Why Democratic Publics Matter in the Age of Algorithms
Part two in an ongoing series
Part two in an ongoing series
In a democratic society, media plays an essential role in informing the citizenry and enabling democratic participation. Media create online public spaces, or “publics,” which are communities of people who gather for discussion, exchange, and collaboration around issues and policies that impact them. People in publics have some connection or commonality, in this case, a connection to media, as audience or participants.
Fenwick McKelvey (2014) connects this crucial media role with the rise of algorithms, and the difficulties they pose when it comes to facilitating democratic publics. Specifically, McKelvey calls for new methods of research, exploration, and inquiry, so as to properly understand the rising power and influence of algorithms when it comes to their impact on democracy. This series seeks to synthesize the latest research so as to further this call.
Newspapers for example have historically had a significant impact on elections, politics, and public policy (Shanahan et al, 2011). As a consequence, newspapers have been subject to both peer based self-regulation, as well as the need to conform to Tort laws around libel and defamation.
Radio and television are also recognized for the power and influence they wield, and are generally subject to heavy regulation, so as to prevent their abuse or ability to undermine democratic processes and institutions (Feintuck et al, 2006) (Lunt et al, 2011).
The publics created by these legacy media were relatively easy to observe and understand in part due to their relatively limited number. It was prohibitive and expensive to create these kinds of media, and the regulatory regimes in part reflected that.
Algorithms, on the other hand, are relatively easy to produce and, combined with Internet-based distribution, allow for access to a potentially global audience. Not everyone may have the necessary literacy to create their own algorithm (Willis, 2007), however most users of social media do find using algorithms to create media incredibly easy and accessible. As a result, we’ve seen a growing rise in citizen journalism, offering significant potential towards increasing overall participation in democratic processes and institutions (Kaufhold et al, 2010).
Algorithms enable an automated media production that eliminates much of the technical or skilled labour involved in media production so as to allow the creator to tell their story or share their content as easily as possible. Various Internet platforms like Instagram, Snapchat, and YouTube, employ algorithms as part of the tools provided to create media so that users are able to publish instantly and in real time.
Algorithms also connect these creators with customized audiences that use algorithms to select, sort, and rank content that may be of interest to them. These distribution models have almost eliminated any barriers to entry when it comes to global media markets. The aforementioned examples are just a few of hundreds of kinds of algorithmic media that have displaced traditional media to become platforms that connect producers with consumers who themselves become producers as they interact with available content.
Therefore, just as algorithms create an incredible increase in the volume of content created by a growing number of creators, they also create an explosion of publics (Langlois et al, 2009). These publics form as people gather around content, interact, share, and engage with issues and topics that impact our cultures and societies.
An oral society had a town square, print societies had books and a few newspapers, and even early electronic societies had a handful of radio and television networks, limiting the creation of publics to these relatively few and easy to monitor spaces.
Algorithms on the other hand create an exponentially larger number of publics. For example, is the Facebook newsfeed a single public, or in fact are there as many publics as there are Facebook users? Sometimes these publics merge and collide when users interact with each other, however each newsfeed is a customized creation for every user, and thus each creates their own privatized public.
While privatized and customized, these are still publics. They remain the kind of town square that informs and creates the public, enabling voices and participation, that enable a democratic society. While these publics may have a population of one, they can and often do combine together into larger networked publics that wield considerable power (boyd, 2007). There is considerable democratic potential in how these publics enable a deliberative culture within a dynamic public sphere (Gimmler, 2001) (Milan, 2015).
However, the manipulation of these publics is also a growing concern (Epstein et al, 2015), and while there are attempts to reverse engineer how they work (DeVito, 2015), this is a major reason why transparency is essential to understanding how these publics are changing our relationship with democracy, and each other (Kennedy et al, 2015).
We often make the mistake with technology of treating it as magic, and suspending the sort of skepticism that we should have when engaging in public discourse. An example of this is believing algorithms are objective, when as a human construct they are subjective. In particular, algorithms have a kind of invisibility, as they are designed to operate in the background, out of sight, and out of mind.
In order to make the argument as to why they should be visible and transparent, we need to first understand the power they possess, and their impact upon our institutions.