Active Citizenship and Hate Speech on Social Media in the Context of Romanian Family Referendum

3 Assistant Professor, Ph.D, West University of Timişoara, Department of Philosophy and Communication Sciences, Timisoara, Romania, E-mail: corina.sirb@e-uvt.ro Abstract: Active citizenship is one of the most interesting characteristics of young generations worldwide and is strongly connected to digital media usage. Internet gives people a voice to speak up about stressing community problems, unite in groups with the same views for a stronger impact and conduct true lobby campaigns to promote and sustain their ideas. However tangible, utopic or doubtful this democratic way of civic involvement may appear for a global digital community, its manifestation is still problematic, as it nurtures, in some cases, hate speech and extreme polarization of opinions. To support this idea, we conducted a study that analysed civic activism and the values promoted in the context of Romanian Family Referendum. Using a qualitative methodology based on communication sciences techniques and semiotic investigation, we analysed the type of discourse used in social media and the frequency of hate speech and undemocratic behaviour. At the end of this study, we will outline some concluding ideas regarding the discussed topic.


Conceptual clarifications
Civic activation is one of the most interesting characteristics of young generations worldwide and is strongly connected to digital media usage (Sirb, 2017). Online media, starting with blogs and continuing with the explosion of social networks and multimedia user generated content, became the biggest and most powerful agoras ever seen, where citizens discuss political issues, with the intent of making their voices heard and generate social change (Bader, 2018;Momoc, 2014;Sirb, 2017). That is why political and social topics are one of the most popular topics on social media in Romania (Sirb, 2017). However, the present paper is not focused on the problem of finding out whether online hate speech should be censored or regulated in any other forms, but to show that, contrary to the common belief and its immense potential, sometimes Romanian social media is not used as a space to debate public issues, but rather as a tool for unloading frustration. Moreover, it appears that sensitive issues bring about extreme narratives rather than sensible discussions. This was also the case for the Romanian Family Referendum, as we will further see in the case study.
The main questions that guided our research were the following: 1. What is the frequency and distribution of hate speech in the context of approaching a major social issue on Romanian digital media? 2. Is hate speech connected to certain political or moral values? 3. Does hate speech has a specific narrative structure and/or function?
Before diving in our quantitative and qualitative analysis, it is necessary to define the key terms used here. There are many definitions for hate speech, from very specific ones to more vague depictions, but all refer to discourse designed to call attention to, and to manipulate, social differences. (Foxman & Wolf, 2013;Waltman & Haas, 2011). This concept has been an important issue in public discussion for a long time, but it has become considerably more visible once digital communication became an inherent feature of our society. This is quite natural, if we think about the way social media works: it gathers people with similar interests and way of thinking in small communities and feeds them with information and ideas which they most likely agree with. Digital communication forms groups, one way or the other and hate speech is all about group communication: "Hate speech discursively constructs the in-group in extremely positive terms while also constructing the out-group in dehumanized terms, including characterizing the out-group as posing a threat to the in-group and cherished in-group values and traditions. Hate speech may serve a variety of different social and political purposes. First, hate speech may serve the simple purpose of intimidating those who belong to an out-group because of their race, ethnicity, religion, gender, gender identity, or sexual orientation. (Waltman & Haas, 2011).
Moreover, digital media has been the main space to nurture hate speech, as users feel: 1) detached (in contrast to direct communication, digital dialogue is regarded as distant, which explains the courage of expressing oneself); 2) protected (the effects of an insult are insignificant for the aggressor); 3) empowered (by the reactions from the people from the in-group; ) 4) free (due to lack of legal reglementations). As it will be seen in our research, the analysis we pursued reflects Waltman & Haas perspective on hate speech as always being a discourse constructed between an in-group and an out-group.
In addition, we dare to claim that our research brings more depth to the problem of hate speech by its interdisciplinary approach on the one hand, and by combining quantitative methods and semiotic analysis, on the other.
As far as relevance is concerned, our paper draws attention towards a sensitive, controversial issue of global importance: digital media nurtures the exchange of opinions and civic activism, but it does not solve the problem of protecting diversity and minorities. On the contrary, it sometimes offers the tools and context to facilitating hate speech. While the European Union has been struggling to find the proper means and measures to regulate digital dialogue, it appears that when it comes to subjects of high public interest, social media still fosters both constructive dialogue and bigoted attitudes.

