Digital media, political polarization and challenges to democracy

Today democracy is under attack and there are many points in which it is weak and vulnerable to those attacks. In almost every democracy citizens are losing confidence and trust in parties and in governments. There are declines in news audiences and press readership. In Europe only half of the citizens born after 1980 think that it is essential to live in a democracy.

It was with these reflections that W. Lance Bennett (University of Washington) opened his public lecture ‘Who Are the People’ (Bennett, 2017a), which was broadcast by the Austrian Broadcasting Corporation (ORF) as part of an international symposium on ‘Digital Media, Political Polarization, and Challenges to Democracy’ hosted by the Austrian Academy of Sciences and the University of York.

The same symposium also gave rise to this special issue of ‘Information, Communication and Society’, which consists of six selected articles based on papers presented at the conference. Held at the Austrian Academy of Sciences in Vienna from 21 to 22 September 2017 the symposium brought together social scientists from 20 countries and four continents to address the complex interplay between democracy, political participation, political polarization, and the new media environment or as Michael X. Delli Carpini (Walter H. Annenberg Dean at the Annenberg School for Communication, University of Pennsylvania) has termed it, the new ‘media regime’. In his keynote speech, Delli Carpini argued that, from a historical perspective, distinct ‘media regimes’ emerge over time, on the back of economic, political, cultural, and technological changes, and that each regime is associated with different notions of free speech, a free press, democratic citizenship, civic engagement, and the responsibility to participate in the political process. The problem is that we are familiar with the rules and norms of the previously dominant broadcast-based regime, we cannot know with any certainty what is most beneficial or most problematic about the emerging new regime (Williams & Delli Carpini, 2011). While the conference itself had a much broader focus, encompassing a range of topics from theoretical reflection on the changing dynamics of democratic processes and political communication to empirical studies on the radicalization of public debates and campaigns, nationalism, (right-wing) populism, and hate speech, the present special issue focuses on the interplay between social media, political polarization, and civic engagement. It also considers what needs to be done to ensure the democratic character of the new, and thus far largely unknown media regime and explores its potential to enhance and undermine democracy.

The conference was marked by a unifying, invigorating, and captivating sense of a new departure, underlined by a mutual feeling of social responsibility. In this spirit, the various theoretical reflections and empirical studies presented at the conference ultimately revolved around the highly topical and urgent question, ‘What is to be done?’ However fragmentary any answer to this question must be, as W. Lance Bennett noted, it will on the one hand have to set out from the fact that political polarization cannot be grasped in isolation from its entanglement with neoliberal interests; on the other, it will have to address the overarching question, ‘Who are the people?’ i.e., the question of democratic sovereignty (Bennett, 2017b) . This question calls for far more than a simple examination of populism as today’s allegedly central political problem. As Bennett affirms, ‘It’s rather a question that gets to the actual survival of democracy itself’ (Bennett, 2017a). Against this backdrop, the conference saw discussions of a range of approaches based primarily on analyses of the dynamic interplay between changing conceptions of democracy, changing forms of political participation, and changing media usage behaviour. There is a clear evidence of a transition toward a more participatory, commons-based understanding of citizenship, particularly among young people, a transition that serves to shift the focus from ensuring system performance (via voting and party membership) to supporting the system as such. This transition needs to be understood in relation to the ways in which digitally mediated communication influences – and perhaps organizes – our personal relationship to society and public life (Bennett & Segerberg, 2012). Political actors and researchers therefore have to confront the challenge of how to manage this contemporary shift from established structures of representative democracy – with their group-based identities and mainstream issues – toward a wider repertoire of personalized political experience and individual motivated concerns within a digitally networked society marked by an increasingly participatory understanding of democracy (Seethaler & Beaufort, 2017).

The present issue aims to reflect the various multidisciplinary perspectives that shaped the conference discussions. It contains both innovative theoretical interventions and empirically grounded research that addresses the interplay between social media, political polarization, and civic engagement in very different countries with varying degrees of democratization, diverse media environments, and different cultural specificities. Both the three Western and three non-Western contributions to the issue share a common aspiration to understand the (anti-) democratic potential of the new media environment.

