3. The Role of Technology
During the past decade there has been a remarkable shift in how the relationship between democracy and technology is perceived. In 2010, there was much discussion of what Larry Diamond, the editor of the Journal of Democracy, called ‘liberation technology’.30 New digital tools, he wrote, would empower citizens to ‘report news, expose wrongdoing, express opinions, mobilize protest, monitor elections, scrutinize government, deepen participation, and expand the horizons of freedom’. In short, digital technology was seen as being on the side of democracy, which it would help consolidate and deepen. The role of social media in the Arab Spring a year later seemed to confirm this view of digital technology as a liberating force.31
In recent years, however, this optimism has given way to pessimism as utopian thinking about digital technology has been replaced by dystopian thinking. In particular since the decision by the British people to leave the EU and the election of Donald Trump as US president in 2016, technology has become widely seen as a threat to democracy. Social media is now often described as a vehicle for disinformation rather than as a way to democratize information and expose the truth. It is also often seen as a factor in the rise of populism and as a cause of democratic deconsolidation. In short, digital technology is now widely seen as a driver of the crisis of liberal democracy.
This chapter explores whether this is really true. It examines various aspects and forms of digital technology that are currently having, or may in future have, an impact on democracy in Europe. Much of the debate about the role of digital technology in the current crisis focuses on the impact of social media. But it is widely accepted that the forms of digital technology that may already be having an impact on democracy in Europe are merely the beginning of a bigger revolution that could transform society in even more dramatic ways in the medium and long term – particularly through artificial intelligence (AI). The chapter therefore also examines ways in which other forms of digital technology could have an impact on democracy in Europe in the future, though this analysis is inevitably somewhat speculative.
Social media and the current crisis
Although the idea of ‘liberation technology’ was based mainly on the effect that social media seemed to be having in authoritarian states, it was also hoped that social media would act as a positive force in consolidated democracies by informing and engaging citizens. In particular, it was hoped that social media would ‘enable greater access to information, facilitate collective organizing, and empower civil society’.32 Social media has indeed enabled much greater participation in politics, particularly by young people.33 In that sense, it has fulfilled its promise. But at the same time, the way that social media undermines the ability of mainstream media organizations to control the flow of ideas and information – which in the early 2010s was identified as a positive change – is now widely seen as a problem.
In particular, it is widely believed that social media has contributed to the emergence of a ‘post-truth’ politics. It is argued that social media ‘increases volume and speed of information flow and the relative volume of opinion over fact’ and thus leads to what has been called ‘truth decay’.34 Social media ‘makes it possible for disinformation and misleading information to spread quickly and widely – and thus feeds Truth Decay by enabling the blurring of the line between information and fact and magnifying the relative volume and effect of opinions and beliefs over objective facts’.35 In short, social media is corroding political discourse. But the danger with such arguments is that they idealize an imaginary past in which ‘democracy and factual knowledge were uncontested’.36
It is also frequently claimed that social media is creating polarization. According to Jamie Bartlett, ‘information overload and connectivity has encouraged a divisive form of emotional tribal politics in which loyalty to the group and anger outrank reason and compromise’.37 This leads to the formation of ‘tribes’ who share a sense of struggle and grievance and leaders who are unable to reach the kind of compromises necessary to the functioning of liberal democracy. However, though these arguments about social media polarization have become widely accepted, they are based on limited evidence.38 Moreover, they are based on the assumption that polarization in politics is itself a problem. However, as discussed in the previous chapter, some argue that in some ways there has not been enough polarization in our politics in the past few decades.
Much commentary has also focused on social media’s use of algorithms to confirm user beliefs and biases and thus create ‘filter bubbles’ or ‘echo chambers’.39 For example, Bartlett writes that the ‘splintering of established mainstream news has allowed people to personalize their news sources in ways that play to their pre-existing biases’.40 Some research has concluded that users are more inclined to share content or opinions they agree with, which creates ‘selective exposure’.41 But more recent research challenges the idea of echo chambers and argues that social media actually increases exposure to alternative views.42 Some even argue that ‘the “echo chambers” narrative’ has itself been ‘amplified and distorted in a kind of echo chamber effect’.43
Whether or not social media corrodes political discourse, many see it as empowering particular political actors – i.e. social media changes the balance of forces within democracy rather than directly undermining it. In particular, social media is widely seen as having empowered populists, who, it is claimed, are particularly ‘adept at using social media to amplify their message, recruit and organise’.44 Yet there is little evidence, for example, that social media helped Trump win the presidential election in 2016.45 Moreover, a wide range of other figures, movements and parties – such as ‘Black Lives Matter’ in the US – have successfully used social media in much the same way. In other words, it is far from clear that social media particularly benefits populists more than other actors.
