Chicago Journal of International Law call for genocide.166 Where the incitement is not explicit, the court reviews cultural context to determine whether the audience clearly understood the statement as a call for genocide.167 To that effect, a systematic propaganda campaign to dehumanize a protected group may constitute direct incitement.168 Statements which in isolation fall short of incitement may, in the aggregate, satisfy the direct requirement. This approach is consistent with a multi-staged conception of genocide, which begins with the systematic dehumanization of a protected group. It recognizes that a campaign of implicit statements may more effectively incite genocide than one explicit call. The courts in Streicher and in the Media Case understood as much, emphasizing the systematic nature of the incitements in those cases. Social media plainly allows for the possibility of direct incitements. Like print newspapers and radio broadcasts, social media provides inciters with a platform to disseminate their messages. Social media may actually enhance the inciter’s capacity to dehumanize a protected group. Unlike conventional media, it deploys algorithms with a tendency to amplify content tailored to individual predispositions, whatever they may be.169 Such algorithms may contribute to the exacerbation of existing social tensions within a nation.170 In extreme cases, this could result in the dehumanization of a protected group and therefore help set the social conditions necessary for genocide.171 But even if a court ignored the role of algorithms, social media’s capacity to circulate incitements is sufficiently analogous to the conventional media used in Streicher and the Media Case to facilitate direct incitement, through either the explicit call for genocide or the implicit but systematic campaign to dehumanize a protected group.172 166 167 168 169 170 171 172 CLICC, supra note 56, at 271 n.272. Id. See Nahimana, Case No. ICTR 99-52-T at ¶ 486; Streicher, Judgment at 301–03. See Max Fisher & Amanda Taub, How Everyday Social Media Users Become Real-World Extremists, N.Y. TIMES (Apr. 25, 2018), http://perma.cc/9TZH-TXMS. Everyday users might not intend to participate in online outrage, much less lead it. But the incentive structures and social cues of algorithm-driven social media sites like Facebook can train them over time — perhaps without their awareness — to pump up the anger and fear. Eventually, feeding into one another, users arrive at hate speech on their own. Extremism, in other words, can emerge organically. Id. See Mozur, supra note 165 (“The purpose of the campaign, which set the country on edge, was to generate widespread feelings of vulnerability and fear that could be salved only by the military’s protection, said researchers who followed the tactics.”). See Nahimana, Case No. ICTR 99-52-T at ¶ 486; Streicher, Judgment at 301–03. 108 Vol. 21 No. 1