Source: extracted from Gallardo-Paúls (2018). Self-made.
10.4185/RLCS-2020-1440
Artículo
THE PSEUDO-POLITICAL DISCOURSE OF THE SECOND SCREEN. #ELDEBATEENRTVE SEEN THROUGH ITS PROSUMERS
EL DISCURSO PSEUDOPOLÍTICO DE LA SEGUNDA PANTALLA. #ELDEBATEENRTVE VISTO A TRAVÉS DE SUS PROSUMERS
Paz Villar-Hernández1
1University of Valencia. Spain
ABSTRACT
Introduction. Our object of analysis is the conversation held on Twitter in one of the election debates in the campaign in the Spanish General Elections of April 28th, 2019; precisely, the one that took place on April 22 on the public broadcasting corporation Radiotelevisión Española among the leaders of the four main political parties: the ruling Socialists, the conservative People’s Party (PP), the centre-right Ciudadanos and the left-wing Unidas-Podemos.
Methodology. The methodology used in the work is the Pragmatic Discourse Analysis, and quantitative and qualitative analysis techniques are used. The data was coded and analyzed with Atlas.ti software. For the analysis, a total of 1000 tweets were randomly selected from among the most popular Twitter debate hashtags: #ElDebateEnRTVE and #ELDEBATEenRTVE.
Results. The results indicate that the four parties participated in the second screen conversation generated on Twitter. The parties mostly preferred to praise their candidate, instead of attacking others that was the sample results. On the other part was Ciudadanos, who decided to the attack the PSOE and only afterwards to praise their candidate.
Debate and conclusions. As for the issues discussed by audiences, they were not exactly political but pseudo-political. The memes, ironic messages and jokes were not especially prominent in this transmedia debate.
KEYWORDS: Spanish general election; April 28th; electoral debate; Twitter; pragmatic discourse analysis; pseudo-politics.
RESUMEN
Introducción. Nuestro objeto de análisis es la conversación mantenida en Twitter en uno de los debates electorales de las elecciones generales del 28 de abril: el que tuvo lugar el 22 de abril en Televisión Española entre los candidatos de los cuatro principales partidos políticos: PP, PSOE, UP y Cs.
Metodología. La metodología empleada en el trabajo es el análisis del discurso de base pragmática, en la que se emplean técnicas cuantitativas y cualitativas de análisis. Para este, se seleccionaron de forma aleatoria 1000 tuits de entre los más populares que habían empleado los hashtags promocionales del debate: #ElDebateEnRTVE y #ELDEBATEenRTVE.
Resultados. Los resultados indican que los cuatro partidos analizados participaron en el debate paralelo generado en Twitter y que, si bien la mayor parte de las intervenciones en la segunda pantalla fueron de ataque, los partidos prefirieron sobre todo elogiar a su candidato, a excepción de Ciudadanos, que optó por el ataque al PSOE y luego por el elogio propio.
Discusión y conclusiones. En cuanto a los temas tratados fueron no tanto políticos sino pseudopolíticos. Los memes, mensajes irónicos y bromas ocuparon un espacio reducido en esta conversación transmediática.
PALABRAS CLAVE: elecciones generales; 28 de abril; debate electoral; Twitter; análisis pragmático del discurso; pseudopolítica.
Correspondencia:
Paz Villar-Hernández. Universidad de Valencia. España.
paz.villar@uv.es
Received: 12/09/2019.
Accepted: 20/10/2019.
Published: 30/04/2020.
This article has been prepared within the framework of the research project: PRODISNET-02. Discursive processes on the Internet: Enunciative displacements and hyperbolic effects in political discourse. RTI2018-093523-B-100, of the Ministry of Science, Innovation, and Universities.
How to cite this article / Standard reference
Villar-Hernández, P. (2020). The pseudo-political discourse of the second screen. #ElDebateEnRTVE seen through its prosumers. Revista Latina de Comunicación Social, (76), 121-141. https://www.doi.org/10.4185/RLCS-2020-1440
CONTENTS
1. Introduction and Theoretical Framework 2. Methodology. 3. Results. 4. Discussion and conclusions. 5. Bibliography.
Translation by Paula González (Universidad Católica Andrés Bello, Venezuela).
1. Introduction and Theoretical Framework
The televised electoral debate between candidates for the presidency of a country is a deeply rooted tradition in places like the United States -where the model emerged- Germany or France, although until recently it was not as much in countries like Spain (Cantavella-Blasco et al., 2008). In this country, the first televised debate took place on May 24th, 1993, in the studios of Antena 3. The president of the Government at that time, Felipe González, and the candidate of the main opposition party, José María Aznar, were summoned. This debate was followed by another, days later (on May 31st), in the other large private network, Tele 5. However, after those two initial debates, the silence was made and we had to wait 15 years to live a new dialectical confrontation between candidates to the Presidency of the Government of Spain.
The year 2011 was especially transcendent for two reasons: on the one hand, because one of those debates incorporated representatives of other political forces, starting a trend that has been consolidated over the years and, on the other hand, because the electoral debate underwent a transformation because of the influence of the parallel debate on Twitter. Although it will be in the debates and campaigns of 2015 and 2016 when that relationship is boosted and consolidated, and the conversation of the second screen leads the way to Social TV or Television 2.0 (Ruiz del Olmo and Bustos Díaz, 2017). This is how the debate that we analyzed here occurred, that of April 22nd, 2019, which brought together the candidates for the Presidency of the Government of the PSOE, PP, Cs, and Unidas Podemos.