What may emerge from this era of online collective action is a ‘chaotic, turbulent pluralism’.
A more persuasive argument is that the effect of social media is to empower all kinds of outsiders rather than populists as such. Mounk argues that social media closes the technological gap between insiders and outsiders and thus reduces the advantage that elites previously had in shaping political discourse. Thus social media favours change over the status quo, and the forces of instability over the forces of order. ‘In empowering outsiders, digital technology destabilizes governing elites all over the world and speeds up the pace of change,’ he writes.46 Yet even this conclusion may be premature. After all, authoritarian governments are also increasingly using social media to increase social control and maintain ‘stability’. In particular, China is working with social media companies to develop its social credit system. Authoritarian governments may yet turn out to be more successful in their use of social media than those who seek to overthrow them.
A different way of thinking about the effect of social media is in terms of increased political turbulence. A group of four academics (two political scientists, a physicist and a computer scientist) have shown how social media ‘reshape the context in which citizens operate and influence their decisions about whether to participate politically’.47 What may emerge from this era of online collective action is a ‘chaotic, turbulent pluralism’.48 Social media enables citizens to mobilize more quickly than before – though it is difficult to predict how and when this will happen. Moreover, the movements that emerge from this activity are often leaderless and unsustainable.49 In short, the only thing we can say about social media with any certainty is that it increases uncertainty in politics.
Another critique of social media is that it is contributing to democratic deconsolidation. According to this view, people may be becoming impatient with democracy because it is too ‘analogue’. As Mounk writes: ‘The daily experience of liking and sharing posts on social media may habituate users to a simulated form of direct democracy that makes the existing institutions of representative democracy appear intolerably outmoded.’50 But if technology has in this sense contributed to this crisis of liberal democracy, it may also have a role in enabling its reconsolidation. For example, there has been much discussion of the potential of digital technology to facilitate democratic participation. Some have even seen the possibility of using it to scale up and update the kind of direct democracy that existed in ancient Athens – in effect creating a kind of ‘virtual agora’.51 This possibility is examined in more depth in Chapter 6.
A future threat to democracy?
Even if there is little conclusive evidence that digital technology has contributed to the current crisis of liberal democracy, this does not mean that it could not threaten democracy in the future. The remainder of this chapter examines possible ways in which the further development of digital technology could undermine the necessary preconditions for democracy in Europe to function. Some fears focus on the nature of the development of technology itself – particularly around its potential to manipulate human beings and undermine their ability to make informed, autonomous choices. Others have to do with the increasing power that technological change is giving to corporations, which are already taking on some of the functions of government – in particular law enforcement – but without democratic legitimacy.
Perhaps the most obvious way in which digital technology could undermine democracy is through the further development of computational propaganda. In particular, there is now much concern about what are termed ‘deep fakes’ – that is, ‘hyper-realistic digital falsification of images, video, and audio’.52 A number of such ‘deep fakes’ have already gone viral – such as an image appearing to show Emma González, a survivor of the February 2018 mass shooting in Parkland, Florida and a campaigner for gun control, tearing up the US constitution. The fear is that, as technologies for manipulating audiovisual content evolve further, it could become even more difficult for citizens to identify ‘fake news’. This could distort policy debates and even elections and erode trust in institutions.
