Celina Navarro Discussion and mediation of social https://orcid.org/0000-0002-3127-7358 celina.navarro@uab.cat outrage on Twitter: The reaction to Universitat Autònoma de Barcelona the judicial sentence of La Manada Òscar Coromina https://orcid.org/0000-0002-6306-4154 Abstract oscar.coromina@uab.cat Universitat Autònoma de Barcelona In recent years, digital feminism has been prominent in the international public sphere and the massive release of individual testimonies of violence against women has had an important Submitted April 29th, 2019 presence on social media. In Spain, the reaction to the judicial Approved sentence of the La Manada case, which only convicted the five October 21st, 2019 men accused of abuse, rather than rape, provoked a public outrage that was negotiated on social networks and was later © 2020 displayed in mass demonstrations across the country. The aim of Communication & Society this study is to analyze the reaction that took place on Twitter on ISSN 0214-0039 E ISSN 2386-7876 the day of the ruling of the sentence (26 April 2018) and how the doi: 10.15581/003.33.1.93-106 dominant frames were initiated and stabilized on Twitter. With a www.communication-society.com sample of almost 500.000 tweets, the posts achieving most retweets per hour and the most salient media objects (including 2020 – Vol. 33(1) hashtags, visual content and links) have been analyzed based on pp. 93-106 the framing theory. Media objects have been explored through flowcharts allowing the visualization of the evolution of the How to cite this article: dominant messages. The results show a hegemonic conversation Navarro, C. & Coromina, Ò. (2020). on Twitter against the ruling of the court with a high presence of Discussion and mediation of social outrage on Twitter: The reaction to feminist demands. Furthermore, the analysis of the evolution of the judicial sentence of La Manada. the main frames shows that the different Twitter affordances Communication & Society, 33(1), 93- allow rapidity in the negotiation of the dominant messages. This 106. paper contributes to the understanding of the articulation of framing from a complex perspective including media objects such as images and URLs that broaden the layers of meaning of the Twitter discourse on collective actions. Keywords Feminism, Twitter, framing, cyberfeminism, activism, social networks. 1. Introduction In recent years, the feminist movement and its demands have become predominant in the agenda setting and in the public sphere that has enabled significant street mobilization at an international level. It is difficult to explain this current context without highlighting the role that social networks have played in the visibility and condemnation of sexual violence against women. One of the cases with the greatest impact began in 2017 when the American actress Alyssa Milano encouraged women victims of sexual aggression to use the hashtag #metoo on social networks to show the magnitude of the problem, an expression that was initially launched by the activist Tarana Burke in 2006. This initiative ended up becoming a massive campaign on a global scale to condemn sexual harassment in which celebrities and 93 ISSN 2386-7876 – © 2020 Communication & Society, 33(1), 93-106 Navarro, C. & Coromina, Ò. Discussion and mediation of social outrage on Twitter: The reaction to the judicial sentence of La Manada anonymous women participated and explained their stories (Mendes et al., 2018). From these personal experiences, structural problems of society were exposed (Baer, 2016; Summers- Effler, 2002). In Spain, feminism has also been more visible and has had a greater impact being at the forefront of the international movement, particularly with the success of the two feminist strikes in 2018 and 2019 convened on March 8, International Women’s Day. The first feminist general strike led to major demonstrations, which were even greater the following year, and put a diversity of issues related to gender differences on the public agenda. Two months later, in April 2018, the reactions around the controversial case of La Manada, the object of study of this research, put feminism back at the center of the media agenda. In 2016, a woman reported five men for rape during the popular San Fermines festival, who shared a WhatsApp group with the title La Manada [in English “Wolf Pack”]. The case had a high media profile from the initial accusation and report and immediate arrests, through the trial where the defense lawyers put into question personal aspects of the victim to generate doubt of her version of events and finally the sentencing. The ruling was broadcast by most of the television and generalist radio stations and main Spanish digital media. At around 1pm on April the 26th, 2018, the provincial hearing of Navarra announced that the five men accused of rape and sexual assault were convicted only of sexual abuse, a crime with lower penalties than those of rape. In addition, one of the three members of the court issued a dissenting vote requesting the acquittal of the accused. This sentence led to a great outcry in the traditional media, demonstrations in the main Spanish cities and an anger that was also reflected across digital platforms. This investigation analyzes the reaction that took place on Twitter on the day of the judicial ruling of the La Manada case. Specifically, this study proposes a methodological approach to analyze the evolution of collective indignation and the role played by the Twitter platform and its different affordances in the negotiation of the dominant messages based on the theory of framing. The analysis explores, hour by hour, the evolution of the main frames in the tweets, hashtags, images and links that were more salient. 