or thing that is likely to mislead or deceive an elector in relation to the casting of a vote’, this provision has been interpreted rather narrowly to apply to the casting of a vote rather than the formation of a political opinion.56
These legislative amendments have been accompanied by an information campaign, ‘Stop and Consider’, administered by the Australian Electoral Commission (‘AEC’), which was run during the 2019 Australian federal election. The campaign involved advertising material authorised by the AEC that urged voters to stop and consider the source of political information distributed on the internet and through social media. Evaluated by the AEC as largely successful, the campaign delivered 56 million social media impressions and more than 100,000 click-throughs to AEC material. Market research commissioned by the AEC found that 40% of those recognising the campaign claimed that they would take action on account of seeing it.57
Interestingly, in its material publicising the campaign and subsequent evaluations of it, the AEC downplayed the scale of concerns surrounding political misinformation in Australia. For example, a press release announcing the campaign with a statement from Electoral Commissioner Tom Rogers ended with the following: ‘Mr Rogers added that, thus far, there was no serious evidence of people or organisations seeking to disrupt Australian elections through disinformation, but given apparent events in other parts of the world, it was prudent to be vigilant’.58 This suggests that while political disinformation is undoubtedly a threat to democracy on a global scale, instances in Australia have not warranted the concern of electoral administrators, and does stand somewhat at odds with the concerns expressed by Australian politicians.
While misinformation is undoubtedly a legitimate concern and a primary object of attention for governments, the lack of a systematic evidence base indicating that
54 Commonwealth Electoral Act 1918 (Cth) s 321D; Commonwealth Electoral (Authorisation of Voter Communication) Determination 2018 (Cth).
55 Note, however, that truth in electoral advertising laws have been enacted in South Australia and the Australian Capital Territory. Section 113 of the Electoral Act 1985 (SA) creates an offence for an electoral advertisement to carry an inaccurate or misleading statement. In 2020, the Australian Capital Territory Legislative Assembly passed legislation to insert a similar provision into the Electoral Act 1992 (ACT): section 297A, as inserted by Electoral Amendment Act 2020 (ACT) section 13.
56 Commonwealth Electoral Act 1918 (Cth) s 329(1). See Evans v Crichton-Browne [1981] HCA 14; (1981) 147 CLR 169.
57 Australian Electoral Commission, ‘Stop and Consider Campaign: 2019 Federal Election’ (External Flyer) 2
<https://emailfooter.aec.gov.au/elections/electoral-advertising/files/stop-and-consider-external-flyer.pdf>.
58 Australian Electoral Commission, ‘AEC Encouraging Voters to “Stop and Consider” This Federal Election’ (Media Release, 15 April 2019) <https://www.aec.gov.au/media/media-releases/2019/04-15. htm>. See also ibid.
972 UNSW Law Journal Volume 44(3)
it is a prevalent practice with reference to elections in Australia, coupled with research that shows a capacity for moderation and deliberation in political debates in online spaces, suggests that we should be cautious in treating misinformation as the primary driver when thinking about how digital politics ought to be regulated in the future. At the very least, more empirical research needs to be undertaken on the scale and scope of mis- and dis-information activities. This caution is echoed by a recent report that systematically analysed the regulation of political misinformation online in Southeast Asia, which found
worrying effects of government over-reach through legislative responses to misinformation. The governments of both Singapore and Indonesia have been accused of misusing their anti-misinformation laws to censor or silence a wide spectrum of critics, including journalists, political dissidents and human rights campaigners, as well as religious leaders.59
As I have argued in this section, the expansion of political participation and the need to acknowledge, foster and protect diverse modes of political expression, as well as the need to conceptualise political equality as more than simply a disparity in financial resources, pose equally significant threats to democracy now and into the future.
