Digital Prospects of the Contemporary Feminist Movement for Dialogue and International Mobilization: A Case Study of the 25 November Twitter Conversation

: The feminist movement is experiencing the rise of a new generation characterized by speciﬁc phenomena linked to technological progress, such as hashtivism, i.e., mobilization through social media. With the aim of contributing to extending our knowledge of the implications of Twitter for this as well as other social movements, this article examined eight of the most common Spanish-and English-language hashtags used to commemorate the 25 November 2018 event, the International Day for the Elimination of Violence against Women. Employing big data analysis to study social and communications phenomena, the results offer a picture of contemporary feminism through the kind of international digital dialogue or conversation that it creates, as well as questioning Twitter’s validity in terms of cohesion when it comes to uniting forces in relation to one of the movement’s most urgent struggles: eliminating violence against women in all its forms. A.L.U. J.O.T.; methodology, A.L.U., J.O.T. J.M.iG.; J.M.iG.; validation, J.O.T. J.M.iG.; analysis, A.L.U., J.O.T. J.M.iG.; inves-tigation, A.L.U., J.O.T. J.M.iG.; resources, A.L.U., J.O.T. J.M.iG.; data curation, J.O.T. J.M.iG.; writing—original preparation, A.L.U., J.O.T. and J.M.iG.; writing—review and editing, A.L.U.; visualization, J.O.T. and J.M.iG.; project administration, A.L.U.; funding acquisition, A.L.U.


Introduction
The rise of the World Wide Web in the 1990s offered the feminist movement a new way to express itself, in addition to using traditional mass media, and it was at this time that the foundations of "cyberfeminism" (Wilding 1998;Pozner 2003;Kelly 2005; Sánchez-Duarte and Fernández-Romero 2017) and a new, fourth wave of feminism (Munro 2013;Lane 2015) were laid. Through cyberfeminism, the movement is participating in the technological transformation that contemporary society is experiencing. As Locke et al. (2018) pointed out, "cyberspace is a gendered space (...) as readers and scholars, we may argue that this space provides a fourth way of feminist practice (...) we may envisage the online space as a potential utopian postfeminist one in which we share, queer and make radically different interpretations". The movement's organization within the context of the Internet implies a particular kind of activism as well as new ideas and practices linked to issues of gender, participation and collective identity.
As well as indicating the differences in collective identity between feminist activists offline and online, Michael Ayers (2003) showed that not all social movements work in the same way in cyberspace and that feminism, in its form known as "cyberfeminism", demonstrates major benefits. More specifically, this author highlighted cyberfeminism's advantage in promoting the feminist movement's collective approach and emotionality, by reshaping it in terms of community and solidarity, beyond individualism and selfawareness.
The presence of feminism on the Internet situates this social and political movement at the cusp of a new cycle that promises greater opportunities in terms of promoting a freer solidarity among women. After all, the interactive potential of the online context is much more favourable than the offline one in terms of the establishment of action networks aiming to expand feminist activism. As Ting Liu (2008) explained, stronger cooperation among different feminist groups plays an important part in enhancing broader public involvement and the power of civil society in terms of women-related issues.
Following this moment of cultural and social progress, achieved with the rise of the Internet in the early 1990s, the consolidation of social media (Twitter, Facebook, YouTube, etc.) around 2004 represented an additional stage of social struggle, to the extent that visibility, mediated through social media, became a decisive instrument of our era for the articulation and development of the efforts of social and political movements (Downing 2001;Thompson 2005;Earl and Kimport 2011;Bruns et al. 2015).
Social media have become representatives of a social reality that lends itself to analysis from different disciplines within the social sciences-sociology, psychology and social communication among others-as well as from different perspectives, including the gender perspective, constantly being renewed through feminist theory and praxis (Baer 2016;Gill 2016;Stubbs-Richardson et al. 2018).
In fact, the feminist movement is even more present and active in social media than in the offline world. In this offline context, the feminist movement has historically had to fight for visibility. This visibility has been excessively dependent on the kind of representation offered by the media, which are particularly interested in concentrating the focus of public opinion on feminists' more sensationalist and seditious acts, rather than on their demands for equality. As different studies have shown, media representations of the main feminist waves of the last century-the suffragette movement in the early 1900s (Cancian and Ross 1981;Gamson and Wolfsfeld 1993) and the women's liberation wave of the sixties and seventies (Cancian and Ross 1981;Huddy 1997;Bradley 2003)-contributed to their public definition in contradictory terms, as a subversive and circumstantial movement, linked to achieving specific goals, and in certain contexts were also linked to left-wing political struggles, as happened in Spain during the pre-democratic period (Larrondo 2019).