Case Study: Quantitative analysis based on keywords related to the referendum in online media
Between 6 -7 October 2018 in Romania took place a so-called "Referendum for Family" initiated by the government party, Social Democrat Party (PSD). The aim was to define a marriage only when two persons of different sex are involved (heterosexual definition of family). Debates about this item degenerated and the for and against opinions on both sides arrived in online media in forms and manners which often went beyond a civilized debate. To follow up the tendencies in online discourse about this event we appealed to a professional analysis offered by courtesy of a specialized monitoring company of online media, Zelist Monitor.
The greatest number of posts connected with this item were on facebook (125083); understandable, as it is the most used social media network in Romania. Great number of posts, almost similar, appear in online press and aggregators 80541 and 79102); much less on Twitter (28398) and very few on Pinterest (70). Following the evolution in time, the greatest number of posts was in 7 October 2018, the second day of referendum (a total of 25139), which represented a peak of interest (see fig.  1). The daily evolution can be seen on: https://www.zelist.ro/live/referendum To pursue an analysis of the two presumed opposite tendencies we choose as keywords /familiatraditionala/ and /gay/ and observed their occurrence between 24 September -5 October 2018. We chose these two keywords as representative because the first was the sintagma for naming the referendum and mostly used by those who manifests a conservative mentality purely heterosexual. The second keyword, /gay/, was chosen because in Romanian culture of 2018 the term LGBT is seldom used and /gay/ is often used with a pejorative and discriminative connotation, not just simply denotative for homosexuality. In this way, we tried to evaluate quantitatively the discriminative discourse in the studied period.  . 2).
Surprisingly, the keyword /gay/ surpass the first one: 10506 posts, most of them on facebook (4459). The users and viewers had the same demographic distribution (30 -34 y.a., Bucharest, 62,8%) like in figure 3. That shows the pejorative connotation was more used than the affirmative term /familiatraditionala/ which means that the number of posts with a potential hate speech was great then political correct speech.

Content analysis
When we examined the exported content of the selected posts, we identified two main three main trends, as far as communication functions are concerned.
Firstly, there are posts with a clear pragmatic function, that aimed to modify other users behaviour and produce real effects: either to participate in the referendum or promote boycotting it.
Secondly, there are posts with a self-descriptive function. In other words, users took advantage of the public interest in the referendum to say something about themselves, rather than the topic of discussion. The social issue was used solely as a pretext to build a personal image. This is mostly the case for artists, politicians or digital influencers.
Last, but not least, and strongly connected with the previous communicative function, we identified posts with a social function. These were constructed in such a manner that they could attract similar views and create micro communities around them. They are more mobilising, propose keywords and use a more powerful vocabulary.

Hate Speech Narrative Patterns and Rhetorical Tactics
As far as the persuasive tactics are concerned, we identified a few major rhetorical and narrative structures:

The Disclaimer: "I am not homophobic/atheist, but I believe that…"
This pattern could be identified in both sides of the discussed topic and is situated at the low-end of hate speech. In other words, even though it is one of the least aggressive type of comments, it led to the same bigoted content. Most often, this type of content was used with a self-descriptive function.

Appeal to personal experience: "I am talking this way because I know them"
This persuasive tactic was more powerful than the first and it most often invoked an extreme behaviour or a specific event of the out-group, followed by a generalized conclusion that stated that the particular manifestation was the norm. In some cases, this flawed argument was continued by a slippery slope, arguing that allowing that kind of behaviour would finally degenerate in a disastrous situation. These posts were mostly used with a pragmatic function.