The issue opens with a much-discussed topic in current social debates. In their proof of concept study, ‘Parametrizing Brexit: Mapping Twitter Political Space to Parliamentary Constituencies’, Marco Bastos and Dan Mercea aim to validate the use of social media signals as a means of modelling the ideological coordinates of the Brexit debate. They find that their model explains 41% of the variance in the referendum vote and show that the referendum outcome can primarily be explained by outrage at material inequality and a nationalistic shift. The authors conclude with a discussion of the conceptual and methodological challenges involved in signal processing social media data to measure public opinion. As they see it, the strength of their approach lies not so much in its capacity to generate predictions, but in the possibilities, it offers for the exploration of ongoing developments, which would otherwise require the extensive, continuous, and expensive use of traditional survey methodologies. While this first article focuses on the methodological possibilities of the ongoing analysis of social media content in the social science, the authors of the following two contributions focus on this content itself. They consider the extent to which the rise of social media as a dominant force in news distribution has led to the stronger isolation of different outlooks through filter bubbles from algorithmic filtering and echo chambers that arise in information environments dominated by like-minded people, as well as assessing the consequences of such developments.

In ‘Facebook News and (De)Polarization: Reinforcing Spirals in the 2016 US Election’, Michael Beam, Myiah Hutchens, and Jay Hmielowski offer a theoretical and empirical framework for the relationship between social media usage, partisan news exposure, and attitude formation (including affective polarization, depolarization, and homeostasis), by revealing a depolarization process that obeys a model of reinforcing spirals. In order to move beyond correlational relationships, they use 3-wave panel data from the 2016 US Presidential campaign, to show that Facebook message use is related to a moderate over-time depolarization spiral and that people who use Facebook for news are likely to be confronted with both pro-attitudinal and counter-attitudinal messages. They find no evidence of a partisan reinforcing spiral that would lead to increased affective polarization. On the basis of previous research that found that people are more likely to engage in counter-attitudinal exposure and information processing when they receive personalized social and algorithmic recommendations (Messing & Westwood, 2014), the authors contend that the depolarization effects they observed are primarily driven by users encountering counter-attitudinal news, especially incidentally, on Facebook. The study therefore indicates that Facebook news may not be responsible for rising citizen polarization. Furthermore, the findings debunk the theoretical links between social and algorithmic news recommendations on social media. This is also the starting point for the following study, which considers the extent to which the algorithms used to generate news recommendations allow for content diversity. In ‘Do Not Blame It on the Algorithm: An Empirical Assessment of Multiple Recommender Systems and their Impact on Content Diversity’, Damian Trilling, Judith Möller, Natali Helberger and Bram van Es examine the effect of multiple recommender systems on different diversity dimensions and find that all of the recommendation logics in question generate a rather diverse set of recommendations that are on a par with those made by human editors. They also show that basing recommendations on user histories can substantially increase topic diversity within a recommendation set. On the one hand, then, the results of both studies speak against the idea that different political outlooks have become more strongly isolated from one another due to filter bubbles and echo chambers. On the other hand, they point to very specific filtering and acquisitional patterns in use among social networks, which fundamentally differ from those in traditional media environments. These differences should not be underestimated in terms of their impact on engagement and mobilization and require further research.

They can also be seen in the fact that online social networks are regarded as more supportive environments for women than traditional media environments, where previous research has shown that male politicians receive greater coverage (Bystrom, Robertson, & Banwart, 2001; Joiner et al., 2016). The extent to which these newer environments can also help to challenge long-held notions concerning the media inferiority of female politicians is considered by Moran Yarchi and Tal Samuel-Azran in their ‘Women Politicians Are More Engaging: Male Versus Female Politicians’ Ability to Generate User Engagement on Social Media During an Election Campaign’. Yarchi and Samuel-Azran examine whether social media offer a more equal platform for female politicians to generate user engagement and mobilize their supporters than traditional media. Taking the 2015 Israeli election campaign as a case study, the authors examine female and male politicians’ social media posts with respect to their content and the number of ‘likes’ and ‘shares’ they received. They conclude that social media offer greater opportunities for female politicians to promote themselves and raise their status in the political power-play. While this study focuses on the potential of social media in political public relations and voters’ mobilization, the next contribution addresses their role in facilitating bottom-up social engagement.