It can also be understood that social media contribute to destigmatizing the term "feminism", and furthermore, that the liberty in terms of expression and relations offered by social media cannot easily be countered (Shulevitz and Traister 2014) by the effects of certain media messages that distort or minimize the feminist feeling within society (Kitzinger 2000;Roate 2015), as has occurred during other historical periods.
After a loss of mobilization potential in the last three decades (Schneider 1988;Epstein 2002), this fourth feminist wave has been favoured by technological advances represented by networking services such as Twitter. The strength of this networked community is found in the possibilities that such services offer for a freer expression that allows many individual voices to participate in a dialogue with each other and construct collective subjects based on particular campaigns or hashtags.
From a strictly methodological point of view, the analyses of feminist presence in social media, which have a clearly interdisciplinary nature, have shown a preference for the microblogging medium Twitter because of its greater impact and influence on the opinion of a broad public (Murthy 2010). Considering this use of Twitter as a tool for social analysis, it is important to bear in mind that research carried out until now has prioritized qualitative analysis of both the content of tweets with the presence of certain keywords and of the discussion promoted by these tweets, systematized based on the use of hashtags; this has generally been undertaken with the goal of determining the capacity for influence (indegree influence) and the dissemination or viralization (retweets) of messages.
However, feminist and gender research is making an effort to go further and focus on the parameters that go beyond influence as measured in numbers of followers and retweets, because in a hyperconnected interactive world, social activism is above all a conversation or a dialogue. Different contributions have drawn attention to this need for progress in feminist studies of Twitter (Baer 2016;Sánchez-Duarte and Fernández-Romero 2017;Stubbs-Richardson et al. 2018;Turley and Fisher 2018), given the usefulness of this network to "disseminate feminist ideas, shape new discourses, connect different and diverse groups and allow new and creative forms of protest and activism (...)" (Locke et al. 2018).
With this in mind, this study contributes an analysis of the dialogues or interactions among different groups based on variables that demonstrate a particular interest in the feminist movement (Baer 2016;Boyd and Ellison 2007). We refer to the use of hashtags not only as a cohesive element and one that builds community, but also as something that can promote the independence/autonomy of the movement (denesting)-a circumstance that has hardly been studied up until now (Fotopoulou 2014).

Research on hashtag feminism
In theory, the concept of hashtag feminism has created a virtual space where victims of inequality can coexist together in a space that acknowledges their pain, narrative and isolation (Dixon 2014).
What has been called feminist hashtivism has become representative of the kind of activist struggle that characterizes this fourth feminist wave of our times. As Kitsy Dixon (2014) explained, we are seeing a new wave of feminism that operates by hash tagging. Hashtag feminism is being used "to rise above the limitations not only of women's mobility, but the limitations of being a feminist." It demonstrates its usefulness in forming communities for women who are seeking a place to express their beliefs, globally, with other women who share in their social identity. "However, as this online community is formed, questions of safe spaces, identity, and the redefining of feminism accommodate the mass popularity of feminist hash tagging" (Dixon 2014).
In her work, Taryn Riera (2015) discussed how online feminist communities are communities of validation and support, entities that allow them to do activist work by helping young women to feel affirmed in their beliefs: "They offer respite when daily marginalization, both online and off, takes its toll. Although it shares these functions with the way feminism has historically taken place offline, online communities of validation and support reach women who are not encompassed by physical-world feminist networks, and its translation occurs through memes and other Internet artifacts".
The #MeToo phenomenon is paradigmatic of this new women's movement of a postfeminist kind. It can be understood as a symbol of generational renewal (Gillis et al. 2004), a change that has led to a more collective feminist feeling (Budgeon 2001;Braithwaite 2002; McDuffie and Ames 2021) but which applies, more than ever, the feminist maxim that "the personal is political", transcending the individual subject in order to build a shared subject of struggle. #MeToo was one of the first digital challenges to consider whether "sisterhood" had finally gone global or whether it was just another product of neoliberal feminism (Ghadery 2019).
In any case, as Anabel Quan-Haase et al. (2021) pointed out, beyond #MeToo there is a need for more cross-cultural analysis in order to gain a better understanding of the movement as it evolves over time and moves into different spaces.