There has also been much discussion of the way in which technology companies such as Facebook and Google carry out surveillance on users in order to predict and even manipulate their behaviour.53 The development of artificial intelligence (AI) based on big data and machine learning could take such manipulation even further.54 Governments have already begun to ‘nudge’ citizens towards healthier or more environmentally friendly behaviour.55 But what is sometimes called ‘big nudging’ – that is, the combination of big data with nudging – could in future be used to control political preferences in ways that would undermine democracy. As a group of European computer scientists and academics have warned: ‘The more is known about us, the less likely our choices are to be free and not predetermined by others.’56
Some thinkers like Yuval Noah Harari go even further and speculate that the development of AI could ultimately lead to the end of the human agency upon which democracy depends.57 ‘Together, infotech and biotech will create unprecedented upheavals in human society, eroding human agency and, possibly, subverting human desires,’ he writes. ‘Under such conditions, liberal democracy and free-market economics might become obsolete.’ In particular, he argues, AI could lead to a shift in the relative efficiency of democracies and dictatorships. ‘The main handicap of authoritarian regimes in the 20th century – the desire to concentrate all information and power in one place – may become their decisive advantage in the 21st century.’58
Even if democracies survive this competition, Harari argues, they could become increasingly hollowed out as the advance of AI leads citizens to delegate decisions to intelligent machines. ‘We might willingly give up more and more authority over our lives because we will learn from experience to trust the algorithms more than our own feelings, eventually losing our ability to make many decisions for ourselves,’ he writes.59 A hint of how this might work can already be seen in voting advice applications such as StemWijzer in the Netherlands and Wahl-O-Mat in Germany, which advise users on who to vote for based on their preferences or views on particular policy issues. As AI progresses further, it is possible to imagine citizens outsourcing voting choices entirely to digital tools.
Democracies could become increasingly hollowed out as the advance of AI leads citizens to delegate decisions to intelligent machines. Indeed, it is possible to imagine citizens outsourcing voting choices entirely to digital tools.
A final, more indirect way in which technology could challenge democracy in future is through the effect of automation on work and thus on income inequality. Automation promises to increase productivity and efficiency and to create new jobs. But it also threatens huge disruption that could displace workers and lead to changes in the occupational, sectoral and skills-based composition of employment, as well as in the level and distribution of wages. In the worst-case scenario, the result could be a kind of ‘jobs apocalypse’ in which many jobs become obsolete.60 In particular, as machines become more adept at cognitive as well as manual tasks, the next wave of automation could lead to the loss of service sector jobs much as the previous wave hit manufacturing sector jobs.
There is little consensus about how many and which jobs are vulnerable to the next wave of automation. An influential 2013 study predicted that as many as 47 per cent of all jobs in the US were at risk from ‘computerisation’.61 More recently, others have argued that studies such as these, which rely on an ‘occupation-based approach’, overestimate the threat from automation. A 2016 study for the Organisation for Economic Co-operation and Development (OECD) that focused on specific tasks rather than occupations suggested that only around 9 per cent of jobs in advanced economies are ‘automatable’.62 The impact of automation is also likely to differ from one European country to the next, depending on the structure of each economy and the vulnerability of specific sectors within it.
The next wave of automation is relevant to the future of democracy in Europe because of its potential to widen inequality. A certain degree of economic equality is commonly seen as a precondition of democracy – which is why levels of inequality in many European countries are such a problem. This connection between equality and democracy goes back to the Greek concept of isonomia, or ‘equality of political rights’.63 The degree and kind of equality needed for a functioning democracy are debatable – different countries within Europe may have different levels of tolerance of inequality. But if inequality grows too much, democracy risks becoming plutocracy.
In this sense, the question is not so much how many jobs may be eliminated by automation but what kinds of jobs – and in particular, whether automation will tend to disproportionately replace worse-paid jobs or also affect better-paid jobs. Although the 2016 OECD study concluded that automation was unlikely to destroy large numbers of jobs, it also predicted that ‘low qualified workers are likely to bear the brunt of the adjustment costs as the automatibility of their jobs is higher compared to highly qualified workers’.64 Thus, the authors argue, the challenge for the future is to cope with the rising inequality that will result. In the worst case, automation could lead to a new ‘useless class’.65 Some argue that this will necessitate a universal basic income in order to maintain social cohesion.
However, it is also possible that the next wave of automation could have a flattening effect on income scales because some highly paid jobs could be more vulnerable to automation than lower-paid ones. One German entrepreneur even predicts that in future nurses will get paid more than doctors because their function will be harder to replace by algorithms and robots.66 This example is speculative, but it illustrates the idea that automation could affect jobs in complex and surprising ways. Rather than increasing inequality as the previous wave of automation did, the next wave of automation could actually decrease it. As with other technological developments, the potential impact of automation on democracy in Europe is uncertain.