Do different social media usage types correspond to different forms and levels of engagement and mobilization? In their ‘Commitment in the Cloud? Social Media Participation in the [Taiwanese] Sunflower Movement’, Yuan Hsiao and Yunkang Yang investigate this question in relation to the commitment of social media participants in digitally enabled protests. They find that the commitment of so-called ‘cloud activists’ (who operate exclusively online during a movement) is at least on par with most offline participants with respect to their identification with the movement and their efficacy within it. This finding contradicts the traditional notion that only offline participation involves a high level of commitment and suggests that commitment can be exhibited in online forms through sustained or highly intensive organizing activities that may enable (offline) movements to persist. The activities of ‘cloud activists’ go far beyond those of mere clicktivists (i.e., those who engage in online peer production) and such activists might be regarded as ‘virtual connective leaders’. Their use of social media is negatively associated with offline participation, indicating a non-linear relationship between social media usage and offline participation. In contrast to previous studies, the authors find that social media usage serves to increase offline participation only among low and medium level social media users. In general, the authors show that the virtual space consists of different types of actors, whose interrelated usage patterns are non-linear with respect to different forms and levels of engagement and mobilization.

The forms taken by such usage patterns in digital social movements in authoritarian states and the ways in which they contribute to or even enable pro-democracy activism are addressed in the issue’s final contribution, entitled ‘Counter-Surveillance and Alternative New Media in Turkey’. Here, Bora Ataman and Baris Coban discuss opposing conceptions of surveillance culture and offensive and defensive counter-surveillance tactics in the context of an alternative new media landscape. Ataman and Coban describe how this alternative, independent, critical, and investigative new media landscape has developed away from the traditional mainstream media environment and has come to form a kind of counter-surveillance ‘institution’. They make clear that it is only through social media networks – due to their structural characteristics (low access threshold, speed of dissemination, temporal, and local flexibility) and their philosophy (involving anonymity and the possibility of avoiding and changing identifying features, free from formal constraints) – that such a form of counter-surveillance can arise and be maintained. The authors ascribe activists a key role in exposing undemocratic activities. The usage patterns they describe reflect the various actor types elaborated in the previous contribution, along with their interrelations with respect to online and offline activities. In addition, they indicate the particular security demands that have to be taken into account where digital footprints are concerned (Sambrook, 2016) when gathering, producing, and disseminating news (that is considered illegal) in the public debate.

Editing this special issue and collaborating on it with so many committed people has been a pleasure and a privilege.

This collection developed out of the aforementioned Vienna 2017 symposium, which was organized by Brian D. Loader (Department of Sociology, University of York), Josef Seethaler, and myself (Institute for Comparative Media and Communication Studies, Austrian Academy of Sciences and the University of Klagenfurt). Nevertheless, the articles in this special issue do not simply represent the proceedings of the 2017 conference. All of the papers were revised and passed through a double-blind peer-review process and multiple rounds of revisions. I am now delighted to present these six exceptional studies.

I would therefore like to express my deepest gratitude to all the contributors to this special issue for their thought-provoking and insightful contributions. I would also like to thank the anonymous reviewers who, despite the short turn-around time, offered generous and constructive feedback to the authors. I am very happy that this collection is appearing as a special issue of ‘Information, Communication and Society’. Many thanks to Sarah Shrive Morrison and Brian D. Loader from the University of York, and Josef Seethaler from the Austrian Academy of Sciences for their valuable help during the editing process. I am especially indebted to W. Lance Bennett for our inspiring discussions during the conference – and that they will continue. Last but not least I owe particular thanks to Marlene Gsenger and Valentina Dopona from the local organizing committee, the Kapsch AG for generously funding the conference, and the Austrian Broadcasting Corporation for the pleasant cooperation.

Disclosure statement

No potential conflict of interest was reported by the author.

Notes on contributor

Maren Beaufort is a junior scientist at the Institute for Comparative Media and Communication Studies (CMC) at the Austrian Academy of Sciences and the Alpen-Adria-University Klagenfurt. Her research interests include social media, political participation, media effects, and media literacy. She participates in the European Commission's ‘Media Pluralism Monitor’. Her PhD thesis is concerned with ‘media performance and democracy’[email: [email protected]].