In fact, the complexity that online feminist activism covers in practice is one of the main reasons for continuing the recent line of analysis of feminist hashtags. The usefulness of studies into certain feminist hashtags has been demonstrated by Mendes et al. (2018), who stated that these "provide some significant insights into the promise of digital feminist activism for raising feminist consciousness and producing solidarity". The present study also aims to contribute to this line of research into what is known as feminist hashtivism by considering factors that influence this solidarity.
As well as the analytical work on rape culture tweets by Megan Stubbs-Richardson et al. (2018), especially relevant in this field are those studies which look into the usefulness of feminist hashtags for creating mobilizing conversations and promoting certain constructions and resisting others. This area has become central in terms of researching the online feminist movement from different perspectives, as is shown by the special issue "Feminisms and Social Media" of the journal Feminism and Psychology (Locke et al. 2018).
The academic literature has recently examined important occasional feminist campaigns based on hashtags that have been trending topics and which, therefore, have demonstrated a high mediatization impact for the movement: #YesAllWomen (Thrift 2014;Jackson et al. 2018), #SolidarityIsForWhiteWomen and #FemFuture (Loza 2014), #why-Istayed (Clark 2016;Linabary et al. 2020), #iamafeminist (Jinsook 2017), #MeToo and #BeenRapedNeverReported (Mendes et al. 2018). All these hashtags are representative of the moment that feminism is going through, based on a "call-out culture" that reaches broad sectors of society that are aware of a need for change; in other words, a "willingness to engage with resistance and challenges to sexism, patriarchy and other forms of oppression via feminist uptake of digital communication" (Mendes et al. 2018). In terms of Twitter's main advantages, the cited studies recognize its usefulness as a simple, free space to communicate feminist points of view and connect with people with similar ideas.
This validity of Twitter has also been recognized particularly by one of the special issues of the Feminist Media Studies journal covering the feminist struggle on Twitter and the use of feminist hashtags against gender violence (Berridge and Portwood-Stacer 2015) More recently, Jackson et al. (2018) have said that "feminist hashtags have been successful in creating an easy-to-digest shorthand that speaks to complex concerns", as is the case with regard to violence against women. According to these authors, "these hashtags provide a source of discursive and collective energy that catalyses both on and offline movement work, leading to powerful cultural repercussions and yes, change".
So, taking everything into consideration, feminist hashtags show both advantages and disadvantages (McCafferty 2011;Shulevitz and Traister 2014;Stubbs-Richardson et al. 2018). With regard to their usefulness in terms of their unifying effect, various critical voices have suggested that the very ontology of virtual relationships hinders the creation of solid organizational structures (McCafferty 2011), converting feminist messages into simple echoes within a cloud of criss-crossing messages (Shulevitz and Traister 2014). There are also those who consider that Twitter is a space that is inherently toxic and negative, not only with regard to the feminist movement, but with regard to the female subject (Shaw 2014; Jane 2017). As Dixon (2014) states, "in identifying online communities such as Twitter [...] as safe spaces for expressing feminism [sic] views and politics, its ramifications present dire consequences which lead to online harassment, hate speech, disagreements, and a miscommunication in rhetoric. It is with these consequences that the academic discourse becomes lost in transmitting the message of what feminism is and how feminists are identified." Criticisms of Twitter also refer to the ephemeral nature of hashtags, although some of them, despite being occasional, are able to offer an annual revision that allows important gender matters to be reconsidered (Clark 2016;Jinsook 2017), as is the case with the 25 November day against sexist violence. According to a recent World Health Organization report (World Health Organization WHO), about one in three (35%) women worldwide have experienced either physical and/or sexual violence from an intimate partner, or sexual violence at the hands of a non-partner in their lifetime. The fight against sexist violence continues to be one of the main areas of struggle for the feminist movement, particularly in certain socio-cultural contexts, such as the Spanish-speaking world.
Furthermore, this struggle is supported by other spheres, including academic research. According to recent work by Mendes et al. (2018, p. 244), "while we are confident that conversations around digital feminist activism against rape culture, sexism and harassment will continue, we encourage researchers to continue to explore the experiences of those who are participating in such initiatives, so we can understand the fuller picture and long-term effects and impacts of such feminist activisms".