2. Theoretical framework 2.1. Cyberfeminism Since the beginning of 2010, the rise of digital feminism has placed the movement, online and offline, in the public sphere both nationally and internationally (Pruchniewska, 2018). Whereas during the two previous decades post-feminism had established itself as the social norm by being more focused on individualism and personal freedom (Gill, 2007; McRobbie, 2004), currently the fourth wave of feminism, which is characterized by the use of social networks, is reformulating the movement bringing the possibilities of being more intersectional. That is, including different realities according to race, sexual condition or social class (Zimmerman, 2017), and the collectivization of personal experiences into the feminist message (Mendes et al., 2018). In the case of the feminist struggle, social networks are facilitating their position in the international public sphere. Hashtag feminism is one of the centers of this new type of action by the current feminist movement. In the academic field, studies on the hashtags #YesAllWomen in the context of the United States (Rodino-Colocino, 2014; Thrift, 2014), #EverydaySexism in the United Kingdom (Eagle, 2015), #Iamafeminist in South Korea (Kim, 2017) or #Niunamenos en Iberoamérica (Giraldo-Luque et al., 2018) are clear examples of the interest in analyzing how they are shown and what impact these conversations have on social networks. In Spain, journalist Cristina Fallarás promoted in 2018 a similar initiative using the hashtag #cuéntalo, which was used by more than 40,000 women to report sexual aggression 94 ISSN 2386-7876 – © 2020 Communication & Society, 33(1), 93-106 Navarro, C. & Coromina, Ò. Discussion and mediation of social outrage on Twitter: The reaction to the judicial sentence of La Manada against them1. The initiative arose in response to the court ruling on the case popularly known as La Manada that generated a wave of outrage that resulted in mass demonstrations in major cities across Spain. Social media has altered feminist activism provoking a shift in the aims of the movement, how participants interact among them and challenging hegemonic discourses (Baer, 2016; Clark, 2016). In Western societies, the objectives have moved from the main aim of changing legislation to counteracting the patriarchal social imaginary (Salime, 2014). For example, feminist memes are creating a new and dynamic type of engagement with feminism questioning everyday unequal relationships between the genders (Thrift, 2014). One of the main aspects praised about hashtag feminism is the possibility of creating a community from the individual experiences and opinions of Twitter users (Zajc, 2013). It also facilitates intersectionality within the movement since it can create spaces where counterpublics can join forces (Kim, 2017; Loza, 2014). At the same time, social networks pose difficulties for the feminist movement, as well as for all social movements (Kay LeFebvre & Armstrong, 2018; Tufekci, 2013), relating to the architecture and ecology of these platforms. Caro-Castaño (2015) warns of the presence of a digital feminism closer to neoliberal tendencies that tends towards individualism and the perception of collectivity only for a specific purpose, creating “light collectivities” (Zafra, 2010) and hindering social development. However, “while these online, loosely structured networks might have weak ties; they do define a new type of global activism” (Hogan, 2016, p. 6). 2.2. Activism and framing on Twitter Focusing on Twitter, the platform that has traditionally been associated more often with activism and social movements (Peña-Lopez et al., 2014; Papacharissi & Fatima Oliveira, 2012), it is worth noting the role that hashtags play as a basic mechanism for the organization of “ad hoc audiences” through which users coordinate and formulate conversations around specific topics (Bruns & Burgess, 2015). The central role that hashtags play in political mobilizations has given rise to what is known as “hashtag activism,” a general concept that includes hashtag feminism and can be defined as the use of a slogan of a political or social claim preceded by the hash character [#] in the contents published on social networks to build a narrative connected in formal terms and agency (Williams, 2015; Yang, 2016). Hashtags play a crucial role in what Bennet and Segerberg (2013) define as “connective action” to describe how conversation is organized and frames are defined to narrate and interpret a political conflict. As defined by Entman (1993), framing refers to the aspect that has been selected and highlighted in a communicative text on a topic. In the case of Twitter, these practices can question the information that comes from the traditional media opting to show other frames on the subject. “These hashtags that gain widespread adoption, enact, enable, and sustain the framing of select interpretations, aspects, or frames, to an event over time” (Meraz & Papacharissi, 2013, p. 144). Different investigations have described that framing is articulated from hashtags but also retweets (Meraz & Papacharissi, 2013; Coromina, 2017). In the case of repetitions, retweeting a post implies a certain form of validation and approval of that message (Boyd, Golder & Lottan, 2010). The posts that get the most retweets are those that achieve the highest prominence and saliency within the conversation and is one of the fundamental aspects in the framing process. In the same way, visual content (images and videos) is a central element in the display of narratives on social platforms (Highfield, 2017). In addition, particularly in the framing process since they reflect the priorities of the users and is more effective to express and activate emotions within the narratives of protests (Bosch & Mutsvairo, 2017). Emotions play a very important role as an driver of user participation around political and social issues 1 http://proyectocuentalo.org/. 95 ISSN 2386-7876 – © 2020 Communication & Society, 33(1), 93-106 Navarro, C. & Coromina, Ò. Discussion and mediation of social outrage on Twitter: The reaction to the judicial sentence of La Manada (Papacharissi & Fatima Oliveira, 2012) and different studies show that social platforms are a perfect method of sharing and amplifying “grievances,” contributing to the diffusion of the demands of the social movements (Hussain & Howard, 2013). In addition to textual content, hashtags and visual content, a tweet can also include links to other web pages and social networks broadening the sources of the conversation. As a whole, this repertoire of resources facilitates the propagation of secondary material (links, screenshots, photos of television screens, etc.) and first-hand material, user-generated content that attempts to reproduce events as experienced by the user (Bruns & Burgess, 2015). All these additional resources enrich, complete and extend the 280-character Twitter limit and contribute greatly to the articulation of the story and the dominant frames. 