Advances in digital technologies also hold promise for political organisations. Communications scholars Andrew Chadwick and Jennifer Stromer-Galley argue that social media and the internet are fertile grounds for organisational experimentation, enabling organisations such as political parties to harness the personalised forms of online engagement preferred by citizens to effectively aggregate interests and upscale their mobilisation efforts.60 Chadwick and Stromer-
Galley suggest that
it is not at all clear that political parties are dying. In fact, given the interactive effects we see between digital media, changes in citizens’ engagement repertoires, and parties’ organizational practices, the reverse may be true. In some cases, parties are renewing themselves from the outside in. Citizens are breathing new life into the party form, remaking parties in their own changed participatory image, and doing so via digital means. The overall outcome might prove more positive for democratic engagement and the decentralization of political power than has often been assumed.61
Alongside these lines of inquiry, researchers have also asked broader questions around the impact of these technological changes on the nature of political organisations and their ‘position in the democratic sphere’.62 Helen Margetts, for instance, hypothesised the positive impact new technologies would bring on enhancing
59 Andrea Carson, Fighting Fake News: A Study of Online Misinformation Regulation in the Asia Pacific
(Research Report, La Trobe University, January 2021) 85–6.
60 Andrew Chadwick and Jennifer Stromer-Galley, ‘Digital Media, Power, and Democracy in Parties and Election Campaigns: Party Decline or Party Renewal?’ (2016) 21(3) International Journal of Press/ Politics 283, 285.
61 Ibid.
62 Rachel Gibson and Stephen Ward, ‘Parties in the Digital Age: A Review Article’ (2009) 45(1)
Representation: Journal of Representative Democracy 87, 87.
2021 Digital Democracy 973
party competition in democracies by reducing organisational inequality by lowering the ‘start-up costs’ for minor political parties.63 Indeed, social media platforms have considerably reduced the organisational costs for political engagement by providing ready-made templates to mobilise supporters that can be effectively used with very little technical expertise. The ‘flip side’ of this convenience and ease is an organisation’s consequential reliance on a commercial provider for its organisational infrastructure. If this is removed – as highlighted by Facebook’s February 2021 blanket ban on news content in Australia that also captured many advocacy groups – the consequences for political organisations can be dire.
Digital parties (or platform parties as they are otherwise known, owing to their adoption of digital platforms and social media) are the new organisational template seen across new political formations in Europe, including the Pirate Parties of Northern Europe, left-wing populist parties such as Podemos in Spain and campaigning organisations such as Jeremy Corbyn’s Momentum (UK Labour). These diverse groups are all united by their promise ‘to deliver a new politics supported by digital technology; a kind of politics that ... professes to be more democratic, more open to ordinary people, more immediate and direct, more authentic and transparent’.64 A key consequence of this organisational form is that these political actors are essentially intertwined with technology companies and digital platforms as they utilise the ‘applications that have become the most recognisable mark of the present age, from social media like Facebook and Twitter, to messaging apps like WhatsApp and Telegram’.65 To maintain their relevance, political parties must find ways to engage individuals ‘in very personal ways: as consumers, animal and nature lovers, Facebook friends, Twitter followers, and self-styled global citizens who often prefer more direct ways of acting politically than voting or becoming formal members of organizations’.66 Digital parties, and other similar online advocacy organisations, are distinguished by the fact that they ‘bring digital transformation to their very core, to their internal structure of decision-making, rather than using digital communication simply as an outreach tool’.67
One prominent example of political organisations’ uptake of digital products and platforms to provide the basis of a digital political culture and infrastructure is the increasingly widespread use of NationBuilder. NationBuilder is an online community-organising software platform that enables parties and advocacy groups to build campaign sites and websites that incorporate communications, fundraising and volunteer/management profiling functions. It is a ‘leading nonpartisan platform used in the United States, Canada, the United Kingdom, and Australia’.68 Key features of NationBuilder sites include the ability to link to social media and specific issue campaigns, to enable users to easily donate or volunteer and to
63 Helen Margetts, ‘Cyber Parties’ in Richard Katz and William Crotty (eds), Handbook of Party Politics
(Sage Publications, 2006) 528, 530.
64 Paulo Gerbaudo, The Digital Party: Political Organisation and Online Democracy (Pluto Press, 2019) 4.
65 Ibid 4–5.
66 Bennett and Segerberg (n 34) 5. 67 Gerbaudo (n 64) 13–14.
68 Fenwick McKelvey and Jill Piebiak, ‘Porting the Political Campaign: The NationBuilder Platform and the Global Flows of Political Technology’ (2018) 20(3) New Media and Society 901, 901.