In the specific case of feminist praxis in Spain, Sonia Núñez (2011), (Núñez et al. 2017) has examined cyberfeminist virtual communities, including those created on Twitter (Núñez et al. 2019). As these studies show, the use of new technologies has been decisive in the promotion of increased socio-political mobilisation in the fight against violence against women, as well as in the creation in this particular sphere of new relational spaces or online groups. As well as examining the differences in how these online groups were created, Núñez's work shows how they define Spanish cyberactivist praxis from different positions.
Specifically, in their study into digital activism on Twitter against gender-based violence in connection to the 8 March 2017 feminist mobilization in Spain (8M), Núñez et al. (2019) explored a sample of 20,000 messages (tweets, retweets and mentions) with the most commonly used hashtags in this context, such as #8M and #NiUnaMenos ("Not One [woman] Less"). As this work showed, giving testimony on gender-based violence offers digital feminist activism the possibility of acting in an effective, political manner on Twitter, circumventing invisibility. However, perhaps the most important thing that this study showed, in agreement with our principal hypothesis, is that these hashtags do not form strong conversational communities, but rather serve to disseminate messages at a mass scale.
In other words, the most recent research on hashtag feminism confirms how the tweets on women's experiences in a form of "herstory", together with the cultural climate created as these women and their stories came together to join in social protest at a particular moment, reinforce structures of feeling and collective commitments to this movement's issues (McDuffie and Ames 2021). However, it is not clear if those structures of feeling and collective commitments are strong enough to promote or facilitate some type of international dialogue and cohesion within the movement, an achievement that would be revealed as one of the main potentialities of hashtag feminism on Twitter.
According to our study, the Twitter discourses of the agents representing the feminist movement in English-language and Spanish-language contexts are different and, therefore, so is the kind of hashtivist mobilization carried out in the two political-cultural contexts. This is even the case with regard to activism with a strong intention to unite, such as that related to 25 November International Day for the Elimination of Violence Against Women.
So, taking as a reference the dialogue generated on Twitter at the time as the annual 25 November commemoration in the year 2018, the article examines the usefulness and value of hashtivism to the women's movement according to the different political and cultural contexts of the UK (Crozier-De Rosa 2018), representing the English-speaking world and northern Europe, and the culture of Latin and Ibero-American influence represented by Spain.
The approach, in terms of studying these differences, has been carried out based on different research questions.
RQ1 Firstly, the text asks whether geographically distinct feminists enter into dialogue on Twitter (interdiscursive cohesion level among group or individual accounts with a significantly feminist character), that is to say, whether feminist dialogue on Twitter shows an international dimension.
RQ2 The study also asks about the outstanding presence of other institutional and non-institutional actors in the public space for dialogue that Twitter has become for the feminist movement.
RQ3 A further aim has been to understand the presence and role of institutional feminism at the international level in the feminist dialogue on Twitter related to 25 November by means of an analysis of the role of the United Nations (UN) as an organization for uniting shared struggles.
The article analyzes these matters through the specific prism of the social media (Twitter), bearing in mind the importance granted by feminist-activist theorists to these platforms due to their value in terms of boosting cyberfeminism around fundamental matters such as gender violence. In this regard, the article tackles the implications of Twitter relationships or dialogues for online feminist activism in terms of a more interactive and social constructivist model of feminist (cyber)activism.

Data Extraction and Analysis Method
Structural analysis methods such as Social Network Analysis (SNA) have demonstrated their capacity for the analysis of interactions on social media as a mixed methodology, that is, both quantitative and qualitative (Crossley 2010;Edwards 2010;Tremayne 2014), considering the possibility of articulating analytical strategies that combine mathematical and computational rigour with interpretative skills characteristic of social theory.
Social media offer a large amount of data that can be easily computed as a matrix of relationships (e.g., mentions among users, "friend" or "follower" relationships among users or relations set up between users and content), and so the data extracted from social media becomes a big data source through user-platform and user-user interactions (e.g., publications, mentions, likes, swipes or shares).
Specifically, our research focused on the microblogging social platform Twitter and was carried out using the Twitter Capture and Analysis Toolset, developed by the University of Amsterdam's Digital Methods Initiative (DMI-TCAT) (Borra and Rieder 2014). This has been considered to be one of the most useful open Twitter analysis tools (Felt 2016) since it permits the capture of a wide variety of information from the Twitter social networking service in real time or retrospectively This offers an unprecedented level of access to the store of online human activity over time (Del Fresno-García 2014). After all, as pointed out above, Twitter makes it possible to observe the social or interpersonal relationships that provide the connection between individuals or organizations, that is, the forms in which actors are linked, which determines the general structure of the network, its groups and the positions of the individual profiles within it.