3. Methods 3.1. Objectives and research questions The purpose of this research is to analyze how frames are generated and negotiated on Twitter based on the case of La Manada and the reaction to the judicial ruling that generated collective outrage and crowded demonstrations. Twitter has been chosen because the dynamics of this social network are the most suitable for conversations in real time and has been described as one of the pillars of digital activism (Bruns & Burgess, 2015). In addition, it is the most important platform for the fourth wave feminism (Zimmerman, 2017). More specifically, we are interested in observing the evolution of the dominant discourse and how Twitter affordances participate in the building of these narratives. These objectives are summarized in the following research questions: RQ1: Which are the most salient frames related to the study case and what are their distinguishing features? RQ2: What role do the different media objects (hashtags, visual content and links) of Twitter play in the process of gestation and stabilization of the dominant discourses of feminist activism? RQ3: How do these resources intervene in the evolution and dynamics of the story spread through the Twitter platform? 3.2. Data extraction and analysis From 11am on Thursday, April 26, 2018, two hours before the reading of the sentence, until the next day a capture protocol was activated to collect tweets related to the case study. To do this, the DMI-TCAT software (Borra & Rieder, 2014) was used, which connects to the Twitter API and incorporates the tweets that contain a series of previously defined queries into a database. This tool connects to the Twitter Streaming API to collect live tweets. This method allows access to a higher volume of tweets than the Search API, which is limited to the last 5000, at the cost of not having access to the historical database. For this reason and also because one of the objectives was to analyze the evolution of the frames from which the reaction to the court ruling was developed, a dynamic consultation was chosen. The trending topics in Spain were reviewed hourly during the collection period and those that were directly related to the case were added to the list of keywords. In this way, seven different hashtags were detected: #LaManada, #EstaEsNuestraManada [#ThisIsOurPack in English], #YoSíTeCreo [#IDoBelieveYou], #justiciapatriarcal [Patriarchaljutice], #NoesNo [#NomeansNo], # NoEsAbusoEsViolación [#ItsNotAbuseItsRape] and #NoESPno that allowed us to compile a total of 780.418 tweets issued by 301.021 different users. The evolution of the volume of tweets was closely linked to the events that took place in the offline world. Although the sentence was already trending during the morning, with the reading of the sentence at about 1pm the volume of tweets began to increase considerably, with the peak between 2pm (64.581 tweets) and 3pm (62.328 tweets). Five hours later there was 96 ISSN 2386-7876 – © 2020 Communication & Society, 33(1), 93-106 Navarro, C. & Coromina, Ò. Discussion and mediation of social outrage on Twitter: The reaction to the judicial sentence of La Manada a further increase in the number of posts coinciding with the demonstrations held throughout the Spanish territory. With this database, and to be able to analyze the gestation and evolution of the main frames in detail, our analysis focuses on the 12 consecutive hours with the highest levels of activity before and after the sentence. From 11am to 11pm a total of 497.843 tweets were captured. Of these, the vast majority were retweets (86,4%) while the originals and replies only accounted for 12,7% and 0,9%, respectively. Practically all the tweets incorporated hashtags (99,9%), 16,6% visual content and 10,8% links. With regard to framing, this study has used an inductive approach in the categorization of frames based on the theoretical framework presented and knowledge of the context of the case study. We have analyzed the most retweeted content, and the most recurrent hashtags, visual content and links, to define and delve into the different frames that were used to talk about the sentence. The data was reviewed hour by hour from 11am to 11pm to observe the evolution of the conversation and the process of negotiation and setting of the frames during the day in which the sentence was known. A particular area of interest was if the discourses were articulated around the specifics of the case or were able to connect to the generic feminist agenda. In the conversation developed on Twitter, both before and after the reading of the sentence, the dialogue was united in outrage against the sentence, in support of the victim and criticism of the judicial and social system considered to be patriarchal. Taking into account this homogeneity from the most general point of view and with the global analysis of the tweets with the largest number of retweets and most prominent media objects (hashtags, visual content and links), five different frames were established focusing on the emphasis of the messages, highlighted in Table 1. In general, media objects and tweets have been divided into those referring specifically to the La Manada case (regarding the victim of the case, the five accused men and the sentence) and others that use the case to discuss more general issues related to the judicial system and gender violence from a feminist perspective. Furthermore, it is acknowledged that most of the elements analyzed fell under the scope of two or more frames detected. In this sense, there is the belief that the qualitative approach helps to explain and better analyse the real conditions of how the conversation and negotiation of the frames took place. Table 1: Framing categories