Appeal to majority: "Most people think this way, but I am one of the few who have the courage to express themselves"/ "I speak for all/most of us when I state that"
The appeal to majority was most commonly used with a social function, as the users who followed this pattern positioned themselves as representatives of the in-group, that do not hesitate to speak up their mind and defend the interests of the majority they represent.
A relevant aspect in the context of the Romanian Family Referendum is represented by posters promoted during this event, posters that have been broadcast not only online, on various social networks, web sites and blogs, but as well off-line, by placing banners or posters in central urban areas among others. In the following, we shall apply semiotics research for a content analysis of the visual messages transmitted by four posters promoted during the event, of which two are pro referendum posters and two in opposition. We mention that these posters were among the most broadcast and they were the main focus during that period, which motivated our choosing them for this research. The semiotic analysis of these posters shall focus on the content of symbolic messages transmitted through non-verbal visual signs, namely: image, characters, chromatics, and graphics. In this analysis, we are interested in the signification of the visual message as a whole, especially the moral and socio-cultural values transmitted, the political background of which has already been discussed previously.
We will start with the semiotic analysis of the first official Romanian Family Referendum poster distributed online (https://www.facebook.com/ ilfovpsd/photos/a.283035111856841/1085559218271089/?type=3&theater). The poster is comprised of the following types of nonverbal signs: iconic, chromatic and symbolic signs. In the central part of the poster there is a photographic image, symbolically representing the family. The way in which it is represented is an ideological one, in that we are shown a bi-parental family with a child, whose gestures and mimics express the notion of happiness. The colour palette of the poster is made up of red, yellow (proportionally represented the most, to attract the eye), blue, these being symbolically arranged as the flag of Romania. We also notice the black and white noncolours and the grey at the written message level of the poster. The poster is marked on the top and bottom with distinctive graphic signs, representing Romanian folklore symbols. All these nonverbal signs symbolically emphasize the idea of the traditional heterosexual family, the concept of national specificity, conservative spirit, traditional moral, cultural and social values, implicitly marginalizing and thus discriminating other family structures and adherence to values other than the traditional values.
The second pro-Romanian Family Referendum poster draws attention primarily through the size of the central image and the minimal chromatic look (https://www.facebook.com/photo.php?fbid=1021261528 4884894&set=a.4075542334740&type=3&theater).
The central picture is, as in the aforementioned poster, the notion of a bi-parental family, but it is engaged with the possibilities offered by digitization (reinforced by the symbolic positioning of the laptop in the hands of all three members of the family). As far as the poster's colour is concerned, we can see different cool shades of blue, varying from dark blue to very light blue. Under symbolic terms, dark blue expresses loyalty, protection, tradition, conservatism (Chevalier & Gheerbrant, 2009), also emphasised by using it as the font colour for the poster's slogan. Although visually and in terms of narrative structure, this poster is somewhat more moderate than the previous one, nevertheless the non-verbal message has the same aspect of marginalization and discrimination against non-traditional family types as well as the idealization of the ethical values usually associated with the traditional family.
The third poster we submit to the analysis is among the most recognizable in terms of the anti-discrimination message and its support for the social inclusion of various marginalized categories (https://www.anosr.ro/pozitia-anosr-privind-referendumul-for-revizuireaconstitutiei-romaniei/). The graphic concept of the poster is minimalist, composed of visual symbols representing multiple and diverse family groups, rendered as animated characters displayed in a circular pattern. The means of visual expression of the characters in the image (posture, silhouette, anatomy, facial expression, and chromatics) accentuate the antihate speech slogan of the poster, which draws attention to the social need of tolerance, relationship, communion, acceptance and social inclusion of diversity and to enlarge the notion of family (https://www.anosr.ro/wpcontent/uploads/2018/09/Pozi%C8%9Bie-referendum-converted-1.pdf).
We can also find minimalist graphics in the second poster that held an anti-discriminative position (https://www.facebook.com/mozaiqromania/ photos/a.1228537373826607/2357218644291802/?type=3). The central nonverbal sign is that of the white heart (love, compassion) played against the black background, which, symbolically, expresses the prevalence of love as a fundamental human value in relation to other types of values. The brightness and shine of the turquoise blue shade in the backdrop of the poster symbolically represents life, protection, compassion, emotional relaxation, adaptation, spiritual opening, happiness, the pride of taking a stance for something (Chevalier & Gheerbrant, 2009).
In the context of our research, we particularly appreciate the semiotic perspective of visual messages, since nonverbal communication does not only fulfil the function of completing verbal messages and conveying expressivity, but also of retaining attention and persuasion (Barry, 1997;Petrovici, 2014). Alongside verbal messages, nonverbal discursive content can have a considerable emotional and conceptual impact, playing a significant role in developing inter-human and inter-social interactions (Joannes, 2009: 28).

Conclusions
However utopic this democratic way of civic involvement may appear for a global digital community, its manifestation is still problematic, as it nurtures, in some cases, hate speech and extreme polarization of opinions. It is not hard to observe, when one studies Romanian online narratives regarding social and political issues that the above mentioned instruments are not always used as tools to generate social change; that dialogue does not always function as a democratic power and citizens are not always respecting political and democratic values. Romania has one of the most polarized narratives, and this is a fact that can be easily grasped when reading social media feeds, and which is also confirmed by a series of studies conducted by Reuters (Reuters Institute for the Study of Journalism, & Oxford University, 2018). In other words, approaching social issues is often not a manifestation of empowered citizens, but turns into hate speech directed at the opposite teams, regardless the subject of matter.