The use of the Twitter Capture and Analysis Toolset made it possible to monitor the flow of interactions on the Twitter social networking service and to download the data regarding the following hashtags between 22 and 28 November 2018 (that is, from three days before to three days after the 25 November commemoration): #25N; #25nfeminista; #Diacontralaviolenciadegenero; #Iflovehurts; #Niunamas; #Stopviolenceagainstwomen; #Violenciacontralamujer; #Whiteribbonday. Activity related to the following combinations of words was also monitored, capturing any tweet that contained both concepts (not necessarily in the following order): mujeres + violencia (Spanish); women + violence (English); Noviembre + 25 (Spanish); November + 25 (English).
Using this protocol, a total of 881,887 tweets were captured around the world in English and Spanish, in a conversation involving 473,157 unique users. These tweets contained a total of 705,769 unique mentions among 439,859 unique users, which were entered into the Gephi social network analysis software.
In order to facilitate viewing the network and the interpretation of the phenomenon analysed (Venturini et al. 2019), the Force Atlas 2 (Jacomy et al. 2014) algorithm has also been applied. This measures proximity between those nodes that have interacted, and the distance between those that have not done so. In identifying communities, the role of the analyst consists of interpreting the results of a process of grouping nodes based on the patterns that the algorithm itself is autonomously able to identify in the data, using software such as Gephi.
Bearing in mind the goals of this study, as well as the scientific context in which it is set, the problem involved an exclusive methodological design, developed ad hoc to examine this study's specific problem. In order to identify the importance of users, their "degree of entry" was taken; this is the number of times that each user is mentioned by other users of the service. In order to identify the conversation that each user participated in, the Louvain algorithm for community detection (Blondel et al. 2008) was applied.
Furthermore, in this study, the concept of social homophily takes on particular importance; this is the tendency of individuals to relate more or have more relational links with those who have similar social characteristics (Verd et al. 2014). Thus, in this study, the "degree of intercommunity homophily" was identified for each pair of communities using the E-I index (Krackhardt and Stern 1988). This index provides a figure between 0 and ±1, where a value of -1 indicates perfect homophily (users are only linked with actors from their own community), a value of 0 indicates neutrality or indifference (users are equally linked through nodes of their own community and those of the other) and a value of 1 indicates perfect heterophily (users are only linked with users of the other community).
Given that the communities have been calculated using the Louvain Multilevel algorithm which generates communities with homophilic criteria, the key point about this last piece of evidence lies in seeing the differences between pairs of communities. In other words, there is an examination of which pairs of communities are the least homophilic, given that the occurrence of high or very high levels of intercommunity homophily were to be expected.

Results
Looking at Twitter, a total of 11 communities were identified that contained over 2% of the service's users or nodes. These included a total of 44.73% of the nodes. A first result, then, was that over half of the network's participants made their contribution in small or isolated conversations (i.e., they constitute communities that include less than 2% of users, involving very few interactions among very few users). This, in turn, means a strong trend towards horizontality: the conversation's leaders are not hegemonic, but rather are diverse and have low shares of power. In short, what was found was a highly dispersed conversation without a single discourse forming a backbone. The communities with more than the 2% threshold of users are identified in Table 1 and Table  S1 (Supplementary Materials). When these communities were analysed one by one, based on their leaders and most popular tweets, the main topics of conversation in each one were identified: The first community was led by @un_women and offered various messages of an institutional kind related to 25 November. The second community was led by the Spanish activist Bebi Fernández, who stood out for her role in promoting feminist mobilization related to 25 November in Spain. Along similar lines, the third community was led by @huelgafeminista, in this case with an activist style and a goal of organized mobilization. The UN's account in Mexico led the fourth community, contributing to mobilization in the Latin American sphere from an institutional point of view. The fifth community was led by the athlete @luciarguezf, with fewer than a thousand followers. The community focused on sharing memes or graphic content against gender violence. The sixth community was led by a homonym of community number 1, the UN's institutional profile in Spanish. The seventh community @whiteribbonscot is an account that is the mouthpiece of a campaign to engage men in bringing an end to sexist violence. The eighth community, led by @botonica, is the most striking, since it is not an account that specifically covers matters related to violence against women: this is an account that publishes poetry, which achieved a particular engagement with the 25 November event. The ninth community is led by the account @uniciudadanaar, closely linked to protests in Argentina, just as community number 3 is related to protests in Spain. The tenth community was led by the account @rubnpulido, with messages and topics related to the far right in Spain. The last community was led by an account promoting Catalan independence @cupnacional.