extracted from the analysis. Frames Description Examples Specific Victim Topics supporting the Tweets: about La woman who denounced the - Compañera, no estás sola #YoSiTeCreo Manada case group rape and defending the #EstaEsNuestraManada veracity of her complaint. - Toda mi energía puesta ya en el alma y el espíritu de la superviviente de #LaManada. Pase lo que pase No Estas Sola Hashtags: # YoSiTeCreo; #HermanaNoEstásSola Accused Criticism of the five men Tweets: accused of rape. - #EstaEsNuestraManada Ellos se quieren pixelados, pero #NuestraManada los quiere con nombres y apellidos y viéndoseles bien las caras, para que sientan VERGÜENZA y les señalen con el dedo por VIOLADORES 97 ISSN 2386-7876 – © 2020 Communication & Society, 33(1), 93-106 Navarro, C. & Coromina, Ò. Discussion and mediation of social outrage on Twitter: The reaction to the judicial sentence of La Manada - Vergüenza. No se puede sentir otra cosa de estos hombres. #EstaEsNuestraManada #YoSiTeCreo Sentence Criticisms of the sentence of Tweets: the case of La Manada, to - Condenan a #LaManada por abuso sexual y no por violación, specific clauses of the reduciendo así la condena. Al parecer, el juez no considera sentence or the judges of the violento que 5 tíos violen sexualmente a la mujer. Aquí, para court. que vean violencia, hay que rapear. De vergüenza. Llamamos a acudir a las movilizaciones convocadas. - Un juez, hombre, ha votado por la absolución de #LaManada. Tenemos un problema muy serio. - #LaManada el juez González votó por la absolución. - El juez Ricardo González pedía la absolución para todos. ¿Qué cojones le pasa a la gente por la cabeza? #LaManada Hashtags: #NoesAbusoesViolación Generic Judicial Complaints about the Tweets: topics system definition of abuse and rape - #YoSiTeCreo #EstaEsNuestraManada Una justicia que and requirements to change juzgue a los violadores, no a las mujeres violadas the Spanish Penal Code. #NoEstasSola - Un tobillo roto a un guardia civil 50 años. Un guardia civil que viola...9 años. #JusticiaPatriarcal #EstaEsNuestraManada - #EstaEsNuestraManada El mensaje q envía la #JusticiaPatriarcal Es q Violar sale gratis #NoEsAbuso es #violación Hoy todas a la calle Que la culpabilidad recaiga en los agresores Hashtags: #Justiciapatriarcal; #NoesNo Feminism Awareness of sorority and Tweets: community in the face of - Somos fuertes y estamos más unidas que nunca. Juntas lo sexual and gender violence2. conseguiremos #EstaEsNuestraManada - LA MANADA SOMOS NOSOTRAS Nos vemos hoy en el Ministerio de Justicia (Metro Noviciado) a las 20H #EstaEsNuestraManada Hashtags: #EstaEsNuestraManada; #AlertaFeminisga Non-Topic Not containing Tweets: any framing - 9 anys #LaManada expression. - preparada para ver qué pasa con #LaManada. Hashtags: #LaManada Source: Own elaboration. 2 The term sorority, admitted in the dictionary of the Royal Spanish Academy (RAE) in 2018, refers to the sisterhood among women beyond differences in class, race or religion (Varela, 2013). 98 ISSN 2386-7876 – © 2020 Communication & Society, 33(1), 93-106 Navarro, C. & Coromina, Ò. Discussion and mediation of social outrage on Twitter: The reaction to the judicial sentence of La Manada The results of the media objects (hashtags, visual content and links) have been represented graphically with the rankflow tool (Rieder, 2016) that produces flowcharts that allows the easy visualization of the variations in the rankings of the ten most recurrent objects each hour and provides an entry point to analyze the moments at which the most significant changes occur (Rieder et al., 2018). Additionally, the rank-biased distance (RBD) metric has been calculated (Webber et al., 2010), which is an objective parameter to measure the variations in the ordering of results. The statistical results range from 0 to 1, the closer the index to 1, the more variation between that hour and the previous one in the most predominant objects. In the case of links, the rankflow analysis has been done with the domains of the URLs to detect the main sources (media outlets, other social network platforms…) in the conversation and, later, an in-depth analysis of the specific links has been qualitatively analyzed. 4. Results 4.1. The gestation and stabilization of the main frames Starting with an analysis of the 10 most retweeted tweets during each hour of the analysis, a clear evolution of the most hegemonic frames is found. Whereas before the ruling the dominant discourse was through the frame of feminism with secondary topics supporting the victim and against the five accused men, the public reading of the sentence at 1pm produced a change in the most retweeted contents, putting messages about the La Manada sentence and criticism of the judicial system in the center. The tweets focused on the sentence were mostly messages of attack on the judges, highlighting some specific clauses of the sentence and comparisons with sentences for other crimes. In the case of tweets about the judicial system, these emphasized the patriarchal basis of justice and the need to change the Spanish criminal code. Although feminism in Western countries is currently focusing more on changes in the patriarchal structure of society (Salime, 2014), in this case the conversation was centered on laws considered patriarchal and calling for urgent changes to the Spanish Penal Code. Once the initial shock of the sentence was over, from 5pm, the frames of the most retweeted posts focused on feminism frames and criticisms of the judicial system, leaving aside the specific attacks on the La Manada’s sentencing. The importance of the feminism frame on the social networks dialogue, which was also reflected in the demonstrations called for that same day, exerted pressure on the political parties. On the one hand, the parties of the left (Podemos and the Spanish Socialist Workers Party) took the controversy to the European Parliament, where it was debated whether the country met international standards on sexual violence (Álvarez, 2018). On the other hand, the conservative government of that time (Partido Popular), was forced announce the creation of a commission to study the reform of the Penal Code articles relating to sexual crimes. However, it was a very controversial decision to not include any women in the committee of 20 jurists (Diego & Sevilla Lorenzo, 2018). 