In short, there is much we do not know – and cannot know – about how technology might affect democracy in the future. Some developments may affect Europe differently from the US or other parts of the world. Much will depend on the crucial issue of how technology is regulated – which may also be different in Europe compared to other jurisdictions. The EU has sought to take the lead in creating consumer-oriented rules to control digital businesses through a framework based on ‘European values’. But it remains to be seen how much influence this will have, particularly if technological developments are in future dominated by US and Chinese companies.67 Chapter 7 of this research paper returns to the issue of regulation.
30 Diamond, L. (2010), ‘Liberation Technology’, Journal of Democracy, Volume 21, Number 3, July 2010, pp. 69–83, https://www.journalofdemocracy.org/wp-content/uploads/2012/01/Diamond-21-3.pdf.
31 The term ‘social media’ is used in this paper as shorthand for all kinds of internet-based platforms that allow users to share content, including shared-news websites, photo- and video-sharing applications and websites, and closed messaging applications.
32 Deibert, R. (2019), ‘The Road to Digital Unfreedom: Three Painful Truths About Social Media’, Journal of Democracy, Volume 30, Number 1, January 2019, pp. 25–39, here p. 25, https://www.journalofdemocracy.org/articles/the-road-to-digital-unfreedom-three-painful-truths-about-social-media/.
33 See Boulianne, S. and Theocharis, Y. (2018), ‘Young People, Digital Media, and Engagement: A Meta-Analysis of Research’, Social Science Computer Review, 5 December 2018.
34 Kavanagh, J. and Rich, M. D. (2018), Truth Decay: An Initial Exploration of the Diminishing Role of Facts and Analysis in American Public Life, Santa Monica: RAND Corporation, pp. 110–21.
35 Ibid., p. 113.
36 Müller, J.-W. (2019), ‘The Pre-History of Post-Truth’, Project Syndicate, 26 April 2019, https://www.project-syndicate.org/onpoint/the-pre-history-of-post-truth-by-jan-werner-mueller-2019-04.
37 Bartlett, J. (2018), The People vs. Tech. How the Internet Is Killing Democracy (and How We Save It), London: Ebury, Kindle edition, location 398–401.
38 Some data suggest that ‘affective polarization’ has actually decreased in some European countries in recent decades, including in Germany, Norway, Sweden and the UK. See Boxell, L., Gentzkow, M. and Shapiro, J. (2020), ‘Cross-Country Trends in Affective Polarization’, National Bureau of Economic Research Working Paper No. 26669, January 2020, https://www.nber.org/papers/w26669#fromrss.
39 The term ‘filter bubble’ was first used by Eli Pariser. See Pariser, E. (2012), The Filter Bubble: How the New Personalized Web Is Changing What We Read and How We Think, New York: Penguin. The term ‘echo chamber’ was first used by Cass Sunstein. See Sunstein, C. (2001), Republic.com, Princeton: Princeton University Press.
40 Bartlett (2018), The People vs. Tech, location 444.
41 Kelly Garrett, R. (2009), ‘Echo Chambers Online?: Politically Motivated Selective Exposure among Internet News Users’, Journal of Computer-Mediated Communication, Volume 14, Issue 2, January 2009, pp. 265–85.
42 Blank, G. and Dubois, E. (2019), ‘The myth of the echo chamber’, Oxford Internet Institute, 9 March 2018, https://www.oii.ox.ac.uk/blog/the-myth-of-the-echo-chamber/; Blank, G. and Dubois, E. (2018), ‘The echo chamber is overstated: the moderating effect of political interest and diverse media’, Information, Communication and Society, Volume 21, Issue 5, pp. 729–45; and Flaxman, S., Goel, S. and Rao, J. M. (2016), ‘Filter Bubbles, Echo Chambers, and Online News Consumption’, Public Opinion Quarterly, Volume 80, Special Issue, January 2016, pp. 298–320.
43 Guess, A., Lyons, B., Nyhan, B. and Reifler, J. (2018), Avoiding the echo chamber about echo chambers: Why selective exposure to like-minded political news is less prevalent than you think, Miami: Knight Foundation, https://kf-site-production.s3.amazonaws.com/media_elements/files/000/000/133/original/Topos_KF_White-Paper_Nyhan_V1.pdf.