These results indicate that 25 November created feminist mobilization and discourses against gender violence from different perspectives, some of which are institutional and some not, and some of which are alternative and some not, but bearing in mind that in all cases this activism shows a certain degree of organization. Once the characteristics of each community had been examined according to the accounts that lead each of them, the study looked at the level of dialogue that was established among them. In each case, the aim was to identify an approach in terms of the basic characteristics of each community or, where appropriate, to identify certain differences with regard to its cause or justification in the different features, viewpoints or even functions of each feminist community within the movement's debate and even within the social and political debate in which this movement is embedded, in both theory and action.
The following graph (Figure 1) shows the interactions established among the conversation's participants: each user is a node and each mention between users is an edge. The colours indicate the community concerned, whose names are written on the image and in the tables. Likewise, in this graph it is possible to see some of the details set out in the next double-entry or contingency table (Table S1-Supplementary Materials), particularly the enormous disconnect among communities. We are looking at a conversation that has very few points of connection among different communities, including those which seem to have greater ideological or discursive proximity, as is the case of the communities led by Protests in Mexico @onumujeresmx and Protests in Argentina @uniciuddadanaar, or the case of @un_women and @onumujeres. On the other hand, the graph makes it possible to see clearly the many and varied discursive leaderships established in relation to such an important commemoration.  Table S2 (Supplementary Materials) shows almost perfect levels of intercommunity homophily (i.e., close to -1). This is for two reasons: the first is that, as has been pointed out, the conversation is highly dispersed; the second is that the Louvain Multilevel algorithm identifies communities using a procedure that stimulates homophilic grouping. The aim, then, is not to show that homophily exists, since this derives from the very method used, but to identify precisely which intersections between clusters have the least pronounced or lowest homophily values, that is to say, where the homophily is less homophilic.
In this way, it is possible to see that the intersection between the Protests in Spain communities number 1 and 2 is the least homophilic, given the possibilities of homophily described. The members of the Protests in Spain (1) community have also established a  Table S2 (Supplementary Materials) shows almost perfect levels of intercommunity homophily (i.e., close to -1). This is for two reasons: the first is that, as has been pointed out, the conversation is highly dispersed; the second is that the Louvain Multilevel algorithm identifies communities using a procedure that stimulates homophilic grouping. The aim, then, is not to show that homophily exists, since this derives from the very method used, but to identify precisely which intersections between clusters have the least pronounced or lowest homophily values, that is to say, where the homophily is less homophilic.
In this way, it is possible to see that the intersection between the Protests in Spain communities number 1 and 2 is the least homophilic, given the possibilities of homophily described. The members of the Protests in Spain (1) community have also established a significant number of interactions with users of the Memes and Protests in Argentina communities. The members of the Protests in Spain community number 2 have interacted with members of the Catalan-speaking community, the Onumujeres community, and also with the clearly anti-feminist Spanish far right. This last connection makes clear that interactions between communities can be established in negative or very negative terms (insults, etc.). References with derogatory aims also occur. This means that the interactions established are not necessarily positive.
Elsewhere, it is possible to find significant interactions between the clusters in English, UN Women and Activism in English, revealing a linguistic homophily in the network. Those accounts that speak the same language tend to communicate more among themselves, although it is true that UN messages in English and in Spanish create certain common interactions. Thus, connections exist between the Protests in Mexico and the Protests in Spain communities, and the Protests in Spain and Protests in Argentina communities. Finally, there are a number of communities that have slight interaction with the Memes community, which are the Protests in Argentina, the Protests in Spain and the Poetry communities, suggesting a certain synergy among native Twitter contents.

Discussion and Conclusions
It is possible to use the feminist demands made through the hashtivism generated on Twitter around the 25 November 2018 commemorative day to characterize contemporary feminism by the kind of digital dialogue or conversation it creates. In this regard, according to the results obtained in this case study, today's Internet feminism is not characterized according to the features that have historically defined it in the offline sphere. Based on the case analysed, Internet feminism has shown itself to be a movement uninfluenced by major leaders or ideological reference points.