4.2. The role of media objects in the evolution of frames The analysis of the variations in the rankings of hashtags, images and most frequent domains by calculating the RBD at each hour of the extraction period reveals that the ranking of images is the one that registers the most variations throughout the day analyzed with an average RBD of 0.5. This is followed by that of domains (0.44) and, with a significant lower metric, hashtags (0.12). This data indicates that while the hashtags and the linked domains (not the URLs) experience little variation throughout the day, the images represent a more varied and changing ecosystem. 99 ISSN 2386-7876 – © 2020 Communication & Society, 33(1), 93-106 Navarro, C. & Coromina, Ò. Discussion and mediation of social outrage on Twitter: The reaction to the judicial sentence of La Manada Figure 1: Rank-biased distance (RBD) metrics per hour and media object. 1 0,9 0,8 0,7 0,6 0,5 0,4 0,3 0,2 0,1 0 11am 12am 1pm 2pm 3pm 4pm 5pm 6pm 7pm 8pm 9pm 10pm Images Hashtags Domains Source: Own elaboration. Figure 2 shows the rankflow visualization of the changes in the evolution of the most used hashtags in the 12 analyzed central hours before and after the reading of the judicial sentence. It needs to be highlighted that the hashtag #LaManada, the most popular in each hour of the tweet capture, has not been considered in the rankflow, as it does not contain any frames. Although most hashtags do not have enough information to frame them within a specific category because their information is more abstract and general, the analysis of the other elements analyzed in this research will help to shape the negotiation of dominant messages and frames. Figure 2: Hashtags rankflow. Source: Own elaboration. The vertical axis reflects the most frequent hashtags at the top and the least common on the bottom with the height and the colors adjusted according to the numbers to ease comparisons between the different hours that correspond to the columns. In this rank can be observed that at 1pm there is a large variation in the content used to enrich the text. At this time the sentence was made public and, as the data attests, begins to announce the response of the feminist movement to the verdict. From that moment, the hashtags from which the speech is organized is stabilized and the dynamism moves to the ranking of images that take a few more hours to consolidate. Coinciding with the demonstrations, a rebound in the changes in the ranking of 100 ISSN 2386-7876 – © 2020 Communication & Society, 33(1), 93-106 Navarro, C. & Coromina, Ò. Discussion and mediation of social outrage on Twitter: The reaction to the judicial sentence of La Manada the domains of the most shared links was observed and, to a lesser extent, of the shared visual content. As can be seen in the data of the RBD (Figure 1), the greatest change in the use of the hashtags occurred just after the reading of the sentence, between 13 and 14 hours. It is in this period that the hashtag #NoesNo emerges with a clear message of outrage against the ruling. The official profile of the National Police (@policia), with the hashtag #NoesNo, published a tweet that can be interpreted as criticizing the decision to charge the defendants with sexual abuse rather than rape based on the claim that the victim did not fight enough. At 1pm, the hashtag on the judicial system #justiciapatriarcal, the sentence beyond the specific case of La Manada, also comes into scene. Another moment to highlight is between 8 and 9pm, coinciding with the demonstrations occurring across Spain. An hour before, hashtags of cities like Bilbao, Barcelona and Madrid, where the most massive demonstrations took place, began to appear. Figure 3: Images rankflow. Source: Own elaboration. In Figure 3, we can see the variations in the ranking of the most recurrent visual content in the collection of tweets. Although the same visualization in the form of a rankflow is used, on this occasion a capture of the photo or a frame was chosen to be inserted in the case of audiovisual content to facilitate the exploration. The analysis of each of the images allows the identification of 3 different visual stages that correspond to 1) earlier hours and the immediate reaction to the sentence, 2) the hours prior to the demonstrations and 3) the demonstrations and subsequent coverage. Beyond their relationship with the timeline of the day, these 3 stages are also characterized by visual styles associated with the specific cultural practices of social media. This is the case of the callout and online public shaming culture (Highfield, 2017) that can explain the prominence if the image that exposes the identity of each one of the members of La Manada. This image ranked in the top 10 throughout the day. As well as this, another strategy is the publication of screenshots of texts written on other platforms or media to avoid the restriction of 280 characters per message and reproduce longer texts also highlighted by Highfield (2017). This practice was manifested with a special intensity in the hours after the sentence became public knowledge and we can relate to the need to discuss, argue and negotiate the frames around which the feminist response is articulated. 