974 UNSW Law Journal Volume 44(3)
create databases of user activity.69 Supporters are invited to take immediate action through, for example, ‘liking’ a group’s campaign on social media, signing an online petition, donating to the cause or making phone calls.70
While the digital ‘affordances’ that platforms such as NationBuilder and Facebook offer, such as the ability to share and like content, sign petitions, donate and register support for causes, provide citizens with a myriad of opportunities to engage with political organisations, they also provide platforms and organisations with the ability to structure interactions and gather information on their supporters and users. In some instances, this is legitimate – for example, by bona fide parties or advocacy organisations – but in other instances, NationBuilder can be used ‘as a mobilisation tool for hate groups targeting cultural or ethnic identities; ... a profiling tool for deceptive advertising or stealth media ... [and] a fundraising tool for entrepreneurial journalism’.71 Research on the nature of modern-day campaigning websites and the digital infrastructure utilised by political organisations clearly demonstrates that these sites and platforms are no longer passive. Participation online is not like picking up a pamphlet: undertaking one activity can mean that details are harvested and individuals are being actively recruited into activities and movements. From a democratic perspective, this practice goes beyond the framing of digital rights in terms of privacy, and suggests a need to consider not only a right to equality of participation but also one of informed participation.
In an attempt to make politics more accessible and democratic, many digital parties employ digital tools to enable direct forms of decision-making in their organisations through online deliberation and policymaking, and the selection of leaders and parliamentary candidates.72 Also characteristic of these parties is the adoption of a much more flexible understanding of membership, where registering as a member is often as easy as signing up for Facebook or subscribing to an online mailing list.73 Some political parties, responding to evolving demands for more flexible and issue-oriented participation, have even dispensed with the notion of membership altogether, preferring to enlist the support of affiliates or ‘supporters’.74 The membership of political organisations, especially political parties, is one area in which the law has not kept pace with the increasing online presence of politics. Despite the changing nature of membership and what it means to join an organisation, the Commonwealth Electoral Act still requires that political parties demonstrate that they have 500 members in order to qualify for registration.75
69 Ibid 907, 910.
70 See, eg, ibid 910.
71 Fenwick McKelvey, ‘Cranks, Clickbait and Cons: On the Acceptable Use of Political Engagement Platforms’ (2019) 8(4) Internet Policy Review 1, 3.
72 See Gerbaudo (n 64); Bailo (n 4).
73 Gerbaudo (n 64) 17. See also Anika Gauja, ‘The Construction of Party Membership’ (2015) 54(2) European Journal of Political Research 232; Susan Scarrow, Beyond Party Members: Changing Approaches to Partisan Mobilization (Oxford University Press, 2015).
74 A notable example is the governing Liberal Party of Canada, which dispensed with formal membership in 2016. Similarly, the UK Labour Party opened its leadership selection process to registered supporters in 2014: see Anika Gauja, Party Reform: The Causes, Challenges, and Consequences of Organizational Change (Oxford University Press, 2017).
75 Commonwealth Electoral Act 1918 (Cth) s 126(2)(ca).
2021 Digital Democracy 975
While this threshold may not seem overly onerous, it should be examined in the context of a climate where party membership in Australia has fallen to around 1% of the electorate and where a far greater number of citizens prefer to engage with and express their support for parties through, for example, social media.76 If the goal of registration is to demonstrate a base level of political support for a legitimate cause so as to discourage vexatious political competitors, it should arguably not rely solely on this relatively outdated notion of partisan engagement. Reform options may include raising the numerical threshold of registration but incorporating a broader range of affiliation options, such as registered supporters, a base level of electoral performance or requiring the signatures of a minimum number of electors, which is a common approach in the United States.77
While the concepts of liberty, equality and deliberation – and the application to contemporary online politics – have been discussed, a key question that remains is who ought to decide on the framework of laws and regulations that ought to be in place and who should be responsible for enforcing it. In this Part, I discuss three broad models of regulatory governance: self-governance or internal regulation (where technology companies are responsible for regulating the activity that happens on their platforms), external governance (where platform activities are subject to state regulation) and co-governance (an arrangement that seeks to include broader public input on the development of rules and standards). The first two of these approaches lean toward ‘rules-based’ understandings of regulatory governance, whereas as the third focuses more on the development of ‘principles- based’ approaches.