Furthermore, the study's results call into question Twitter's validity as a space for dialogue that can boost feminism as a movement for action in the streets, as well as its validity as a tool that can unite and encourage greater cohesion among members and defenders at an international level. This study, then, joins a new analytical framework in which various analyses of what is known as "feminist hashtivism" question the phenomenon's usefulness for uniting a new generation of activist women as an intermediate step for achieving a greater degree of mobilization and awareness with regard to the struggle for equality (Fotopoulou 2014;Shulevitz and Traister 2014) at the international level.
In general terms, no clear or outstanding dialogue among feminist groups at the international level was perceived, and even less so at the interlingual level (RQ1). A degree of good practice, in terms of cohesion, can be observed among some groups internationally, with a greater vitality within feminism in Spain and the Latin American world, shown by the greater density of relationships among members of certain clusters despite the very compartmentalized scenario. By contrast, in the English-speaking world, no particular volume of similar activity or interactions was observed in any case. This may be due to the tradition of Spanish feminism as a feminism that has come about based on the emergence of different street and protest-based activist movements in the context of the post-Francoist period and the transition to democracy, coinciding in time with the second feminist wave of the 1970s (Women's Liberation Movement).
Furthermore, it is important to bear in mind that in Spain feminism is going through a period of energetic activity, among other things, due to the continuing existence of a high incidence of violence against women (OECD 2019) and an adverse legal climate, as demonstrated by the high-profile case of the gang rape of a young woman during the San Fermín festival in Pamplona in 2016. This form of organizing in Spain, which revolves around mobilization and street action, contrasts with the model in the English-speaking world, where NGOs and lobbies aim to influence society by means of other models: specific campaigns, raising awareness, small-scale protests, etc.
Feminism in the Spanish-speaking world, then, presents a more subversive character that tends towards protest, unlike that in English-speaking countries where the struggle for equality in relation to the 25 November event is channelled through associations and organizations that are integrated into the structure of the social, political and cultural system (RQ1).
The study has also shown how hashtivism increases the presence of non-institutional actors in the public space (RQ2). In the Spanish case, this means the "digital supremacy" of autonomous activist groups linked to the left wing. In the case of the United Kingdom, this means NGOs and lobbies that work in favour of women's rights, as well as other similar initiatives of a more individualist-liberal kind than in the Spanish-speaking world (e.g., men against abuse, women against female circumcision, etc.). This also shows a clear absence of transnational activist cohesion. It is even possible to talk about the existence of structural gaps between activists in different countries and different languages (e.g., activism in English and protests in Spain). Despite these differences, Twitter nonetheless manages to be a channel of feminist dialogue through official mouthpieces such as the UN, making it a kind of interlingual tool for dialogue on the Internet (RQ3).
With regard to the level and degree of homophily examined, there is a certain trend towards linguistic homophily, as well as the use of certain codes or contents, specifically memes, short sentences and protests. The element of interlingual cohesion is the UN and not, curiously, the protests and memes-something that is certainly significant. In fact, the most active feminism of the first and second waves was characterized by using street protest, humour and satire in their public imagery and representations, etc.
Another of the main trends observed is the isolation of the far right, which is hardly connected to any cluster apart from the Protests in Spain community number 2. This is probably due to the "don't feed the troll" maxim that is well known to hashtivists and which produces highly differentiated group behaviours among antagonistic ideological profiles.
In short, according to the general premise that has guided this research which defines hashtivism as "activist efforts that take place online, or utilize online networks to promote communication, action, or awareness" (Gunn 2015), the examined case shows the value of online activism to promote collective solidarity, uniting the feminist movement around one of its most important struggles: violence against women. At the same time, it is effective when it comes to favouring or reinforcing its autonomy on both a macro level (independence of the movement in relation to other movements and social, political, economic or other influences) and a micro level (autonomy of the movement in relation to the feminist struggle's model, based on different cultural and political contexts, for example English-language and Spanish-language feminism).
Even so, in line with previous research (Boyd and Ellison 2007;Page 2012), the results obtained in this case study demonstrate how feminist hashtivism is limited in terms of promoting a solid or strong sisterhood based on consistent dialogues. As results show, this is something difficult to achieve due to the polarization and self-centred uses that Twitter's digital network promotes. Taking into account these constraints and the limitations of this specific case study, in terms of future research lines it might be worth noting here how feminist hashtivism will continue to be a front-line research topic to determine how hashtags can help to promote global dialogue and subsequent online and offline action, beyond its proven usefulness in promoting global communication, visibility and awareness with respect to postfeminist causes.