101 ISSN 2386-7876 – © 2020 Communication & Society, 33(1), 93-106 Navarro, C. & Coromina, Ò. Discussion and mediation of social outrage on Twitter: The reaction to the judicial sentence of La Manada The shared texts serve to spread news published in the media, poems and specially to disseminate information about the different mobilization calls that we can consider as protest materials created for the organization of the demonstrations (Philipps, 2012). Both in the period prior to sentencing and the hours prior to the demonstrations, the diffusion of images with the hashtags promoted by the feminist movement is observed showing that the movement was prepared and organized to respond to the judicial ruling. The third and last visual stage is characterized precisely by the abundance of images of demonstrations and protest materials. These coexist at the top of the rankings with the image of those sentenced for sexual abuse and text captures that reproduce a fragment of the judicial sentence. In it, the judge who issues the vote in favor of absolution discloses the arguments that support its interpretation. Parallel to the consolidation of the hashtag #justiciapatriarcal and the consolidation of the frame that denounces the patriarchal bias of the judicial system, we observe how an infographic image created by the user @mariasande goes on to head the hourly rankings of the most frequent images from 4pm to 11pm (Figure 3). This graph criticizes in a flow diagram the only options, with really strong consequences faced by the victim, to get a heavier sentence in sexual abuse cases in the Spanish judicial system. Figure 4: Domain rankflow. Source: Own elaboration. In the case of the evolution of the shared domains external to Twitter, it can be seen that despite the visual variation in the ranking of the different domains, the vast majority are links to media outlets, highlighting the weight of the digital exclusively media eldiario.es. This medium is described by the desire to incorporate a gender perspective into their journalism. Another important aspect to remark is the low presence of other social networks such as Facebook or Instagram within the most common links in the conversation, resulting in little content created by the users themselves. This is only relevant in the hours during the demonstrations. If the different shared links of each domain are analyzed more deeply, five different stages in the type of linked page can be extracted, two more than in the case of retweets and visual content. This may be explained due to the possibility of external sources adding extra layers of meaning to the Twitter dialogue. In this case, it should be noted that at the beginning of our sample, links were showing media articles about the different possible rulings. At 1pm, when the public reading of the sentence was made, the URLs linked to the live signals of different media which were broadcasting from the Supreme Court of Navarre. 102 ISSN 2386-7876 – © 2020 Communication & Society, 33(1), 93-106 Navarro, C. & Coromina, Ò. Discussion and mediation of social outrage on Twitter: The reaction to the judicial sentence of La Manada After this, the dominant links were redirected to the headlines of different media covering the sentence. Eldiario.es, the Catalan public media corporation (CCMA) and La Vanguardia, a Catalan newspaper, should be highlighted. These links were soon replaced by articles analyzing the sentence and the criminal code in depth. At 5pm the appearance of the scribd.com domain stands out, a platform that is used to share files between users. Here, the full text of the sentence was posted and it was one of the most prominent links. Simultaneously, links to the calls for the demonstrations were prominent until 7pm. It is during this time that links finally appear on the success of street demonstrations throughout Spain. At this moment it is the only one in which Instagram links are part of the dominant discourse which included visual content from the demonstrations. 5. Conclusions To conclude, this research, which is included within the studies of digital feminism, has focused on exploring how frames were negotiated on the day of the judgment of the case, known colloquially as La Manada, was announced. The relevance of this analysis is justified by the need to understand how social movements, or even collective outrage regarding a specific case, create, negotiate and stabilize frames on digital platforms and contribute to the organization of demonstrations. The case of La Manada has proved to be a good example of the rapidity in the articulation of discourse and collective action. Our collection of Twitter data shows how this platform was an essential element in the organization of the numerous and mass demonstrations that took place that same day on the streets of the main cities throughout Spain. In addition, we have observed how the feminist movement remained alert from the beginning of the day to give voice to shared indignation if, as happened, the sentence did not recognize this case of sexual violence as rape. The conversation around La Manada on Twitter was monopolized by a broad sector of society that gave its support to the victim and strongly criticized the fact that the five accused were only convicted of abuse and not of rape. Many of these individuals had demonstrated only some weeks before during the first feminist strike on March the 8th, 2018, in which social networks also played a relevant role. This recent event contributed to the expectant attitude of the feminist movement. However, the permanent mobilization of the movement was not at odds with the ability to evolve and modify the frames by consensus as the day progressed. Another contribution of our study is the methodological proposal of analyzing how the dominant frames used by activism on Twitter are negotiated in a moment of collective grievance and the role played by different resources and media objects of the platforms. Based on the theory of frames, and with a qualitative point of view (Valera Ordaz, 2016), we consider that the analysis of the media objects present in the affordances of the analyzed social platform reflects in a more complex and complete way the frames of the collective indignation and criticism of the sentence. Although the more salient tweets and most popular hashtags, two elements on which most previous studies have been based (Meraz & Papacharissi, 2013; Coromina, 2017), provide basic information about the dominant messages on Twitter and their evolution, the analysis of other media objects such as visual content and URLs complemented with more exhaustive results. This has allowed us to describe the diversity of messages in the public space and see how the different platform cultures such as callout culture and public shaming (Highfield, 2017) or sharing visual captures of texts published in other media contribute to the creation of dominant discourses. Obviously, it