In 2019, Facebook founder Mark Zuckerberg delivered an address to students at Georgetown University on freedom of expression in which he stated:
I understand the concerns about how tech platforms have centralized power, but I actually believe the much bigger story is how much these platforms have decentralized power by putting it directly in people’s hands.78
76 Narelle Miragliotta, ‘Parties and the Mass Membership’ in Narelle Miragliotta, Anika Gauja and Rodney Smith (eds), Contemporary Australian Political Party Organisations (Monash University Press, 2015) 62, 64. Note, however, that the civic participation statistics used by Miragliotta are from 2007 and
were extracted from a 2010 Australian Bureau of Statistics report: see Australian Bureau of Statistics, ‘Democracy, Governance and Citizenship: Civic Participation’, 1370.0 – Measures of Australia’s Progress, 2010 (Web Page, 13 November 2013) <https://www.abs.gov.au/ausstats/abs@.nsf/Lookup/ by%20Subject/1370.0~2010~Chapter~Civic%20participation%20(4.6.7.2)>. See also Anika Gauja and Max Grömping, ‘The Expanding Party Universe: Patterns of Partisan Engagement in Australia and the United Kingdom’ (2020) 26(6) Party Politics 822.
77 Anika Gauja, Political Parties and Elections: Legislating for Representative Democracy (Ashgate Publishing, 2010) 72–3.
78 Mark Zuckerberg, ‘Standing for Voice and Free Expression’ (Speech, Georgetown University, 18 October 2019). The text of this speech is reproduced in Tony Romm, ‘Zuckerberg: Standing for Voice and
Free Expression’, The Washington Post (online, 17 October 2019) <https://www.washingtonpost.com/ technology/2019/10/17/zuckerberg-standing-voice-free-expression/>.
976 UNSW Law Journal Volume 44(3)
Zuckerberg’s comments highlight the democratic potential of big technology and the internet, but they also draw attention to the fact that modern political activity is no longer predominantly focused on – and mediated by – the state. It increasingly occurs on platforms owned and mediated by social media companies, internet providers and digital platforms. Increasing scrutiny is placed on the role of these companies in the aftermath of public scandals and, as Robert Gorwa observes, ‘scholars, policymakers, and the public are increasingly attempting to understand the complex political effects of Instagram, Google Search, Airbnb, Uber, Amazon, and other platforms, amidst calls to make the companies that operate them more democratically accountable’.79
Responsibility for the regulation of digital politics is currently de facto shared by governments and big technology companies.80 Governments provide the legislative framework governing freedom of speech, racial vilification, electoral laws, and so on. However, as service providers, big technology companies are responsible for the mediation of content that appears on their platforms – giving them a great deal of power to control and facilitate political participation and debate. When users create an account or visit a website, they are bound by the terms of service contract issued by the provider, which articulates how platform content will be governed. As Nicolas Suzor argues, we need to view these terms of service in a similar way to how we might conceptualise legislative frameworks for the regulation of politics.81
Indeed, each of the major social media platforms based in Australia – Twitter, Facebook and Instagram – have publicly available terms of service, community standards documents and policies that outline what content is deemed acceptable or unacceptable. Facebook’s ‘Community Standards’ document, for example, prohibits hate speech and explains how this is implemented and arbitrated: ‘We also protect refugees, migrants, immigrants and asylum seekers from the most severe attacks, though we do allow commentary and criticism of immigration policies’.82 Twitter has published a ‘civic integrity policy’ that states the following:
The public conversation occurring on Twitter is never more important than during elections and other civic events. Any attempts to undermine the integrity of our service is antithetical to our fundamental rights and undermines the core tenets of freedom of expression, the value upon which our company is based.
