Fake News in the Digital Public Space

Article by Dominique Cardon, Jean-Philippe Cointet, Benjamin Ooghe-Tabanou, Guillaume Plique of the Sciences Po médialab and Bruno Patino, Jean-François Fogel of the Sciences Po Journalism School, using data from a survey conducted by the médialab with Sciences Po’s Journalism School for the Montaigne Institute.

Fake news circulates all around the digital public space via incessant messages sent by robots, pop-ups targeting web users, Russian influence networks... At a time when there is great concern about the authenticity of public discourse, it is worth considering the structure of digital information production and circuits. Such disturbances are often interpreted as the consequence of information market deregulation (Bronner, 2013). The traditional media, short-circuited by social networks open to all sorts of rhetoricians, has lost the control it once held in shaping public opinion. The “digital” public has become treacherous, uncontrollable, and easily manipulated. 

A 3-stage digital public space

While many comments attribute an almost quasi-direct responsibility to these phenomena in certain contemporary political events (Brexit, the elections of Donald Trump or of Jair Bolsonaro), recent research results call for more circumspection. In the United States, for example, even though a significant amount of “fake news” circulated during the 2016 presidential election, the information was mainly disseminated to an audience of activists who were already convinced of their views (Grinberg et al., 2019; Guess, Nyhan, Reifler, 2016), and the few studies on the effects on users fall short of measuring its impact (Allport, Gentzkow, 2017). Rather than approach the issue of fake news in an isolated manner, it makes sense to reposition it within the overarching architecture of the digital information space. The latter is not an even surface that equally exposes everyone to all possible sources of information. The production of information, its visibility, and the way it exists in different parts of the web depend on the way journalists, politicians, and new voices arising build, and specifically shape, information circuits. It is thus possible to show at a macroscopic level that digital information circulation paths are closely connected to the properties of various national media and political spaces. Three strata of this overarching architecture can be schematically isolated: the authority of information from media websites, the influence resulting from the circulation of this information, especially on Twitter, and the conversation, particularly on Facebook, which, in a more fragmented and covert way, is developing in the relational niches that users create (Fig 1.).

In this very simple diagram, it is possible to identify two dynamics creating tension in the structure of digital public space: a polarisation process and what we call a disconnection process. This distinction shows that a system of forces and constraints allow different circuits of information to emerge, depending on the strategies of media and political agents and the consistency of political and professional perimeters in the public space. This difference is particularly obvious when comparing research on the United States and France.

American polarisation

In the aftermath of Donald Trump’s election, an ideological polarisation process has become apparent in the American media. The disinformation produced on extremist sites that usually have poor visibility – i.e. with little authority and with an influence limited to small parts of social networks – is regularly picked up by media with great visibility and strong authority in the mainstream media space (Benkler et al., 20018). With the dominance of Fox News and of Breitbart as leading media in Republican circles and the election of Donald Trump to office, equipped with his Twitter account, it is from the center of the public space that agents with great authority have sought to “launder” the questionable information produced by lower-visibility disinformation websites. From this “official” level and downwards throughout the digital space, a propaganda feedback loop has appeared (Benkler et al., 2018) and contributed to dividing the American media space and creating a different space – one of alternative facts. This polarisation of the American media system results from a broad and deep political strategy involving political and media agents with many resources and highly significant visibility. It is necessary to have positions of authority to establish the existence, in a polemic way, of the idea of an “alternative world” wherein facts deemed fake by most mainstream media are actually “true”. This global transformation of the American media landscape, following the alt-right takeover of Fox News, played a key role in shaping the media agenda during the 2016 election, leading the New York Times to in turn feature more frequently Hillary Clinton’s email scandal on its front page than the candidate’s platform (Watts & Rothschild, 2017).*

The French disconnect

This process is not apparent in the French media, although many tensions appeared in the mainstream media space with the introduction of new agents and new ways of promoting factually dubious information. The media at the center of the public space quote each other, but in solidarity do not (or barely) quote questionable information produced by counter-information websites (Gaumont et al., 2018). Even if Twitter shares, especially from the far right, can help fake news gain credibility, in the French situation mainstream media generally resist misinformation from the counter-informational sphere. This solidarity in the French media landscape is characterised by journalists’ high level of caution on the reliability of information and their use of fact-checking mechanisms – as attested by the CheckNews initiative, which gathered the editors of various French newspapers during the 2017 presidential and legislative elections.

While these mechanisms have had no effect on the audiences most likely to embrace dubious information (Barrera Rodriguez, 2018), they have brought about mutual monitoring among professionals. Thus the French media system has a limited permeability, with the mainstream media collectively resisting the integration of questionable or fake information.

These conclusions remain partial for lack of better knowledge about another communication space that is more difficult to study: that of conversations on social networks or messaging. Many observations underscore that it is mainly on platforms that limit communication to a small circle (such as Facebook, Snapchat, WeChat, and WhatsApp) that strange, suspect, or manipulative information are most shared. This viral circulation, which under some circumstances can be significant, does not necessarily obey (this remains a question for research to explore) the same principles of organisation of political values and of attention to the legitimacy of sources as the higher levels of the pyramid. This is how the Yellow Vest (Gilets Jaunes) movement might be interpreted. In the absence of a mainstream outlet, the public space does not polarise vertically, but rather horizontally, by opposing highly visible spaces (websites and Twitter) to the conversational space of Facebook. Unlike political polarisation, which, in the American case, reflects an ideological differentiation between two Americas that increasingly oppose one another, the situation created in France by the Yellow Vest movement opposes mainstream media and the “Facebook population”, according to a disconnection logic represented in the following figure.

In this context, what is important is not so much the boundary between truth and falsehood, as the opposition between the bottom and the top, the general audience and the media elites. This anger has been expressed through several acts of violence against journalists. The whole dynamic of the mobilisation attests to a direction of movement from the periphery to the center: from roundabouts to city centers, and then to Paris; from Facebook to the front page of leading mainstream newspapers. Seen from this perspective, the Yellow Vests movement appears to have successfully imposed its protest agenda onto a central public space that was ignoring it.

Bibliography

Allcott Hunt, Gentzkow Matthew, 2017 – Social Media and Fake News in the 2016 Election, in Journal of Economic Perspectives, vol. 31, n°2

Barrera Oscar, Guriev Sergei, Henry Émeric, Zhuravskaya Ekaterina, 2017 –  Facts, alternative facts and fact checking in times of post-truth politics. Centre for Economic Policy Research

Benkler Yochai, Faris Robert, Roberts Hal, 2018 –  Network Propaganda. Manipulation, Disinformation, and Radicalization in American Politics, Oxford University Press

Bronner Gérald – La démocratie des crédules, PUF, 2013.

Gaumont Noé, Panahi Maziyar, Chavalarias David, 2018 – Reconstruction of the socio-semantic dynamics of political activist Twitter networks—Method and application to the 2017 French presidential election in PloS one

Grinberg Nir, Joseph Kenneth, Friedland Lisa, Swire-Thompson Briony, Lazer David,  2019, Fake News on Twitter during the 2016 US Presidential Election in Science, n°363

Guess Andrew, Nyhan Brendan, Reifler Jason, 2016 – Selective Exposure to Misinformation: Evidence from the consumption of fake news during the 2016 US Presidential campaign

Watts Duncan J. , Rothschild David M. ,  2017 – Don’t blame the election on fake news. Blame it on the media in Columbia Journalism Review

This article was originally published in Cogito.

Find out more

• Read the full paper "Unfolding Multi-Layered Structure of the French Mediascape" (June 2019) on the médialab website
• Cogito, the research magazine
• Fact-checking: between beliefs and knowledge