is difficult to assimilate a tweet, an image, a link or a hashtag to a frame that by definition provides an interpretation of a complex reality within a public space and an even broader social movement. For this reason, the classification of these objects exclusively in one of the found frames has been rejected and, instead, a qualitative approach has been used. 103 ISSN 2386-7876 – © 2020 Communication & Society, 33(1), 93-106 Navarro, C. & Coromina, Ò. Discussion and mediation of social outrage on Twitter: The reaction to the judicial sentence of La Manada This dialogue, consensually against the sentence, had five different frames divided into two groups: those that focused exclusively on the specific case of La Manada and those that, based on the sentence, took into account feminist demands, highlighting the importance and relevance of the movement in the public sphere. In the first case, the frames focus on support for the victim, criticism of the members of La Manada and specific criticism of the judicial resolution of the case study. In the second group of frames, the feminist sorority and the criticism of the judicial system show the clear view by the majority of users that the consequences of the sentence did not affect only the specific case but the Spanish social context in general and the legitimization of certain violent behavior towards women. This feminist view was considered to be true even before the reading of the sentence. This journey from the specific ruling to the systemic was evident in the consolidation of the hashtag #justiciapatriarcal [#patriarchaljustice], the diagram that reproduces the options faced by a victim in a situation of sexual violence in the Spanish judicial system and also in the evolution of the most retweeted messages. Based on the evolution and negotiation of the dominant discourses of shared outrage, three key points throughout the 12 hours of the sentence have been found from before the sentence, the response to the reading and finally the street demonstrations that occurred across the country. However, the detailed analysis of the media objects allows us to conclude that each media object on the Twitter platform participates at a different pace in the processes of creating and stabilizing the dominant discourse and this is attributable to the functions performed by each of these resources in the dynamics of the platform. That is hashtags organize the contents, images channel the emotions better and enrich the content and links extend the conversation by directing to other sources. In the case of La Manada, the hashtags had a very fast response to criticism of the sentence immediately after the reading the sentence. After this initial reformulation of the frames, they stabilized during the rest of the day. In contrast, the visual content and links had more variations as demonstrated by the results of the rank-biased distance metric. With these results, it can be inferred that these media objects behave more dynamically and are more difficult to stabilize and simplify. It is a logical fact if we consider that the function of hashtags is to simplify and group information and audiences, while images and links add layers of meaning and manage to create more complex messages. At the same time, for the articulation of the connective action (Bennet & Segerberg, 2013) that led to important mobilizations, links and images were mainly used. Finally, for future research we consider it would be interesting to replicate the methodological design proposed in this paper to detect similarities and differences in the process of framing articulation in the context of mobilizations on Twitter but also on other social media platforms. The comparative analysis of media objects on different platforms should also contribute to understanding the particularities of the language and cultures in the use of each platform in the context of digital activism. References Álvarez, P. (2018, March 3). La sentencia de La Manada desata una tormenta política y social. El País. Retrieved from https://elpais.com/politica/2018/05/02/actualidad/1525286725_262083.html Baer, H. (2016). Redoing feminism: digital activism, body politics, and neoliberalism. Feminist Media Studies, 16(1), 17-34. https://www.doi.org/10.1080/14680777.2015.1093070 Bennet, W. L. & Segerberg, A. (2013). The Logic of Connective Action: Digital Media and the Personalization of Contentious Politics. New York: Cambridge University Press. Borra, E. & Rieder, B. (2014). Programmed method: Developing a toolset for capturing and analyzing tweets. Aslib Journal of Information Management, 66(3), 262-278. https://www.doi.org/10.1108/ajim-09-2013-0094 104 ISSN 2386-7876 – © 2020 Communication & Society, 33(1), 93-106 Navarro, C. & Coromina, Ò. Discussion and mediation of social outrage on Twitter: The reaction to the judicial sentence of La Manada Bosch, T. & Mutsvairo, B. (2017). Pictures, protests and politics: Mapping Twitter images during South Africa’s Fees Must Fall campaign. African Journalism Studies, 38(2), 71-89. https://www.doi.org/10.1080/23743670.2017.1368869 Boyd, D., Golder, S. & Lottan, G. (2010). Tweet, Tweet, Retweet: Conversational Aspects of Retweeting on Twitter. In IEEE (Ed.), HICSS-43. Kauai, HI. Bruns, A. & Burgess, J. (2015). Twitter hashtags from ad hoc to calculated publics. In N. Rambukkana (Ed.), Hashtags Publics: The Power and Politics of Discursive Networks (pp. 13- 28). New York: Peter Lang. Caro Castaño, L. (2015). Construir y comunicar un “nosotras” feminista desde los medios sociales. Una reflexión acerca del “feminismo del hashtag”. Commons: Revista de Comunicación y Ciudadanía Digital, 4(2), 124-154. https://www.doi.org/10.25267/commons.2015.v4.i2.06 Clark, R. (2016). “Hope in a hashtag”: The discursive activism of #WhyIStayed. Feminist Media Studies, 16(5), 788–804. https://www.doi.org/10.1080/14680777.2016.1138235 Coromina, Ò. (2017). Pugna por el relato en los contenciosos políticos. El caso del proceso participativo del 9N de 2014. El profesional de la información, 26(5), 884-893. https://www.doi.org/10.3145/epi.2017.sep.10 Diego, M. & Sevilla Lorenzo, J. (2018, May 4). La comisión