We believe we have a responsibility to protect the integrity of those conversations from interference and manipulation. Therefore, we prohibit attempts to use our services to manipulate or disrupt civic processes, including through the distribution of false or misleading information about the procedures or circumstances around participation in a civic process. In instances where misleading information does not
79 Robert Gorwa, ‘What Is Platform Governance?’ (2019) 22(6) Information, Communication & Society
854, 854.
80 See Natali Helberger, Jo Pierson and Thomas Poell, ‘Governing Online Platforms: From Contested to Cooperative Responsibility’ (2018) 34(1) The Information Society 1.
81 Suzor, ‘Digital Constitutionalism’ (n 33) 2.
82 ‘Community Standards’, Facebook (Web Page, 2021) <https://www.facebook.com/ communitystandards/>. Facebook’s prohibition of hate speech in particular is located in Part III of this document, titled ‘Objectionable Content’.
2021 Digital Democracy 977
seek to directly manipulate or disrupt civic processes, but leads to confusion on our service, we may label the Tweets to give additional context.83
This model of governance, termed ‘self-regulation’ or ‘internal governance’, relies heavily on public and third party pressure to police standards and update policies to reflect community expectations.84 In Australia, this has been the preferred approach with the current development of voluntary industry codes of conduct by the Australian Communications and Media Authority and the Digital Industry Group Inc.85 Companies can be ‘nudged in the right direction’ without the need for legislative intervention, which offers flexibility and speed in dealing with events as they unfold86 – for example, Facebook’s actions to limit live streaming after the Christchurch massacre. A further advantage of this approach is that in keeping regulation in the hands of big technology, there is less concern that government regulations might be used to censor political debate. While this is not an immediate concern in Australia, government regulation has been used to undermine democratic political debate in nations throughout Southeast Asia.87
While the policies and terms of service of platform companies may reflect the normative values of liberty, equality and deliberation, mediating content in practice is more difficult to achieve. Platform companies are – at their core – companies, so commercial imperatives will inevitably drive behaviour. Doubts also exist as to whether transparency and governance initiatives resemble more than simple public relations exercises, resulting in systemic change and modified business models.88 For example, an Australian Broadcasting Corporation investigation conducted in 2019 using documents obtained under freedom of information laws revealed that the AEC had contacted Facebook to seek removal of the page ‘Hands Off Our Democracy’, which appeared not to comply with provisions of the Commonwealth Electoral Act concerning the authorisation of paid political advertising.89 The request was not complied with as both regulators could not reach agreement as to whether or not the page should carry authorisation information. Facebook eventually acknowledged that the page contained paid political content and was therefore subject to the provisions of the Commonwealth Electoral Act, but noted
83 ‘Civic Integrity Policy’, Twitter (Web Page, January 2021) <https://help.twitter.com/en/rules-and-policies/ election-integrity-policy>.
84 See Suzor, Lawless (n 11) 120–5 for an analysis of the role of civil society organisations in promoting governance principles.
85 See Australian Communications and Media Authority (‘ACMA’), ‘Misinformation and News Quality on Digital Platforms in Australia: A Position Paper to Guide Code Development’ (Position Paper, ACMA, June 2020); University of Technology Sydney Centre for Media Transition, ‘Discussion Paper on an Australian Voluntary Code of Practice for Disinformation’ (Discussion Paper, The Digital Industry Group Inc, 2020). The Digital Industry Group Inc (‘DIGI’) is a not-for-profit advocacy organisation representing Australia’s major digital platforms, including Google, Twitter, Facebook, eBay and GoFundMe: ‘About DIGI’, DIGI (Web Page) <https://digi.org.au/about/>.
86 Gorwa (n 79) 863.
87 See Carson (n 59). 88 Gorwa (n 79) 863.
89 Pat McGrath, ‘Facebook Probed by Australian Electoral Commission over Mysterious Political Ads’, ABC News (online, 26 February 2019) <https://www.abc.net.au/news/2019-02-26/facebook-electoral- commission-emails-reveal-political-ad-concern/10834736>.
