Can algorithms be agonistic? If so, what effect does this have on democracy? Does this hurt the democratic process at all? To answer, we turn to Crawford (2016), who goes further in depth with this question. With the digitization of modern society, algorithms are becoming more and more potent within our daily lives, vastly influencing our decision making processes, to such an extent that political decision making is becoming more distorted than ever. In a political context, the idea of algorithmic agonism is more prevalent within the realm of ‘politics’,‘the collection of practices, discourses and institutions that establish forms of order and social organization which are always potentially conflicting’, as opposed to ‘the political’, ‘an antagonistic dimension, a terrain of conflict, which is always present in human societies’ (Mouffe 2003; Crawford 2016, p. 2).
As Crawford explains, algorithms are an ‘effective means for establishing power and authority’ (Crawford 2016, p. 2), power and authority that political parties strive to attain. In today’s political arena, it has become commonplace for political parties to employ algorithms as a means of influencing public opinion, for example, utilising Reddit’s voting algorithm. Although, this algorithm is at times flawed. As Crawford (2016) detailed, a subreddit created during the events of the Boston marathon bombings called ‘findbostonbommers’ that ‘peaked [at] over 272,000 people online [and] 85,000 following the subreddit’ (Crawford 2016, p. 4), whilst scrutinizing surveillance cameras, wrongly convicted a student by the name of ‘Sunil Tripathi’, highlighting the potential flaws of algorithms.
Furthermore, Crawford asserts that an ‘agonistic struggle is…the condition of a functioning democracy’ (Crawford 2016, p. 2), which is to an extent true. Although, with the political arena already highly agonistic, the idea of algorithmic agonism heightens the conflict within, in this case, the democratic process, to a point that is ultimately detrimental to any form of democratic decision making. Algorithms, as defined by Monniaux , are ‘no more than a finite series of instructions for resolving a problem’ (Monniaux 2017, p. 1), although, one needs a certain level of digital literacy to look past these algorithms. The average person is essentially illiterate within a digital context, having made no real attempt to navigate and critically analyse the masses of data that constantly surround them, as such, their agency is essentially stripped away. Thus, in a democratic decision making process, with no human agency, democratic principles of equality and popular sovereignty cannot be upheld, hence, algorithms are democracy’s biggest hindrance.
References:
Beer, D. 201, ‘The social power of algorithms, Information, Communication & Society’, Information, Communication & Society, vol. 20, no. 1, pp. 1-13
Crawford, K. 2015, ‘Can an Algorithm be Agonistic? Ten Scenes from Life in Calculated Publics’, Science, Technology and Human Values, vol. 41, no. 2, pp 77-92. .
Monniaux, D. 2017, Algorithms: a Threat to Democracy?, CNRS News, accessed 25 August, <https://news.cnrs.fr/opinions/algorithms-a-threat-to-democracy>
13542117 