del Gobierno que revisará los delitos sexuales del Código Penal está formada por 20 hombres y ninguna mujer. eldiario.es. Retrieved from https://www.eldiario.es/politica/Gobierno-violacion-Codigo-Penal- exclusivamente_0_767823275.html Eagle, R. (2015). Loitering, lingering, hashtagging: Women reclaiming public space via #BoardtheBus, #StopStreetHarassment, and the #EverydaySexism Project. Feminist Media Studies, 15(2), 350-353. https://www.doi.org/10.1080/14680777.2015.1008748 Entman, R. (1993). Framing: Toward a clarification of a fractured paradigm. Journal of Communication, 43(4), 51-58. https://www.doi.org/10.1111/j.1460-2466.1993.tb01304.x Gill, R. (2007). Postfeminist media culture: Elements of a sensibility. European Journal of Cultural Studies, 10(2), 147–166. Giraldo-Luque, S., Fernández-García, N. & Pérez-Arce, J. C. (2018). La centralidad temática de la movilización #NiUnaMenos en Twitter. El Profesional de la Información, 27(1), 96– 105. https://www.doi.org/10.3145/epi.2018.ene.09 Highfield, T. (2017). Social media and everyday politics. Malden, MA: John Wiley & Sons. Hogan, A. (2016). #tellPearson: The activist “Public Education” network. Discourse: Studies in the Cultural Politics of Education, 39(1), 1–16. https://www.doi.org/10.1080/01596306.2016.1269225 Hussain, M. M. & Howard, P. N. (2013). Democracy’s fourth wave? Digital media and the Arab Spring. New York, NY: Oxford University Press. Kay LeFebvre, R. & Armstrong, C. (2018). Grievance-based social movement mobilization in the #Ferguson Twitter storm. New Media & Society, 20(1), 9–28. https://www.doi.org/10.1177/1461444816644697 Kim, J. (2017). #iamafeminist as the “Mother Tag”: Feminist identification and activism against misogyny on Twitter in South Korea. Feminist Media Studies, 17(5), 804. https://www.doi.org/10.1080/14680777.2017.1283343 Loza, S. (2014). Hashtag Feminism, #SolidarityIsForWhiteWomen, and the other #FemFuture. Ada - A Journal of Gender New Media & Technology (5). McRobbie, A. (2004). Post-feminism and popular culture. Feminist Media Studies, 4(3), 255-264. https://www.doi.org/10.1080/1468077042000309937 Mendes, K., Ringrose, J. & Keller, J. (2018). #MeToo and the promise and pitfalls of challenging rape culture through digital feminist activism. European Journal of Women’s Studies, 25(2), 236-246. https://www.doi.org/10.1177/1350506818765318 105 ISSN 2386-7876 – © 2020 Communication & Society, 33(1), 93-106 Navarro, C. & Coromina, Ò. Discussion and mediation of social outrage on Twitter: The reaction to the judicial sentence of La Manada Meraz, S. & Papacharissi, Z. (2013). Networked Gatekeeping and Networked Framing on #Egypt. The International Journal of Press/Politics, 18(2), 138-166. https://www.doi.org/10.1177/1940161212474472 Papacharissi, Z. & Fatima Oliveira, M. (2012). Affective news and networked publics: The rythms of news storytelling on #Egypt. Journal of Communication, 62(2), 266-282. https://www.doi.org/10.1111/j.1460-2466.2012.01630.x Peña-López, I., Congosto, M. & Aragón, P. (2013). Spanish Indignados and the evolution of 15M: Towards networked para-institutions. Journal of Spanish Cultural Studies, 15(1-2), 189- 216. https://www.doi.org/10.1080/14636204.2014.931678 Philipps, A. (2012) Visual protest material as empirical data. Visual Communication, 11(1), 3-21. https://www.doi.org/10.1177/1470357211424675 Pruchniewska, U. M. (2018). Branding the self as an “authentic feminist”: Negotiating feminist values in post-feminist digital cultural production. Feminist Media Studies, 18(5), 810-824. https://www.doi.org/10.1080/14680777.2017.1355330 Rieder, B. (2016). RankFlow. Computer software. Retrieved from http://labs.polsys.net/tools/rankflow/ Rieder, B., Matamoros-Fernández, A. & Coromina, Ò. (2018). From ranking algorithms to ‘ranking cultures’ Investigating the modulation of visibility in YouTube search results. Convergence, 24(1), 50-68. https://www.doi.org/10.1177/1354856517736982 Rodino-Colocino, M. (2014). #YesAllWomen: Intersectional mobilization against sexual assault is radical (again). Feminist Media Studies, 14(6), 1113–1115. https://www.doi.org/10.1080/14680777.2014.975475 Salime, Z. (2014). New feminism as personal revolutions: Microrebellious bodies. Signs, 40(1), 14-20. https://www.doi.org/10.1086/676962 Summers-Effler, E. (2002). The micro potential for social change: Emotion, consciousness, and social movement formation. Sociological Theory, 20(1), 41-60. https://www.doi.org/10.1111/1467-9558.00150 Thrift, S. C. (2014). #YesAllWomen as feminist meme event. Feminist Media Studies, 14(6), 1090- 1092. https://www.doi.org/10.1080/14680777.2014.975421 Tufekci, Z. (2013). “Not This One” social movements, the attention economy, and microcelebrity networked activism. American Behavioral Scientist, 57(7), 848-870. https://www.doi.org/10.1177/0002764213479369 Valera Ordaz, L. (2016). El sesgo mediocéntrico del framing en España: una revisión crítica de la aplicación de la teoría del encuadre en los estudios de comunicación. Zer. Revista de Estudios de Comunicación, 21(41), 13-33. https://www.doi.org/10.1387/zer.17259 Varela, N. (2013). Feminismo para principiantes. Barcelona: B de Bolsillo. Williams, S. (2015). Digital defense: Black feminists resist violence with hashtag activism. Feminist Media Studies, 15(2), 341-344. https://www.doi.org/10.1080/14680777.2015.1008744 Zafra, R. (2010). Un cuarto propio conectado. (Ciber)espacio y (auto)gestación del yo. Madrid: Fórcola. Zajc, M. (2013). Social media, prosumption, and dispositives: New mechanisms of the construction of subjectivity. Journal of Consumer Culture, 15(1), 28-47. https://www.doi.org/10.1177/1469540513493201 Webber, W., Moffat, A. & Zobel, J. (2010). A similarity measure for indefinite rankings. ACM Transactions on Information Systems, 28(4), 1-38. https://www.doi.org/10.1145/1852102.1852106 Zimmerman, T. (2017). #Intersectionality: The fourth wave feminist Twitter community. Atlantis, 38(1), 54-70. Yang, G. (2016). Narrative agency in hashtag activism: The case of #BlackLivesMatter. Media and Communication, 4(4), 13. https://www.doi.org/10.17645/mac.v4i4.692 106 ISSN 2386-7876 – © 2020 Communication & Society, 33(1), 93-106