978 UNSW Law Journal Volume 44(3)
that the ‘sponsored post attacking Get Up! and The Greens had received hundreds of likes and comments and had been shared widely prior to its disappearance from Facebook’.90
Part of the difficulty stems from a lack of clarity for regulatory responsibility and enforcement – the AEC subsequently formulated clear protocols for social media companies in responding to advertisements that it deemed to be illegal and threatened to seek injunctions if these protocols were not complied with.91 However, part also stems from the lack of transparency surrounding the decision-making process of platform companies themselves. Very little is known about how big technology companies such as Facebook and Twitter mediate political content and arbitrate disputes,92 despite their terms of service and community standards policies. The decisions of platform companies are not subject to the due process afforded by the rule of law – when they make ‘decisions about who uses their networks and how, they have almost unlimited discretion. They are accountable only to the market; there are no checks and balances on how they wield their power’.93
External governance is an alternative regulatory option. Under this model, governments harness policy levers in the areas of privacy and data protection, competition law and through the repudiation of intermediary liability protections. Amending existing laws to cover the digital sphere (for example, authorisation requirements for social media advertisements) is another example of this approach. The European Union (‘EU’) and several European states, notably Germany, have adopted this governance model with initiatives such as the EU’s General Data Protection Regulation.94 The German Network Enforcement Act compels platform companies to remove content violating German law, with the threat of substantial fines (up to €5 million).95 The difficulty of this approach is the speed of the law
– not only the legislative challenge of keeping up with continuously evolving advances in digital technologies, but also the relatively slow speed of responses to specific incidents and high costs if laws need to be enforced through the court system. As Suzor argues, ‘[t]hey’re too expensive and too slow to make a real dent in online abuse and hate, or copyright infringement, or many other problems that involve user-generated content on a massive scale’.96
90 Ibid.
91 Ibid.
92 Suzor, Lawless (n 11) 136; Nicolas Suzor, Tess Van Geelen and Sarah Myers West, ‘Evaluating the Legitimacy of Platform Governance: A Review of Research and a Shared Research Agenda’ (2018) 80(4) International Communication Gazette 385, 393–4.
93 Suzor, Lawless (n 11) 6–7.
94 Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016 on the Protection of Natural Persons with Regard to the Processing of Personal Data and on the Free Movement of Such Data, and Repealing Directive 95/46/EC (General Data Protection Regulation) [2016] OJ L 119/1.
95 Gesetz zur Verbesserung der Rechtsdurchsetzung in Sozialen Netzwerken (Netzwerkdurchsetzungsgesetz)
[Act to Improve Enforcement of the Law in Social Networks (Network Enforcement Act)] (Germany) 1 September 2017, BGB1 I, 2017, 3352, §§ 3, 4(2). See Wolfgang Schultz, ‘Regulating Intermediaries to Protect Privacy Online: The Case of the German NetzDG’ (Working Paper, Humboldt Institute for Internet and Society, January 2018).
96 Suzor, Lawless (n 11) 7.
2021 Digital Democracy 979
Co-regulation presents an intermediate approach. Rather than sharing regulatory responsibilities between platform companies and governments, co- regulation seeks to democratise aspects of the governance and management of big technology companies. Proposals range from third party organisations that might develop ethical and democratic standards for big technology, as well as investigate and mediate complaints, to democratic decision-making processes that feed into the policies of technology companies (such as user voting on policy changes).97 As Gorwa argues, ‘[c]o-governance could also provide even more radical options in the long term, as it lends itself to a philosophy that leads away from major, corporatized platforms and towards various platform cooperatives, decentralized systems, and other forms of community self-management’.98 These decentralised, community-driven modes of governance hold promise in that they leverage market and social norms to influence the policies and practices of big technology.99 In advocating for the development of robust ‘civil society organisations’ to hold big technology to account, Suzor suggests that ‘[u]nderstanding how social pressure can be brought to bear on digital intermediaries is the most important challenge for ensuring that tech companies implement changes and commit to governing their networks in more legitimate ways’.100
A rich body of research on organisational democracy, particularly examinations of intra-party democracy, highlight some of the advantages and problems with this approach.101 When users participate, democratic and community-based governance models can facilitate mass participation, raise feelings of political and policy efficacy and increase the legitimacy of organisations. However, these instances are now relatively rare. Even the digital parties described above have struggled to maintain consistently high rates of participation, leading to political decisions that have been endorsed only by the few.102 Indeed, this is the problem that Facebook faced when it experimented with ‘democratic governance’ from 2009 to 2012.103 Developing a ‘new [model] of governance’ and a ‘large scale democratic process’, Facebook opened policy changes to a vote of users. The first vote attracted only 665,654 participants, or 0.3% of its 200 million users at the time, falling well short
97 See, eg, Tarleton Gillespie, Custodians of the Internet: Platforms, Content Moderation, and the Hidden Decisions That Shape Social Media (Yale University Press, 2018).