44 Bartlett, J., Birdwell, J. and Littler, M. (2011), The New Face of Digital Populism, London: Demos, p. 15, https://demosuk.wpengine.com/files/Demos_OSIPOP_Book-web_03.pdf?1320601634.
45 Sides, J., Tesler, M. and Vavreck, L. (2018), Identity Crisis: The 2016 Presidential Campaign and the Battle for the Meaning of America, Princeton: Princeton University Press, p. 194. See also Groshek, J. and Koc-Michalska, K. (2018), ‘Helping populism win? Social media use, filter bubbles, and support for populist presidential candidates in the 2016 US election campaign’, Information Communication and Society, Volume 20, Issue 9, pp. 1–19.
46 Mounk (2018), The People vs. Democracy, p. 149.
47 Margetts, H., John, P., Hale, S. and Yasseri, T. (2016), Political Turbulence: How social media shape collective action, Princeton: Princeton University Press, p. 11.
48 Ibid., p. 32.
49 Ibid.
50 Mounk, Y. (2018), ‘Can Liberal Democracy Survive Social Media?’, New York Review of Books, 30 April 2018.
51 Mounk (2018), The People vs. Democracy, pp. 57–58.
52 Chesney, R. and Citron, D. K. (forthcoming), ‘Deep Fakes: A Looming Challenge for Privacy, Democracy, and National Security’, California Law Review, https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3213954.
53 Zuboff, S. (2019), The Age of Surveillance Capitalism: The Fight for a Human Future at the New Frontier of Power, New York: PublicAffairs. For a critical review of Zuboff’s book, see Morozov, E. (2019), ‘Capitalism’s New Clothes’, The Baffler, 4 February 2019, https://thebaffler.com/latest/capitalisms-new-clothes-morozov.
54 See Raines, T. (2019), ‘Democratic disruption: technological change’, SITRA, 1 April 2019, https://www.sitra.fi/en/articles/democratic-disruption-technological-change/.
56 Helbing, D., Frey, B. S., Gigerenzer, G., Hafen, E., Hagner, M., Hofstetter, Y., van den Hoven, J., Zicari, R. V. and Zwitter, A. (2017), ‘Will Democracy Survive Big Data and Artificial Intelligence?, Scientific American, 25 February 2017, https://www.scientificamerican.com/article/will-democracy-survive-big-data-and-artificial-intelligence/.
57 Harari, Y. N. (2018), ‘Why Technology Favors Tyranny’, The Atlantic, October 2018, https://www.theatlantic.com/magazine/archive/2018/10/yuval-noah-harari-technology-tyranny/568330/.
58 Ibid.
59 Ibid.
60 Cukier, K. (2019), ‘The Economic Implications of Artificial Intelligence’, in Cummings, M. L., Roff, H., Cukier, K., Parakilas, J. and Bryce, H. (2018), Artificial Intelligence and International Affairs, Chatham House Report, London: Royal Institute of International Affairs, pp. 29–42, here p. 34, https://www.chathamhouse.org/publication/artificial-intelligence-and-international-affairs.
61 Frey, C. B. and Osborne, M. (2013), The Future of Employment: How Susceptible Are Jobs to Computerisation?, Working Paper, Oxford: Oxford Martin Programme on Technology and Employment, https://www.oxfordmartin.ox.ac.uk/downloads/academic/future-of-employment.pdf.
62 Arntz, M., Gregory, T. and Zierahn, U. (2016), The Risk of Automation for Jobs in OECD Countries: A Comparative Analysis, OECD Social, Employment and Migration Working Papers, No. 189, Paris: OECD Publishing.
63 See Nolte (2012), Was ist Demokratie?, pp. 35, 434.
64 Arntz, Gregory and Zierahn (2016), The Risk of Automation for Jobs in OECD Countries.
65 Harari (2018), ‘Why Technology Favors Tyranny’.
66 Cited in Cukier (2019), ‘The Economic Implications of Artificial Intelligence’, p. 37.
67 See Helgadóttir, A. G. (2019), ‘Will Europe scare off Silicon Valley’, The World Today, August/September 2019, https://www.chathamhouse.org/publications/twt/will-europe-scare-silicon-valley.