98 Gorwa (n 79) 864–5. See also Gregory Lastowka, Virtual Justice: The New Laws of Online Worlds (Yale University Press, 2010) 195.
99 Lawrence Lessig, Code: Version 2.0 (Basic Books, 2006).
100 Suzor, Lawless (n 11) 121, 125.
101 See, eg, William P Cross and Richard S Katz, The Challenges of Intra-Party Democracy (Oxford University Press, 2013).
102 Davide Vittore, ‘Membership and Members’ Participation in New Digital Parties: Bring Back the People?’ (2020) 18(4) Comparative European Politics 609.
103 Adi Robertson, ‘Mark Zuckerberg Wants to Democratize Facebook: Here’s What Happened When He Tried’, The Verge (online, 5 April 2018) <https://www.theverge.com/2018/4/5/17176834/mark- zuckerberg-facebook-democracy-governance-vote-failure>, citing
Facebook, ‘Facebook Opens Governance of Service and Policy Processes to Users’, Facebook Newsroom (online, 26 February 2009) <https://about.fb.com/news/2009/02/facebook-opens-governance-of-service- and-policy-process-to-users/>.
980 UNSW Law Journal Volume 44(3)
of the 30% participation threshold for a binding policy decision.104 The experiment was subsequently abandoned.
A further difficulty concerns who can participate in the process. An assumption of the argument for greater public involvement and the democratisation of digital governance is that the public is defined in a similar way to its representation in formal democratic politics – based on established rights (citizenship) and not subject to arbitrary exclusion. However, big technology companies are able to define and redefine their ‘publics’ (or include/exclude users) with impunity. Regulators will need to think creatively about the justiciability of intra-platform decision-making processes. This is an arena where there may be some parallels with disputes within political associations and organisations, in which judicial precedent has developed over time to recognise justiciability based not just on the contractual rights of members, but on the public function performed by each organisation.
Digital platform and internet scholars have argued the importance of developing a principle-based, rather than rule-based, approach to regulation that grounds governance (whether internal, external or cooperative) in shared normative values.105 Digital constitutionalism, for example, offers an approach to regulation that seeks to articulate political rights, governance norms and limitations to the exercise of power online.106 Articulating these rights, norms and values is, however, a more contentious exercise. While previous scholarship has noted the theoretical tension between collective and individualist accounts of democracy and the associated difficulty of arbitrating between liberty, equality and deliberation as core democratic principles, this is also an empirical question. Referring to electoral democracy, Orr notes that ‘there are broad but competing goals that may express themselves differently in different political systems and contexts’.107 Yet, we do not know whether this diversity exists in relation to the practice of politics online, particularly given the fact that political participation transcends national boundaries. This is a question for both political science and legal scholars, which requires further comparative empirical research and reasoning.
I have argued that liberty, equality and deliberation – the principles that have underpinned the legislative and judicial regulation of elections – provide a good starting point for thinking through the regulation of digital activity. However, articulating the contemporary values that underpin modern politics and democracy requires an appreciation of current political practices, citizens’ attitudes and expectations, and an acknowledgment that the regulation of digital politics needs
104 Ibid.
105 Gorwa (n 79) 865; Pieter Nooren et al, ‘Should We Regulate Digital Platforms? A New Framework for Evaluating Policy Options’ (2018) 10(3) Policy and Internet 264.
106 Dennis Redeker, Lex Gill and Urs Gasser, ‘Towards Digital Constitutionalism? Mapping Attempts to Craft an Internet Bill of Rights’ (2018) 80(4) International Communication Gazette 302.
107 Orr, ‘The Law of Electoral Democracy: Theory and Purpose’ (n 30) 168.
2021 Digital Democracy 981
to extend far beyond the sphere of formal democratic institutions and traditional participatory practices. More than a decade of empirical research and political science scholarship reveals numerous contemporary trends in political participation and organisation that pose significant regulatory implications. The diversity of political organisation and practice necessitates moving beyond legislative protections that regulate digital activities with exclusive reference to elections. While the extension of legislation in this way is a start, it fails to recognise that the scale and nature of political participation extends well beyond these events and formal, ‘Big P’ politics that is centred on the institutions of government. Developing protections for equality, liberty and deliberation across a range of participatory spaces that transcend national boundaries, and encompass multiple issues and actors, is therefore necessary to safeguard democratic engagement. A second key regulatory imperative is to rethink and, if necessary, redefine what inequality means in the digital space.
A third is to facilitate, regulate and protect emerging forms of political organisation that no longer subscribe to traditional membership-based models and that rely very heavily on – and in some instances are entirely inseparable from – the organisational infrastructure provided by digital platforms and social media companies. While the proliferation of digital spaces for political deliberation may provide cause for an optimistic view of the health of digital participation, many of these spaces cannot be neatly categorised as either public or private. Providers such as Facebook take great pride in publicising the deliberative potential of their platforms, but, at the same time, exercise unfettered power to control access to their products. The suspension of the Facebook and Twitter accounts of former United States President Donald Trump is a case in point,108 as was the hiatus caused when media outlets had access to their news content suspended by Facebook when it and the Australian Government failed to reach an agreement on payment for news content in February 2021.109 These examples show how the digital realm not only challenges how we conceptualise the normative value of deliberation in an online context, but also the basic distinction between the public and private realms and the necessary justification that this distinction provides for the ‘intrusion’ of regulatory action.
But what model of regulation is best equipped to deal with these challenges? Of the three models of governance that were examined in this article (internal governance, external governance and co-regulation), digital constitutionalism and governance scholars see most potential in co-regulation, particularly in its ability to decentralise and democratise regulatory decisions – giving the public a say in
108 Mike Isaac and Kate Conger, ‘Facebook Bars Trump Through End of His Term’, The New York Times
(online, 18 May 2021) <https://www.nytimes.com/2021/01/07/technology/facebook-trump-ban.html>.
109 Facebook’s ban on posting news content to its pages was reversed several days later, but not before the action caused significant public disquiet and disruption. The ban also inadvertently caught information content from government authorities, emergency services and non-government organisations: Amanda Meade, Josh Taylor and Daniel Hurst, ‘Facebook Reverses Australia News Ban after Government Makes Media Code Amendments’, The Guardian (online, 23 February 2021) <https://www.theguardian. com/media/2021/feb/23/facebook-reverses-australia-news-ban-after-government-makes-media-code- amendments>.
982 UNSW Law Journal Volume 44(3)
the rules that govern them. Political science perspectives reveal, however, that many of the participatory traits and logics that drive citizens to online forms of engagement may also undermine the efficacy of this model as opportunities for voice and participation are not fully utilised. To avoid these pitfalls, co-regulation must be inclusive and careful attention given to who the ‘public’ is in different digital spaces. Is it enough to define the public as users of particular technologies, or should a broader group of participants be included? Or, should these decisions be left to those advocates and representatives with substantial interest and expertise in regulatory concerns? A potential way forward might involve a combination of both groups, with an inclusive approach to identifying principles, and then involve a more focused set of participants to articulate these principles as rules. In both instances, however, accountability and transparency are paramount to the legitimacy of regulatory decisions. The regulation of digital politics can be successfully negotiated, but it will require the same degree of thoughtful and principles-based deliberation in its process and choice of regulatory model as does the content of the regulations themselves.
AustLII:
Copyright Policy
|
Disclaimers
|
Privacy Policy
|
Feedback
URL: http://www.austlii.edu.au/au/journals/UNSWLawJl/2021/34.html