There are many perspectives from which the electoral debates in Spain have been analyzed; debates that, in the opinion of Fernández García (1999, p. 84) are rather a set of “non-dialogic structures” or an achievement of “juxtaposed speeches”. This author, who also analyzed the debates between Rodríguez Zapatero and Mariano Rajoy (2008) (Fernández García, 2009) and that of 2011 between Rubalcaba and Rajoy (Fernández García, 2009) is one of many who have linguistically analyzed this conversational moment. This perspective has a wide tradition in Spain that includes, from the initial works of Blas Arroyo regarding the first debate of 1993 (1998, 1999, 2003), to those of later dialectical contests (Cabrejas-Peñuela, 2015; Cabrejas-Peñuelas, 2015; Cabrejas-Peñuelas and Díez-Prados, 2014; Cantavella-Blasco et al., 2008; Cuenca and Marín, 2015; Díez-Prados and Cabrejas-Peñuelas, 2018). From the communication field, we found researches that deal with its formal, legislative or regulatory aspects (Rúas Araújo, Fernández Holgado, and Alén Amil, 2018), others on the influence of the debate on the intention to vote (Callejón, 2001), its audiovisual narrative structure (Quintas Froufe and Quintas Froufe, 2010), media coverage made by traditional media (García-Marín, 2015), the thematic agenda of some of them (López-García, Llorca-Abad, Valera-Ordaz, and Peris- Blanes, 2018), and the synergies between the televised debate and its commentary on the second screen (Ruiz del Olmo and Bustos Díaz, 2017).
Studies on the electoral debate commented through social networks such as Twitter, logically, have a lower tradition (Vergeer and Franses, 2016). This communicative-discursive practice of generating comments on social networks resulting from the viewing of television programs has been referred to by some as “viewertaria” (Anstead and O’Loughlin, 2011), or “sofalising” (Doughty, Rowland, and Lawson, 2012), and is widely known as the phenomenon of the second or double screen (Doughty et al., 2012; Elmer, 2012). In itself, it has radically changed the way we watch television. Now, audiences can assess in real-time what is happening on television, acting as “backchannel” (Kalsnes, 2014; Pedersen et al., 2015) and providing segmented and detailed information about their response to each of the messages that the candidates transfer from the screen (Vergeer and Franses, 2016), to the point that this practice: “is shaping the political realm across the world” (Gil de Zúñiga and Liu, 2017, p. 215).
Its scope is evident because, despite the different motivations that guide its use, depending on whether it is a country or another (Gil de Zúñiga and Liu, 2017), the truth is that it is an increasingly consolidated practice in all types of televised retransmissions and that it generates a huge volume of tweets. Campaign teams are aware that they should not lose sight of this phenomenon, and not only during the electoral period.
In recent years, research on Twitter’s behavior during televised electoral debates has been appearing. The first appeared in the United States (Shamma, Kennedy, and Churchill, 2009) and Great Britain (Chadwick, 2011; Vaccari, Chadwick, and O’Loughlin, 2015). These are soon joined by others from Canada (Elmer, 2012), Germany (Trilling, 2015), Holland (Vergeer and Franses, 2016), Italy (Bentivegna and Marchetti, 2015), Norway (Kalsnes, Krumsvik, and Storsul, 2014), Spain (Ruiz del Olmo and Bustos Díaz, 2017), and other types of contrastive works arise, such as the one of Wells et al, (2016) in which distinctive features of the discourse on the second screen in France and the United States are compared.
In the line already pointed out by Nee (2013), some have confirmed that the most active users in the political debate on the second screen are also those that show a greater civic and political commitment (Chadwick, O’Loughlin, and Vaccari, 2017; Vaccari et al., 2015; Vergeer and Franses, 2016). Chadwick, O’Loughlin, and Vaccari (2017) point to the existence of gender differences because, while women say they use networks to find information and configure their vote, men mainly seek to persuade others, extend their influence, assert their judgment, by intervening in the conversation on the second screen. Horsch-Dayican, Amrit, Aarts, and Dassen in their research on the Dutch campaign of 2012 also point out this persuasive feature of the anonymous citizen’s interventions on Twitter, although probably the most striking is the negative nature of much of the citizen conversation in this network: “It was interesting to see that negative messages have the biggest share among tweets from citizen users. The analysis of the contents of these tweets suggested that citizens ’negative campaigning is rather an expressive act and, in this way, it differs from online persuasive campaigning efforts” (Hosch-Dayican, Amrit, Aarts, and Dassen, 2014, p. 147).
The discourse that is created in a network (second discursive level) about the electoral debate, finds in these social media, different from the traditional media but, as Campos Freire (2008) indicates, complementary to them, a new different and differentiated diffusion space, because it occurs in another technological environment that implies formal, contextual, linguistic, and content differences.
Being aware of our discrepancies with the restrictive and widely established definition of “political discourse” that Van Dijk formulates as: “the discourse of politicians” (van Dijk, 2003, p. 212), and that dates back to Aristotle, the textual universe of tweets that deal with politics and that take as a conversational reference a political genre such as the electoral debate, could also be understood as political discourse in a lax sense (Gallardo-Paúls, 2018). Beatriz Sarlo (2011) speaks of the sharpness, brevity, and vivacity of the statements as fundamental qualities of the expression in networks, and irony and sarcasm as consubstantial elements. Gallardo-Paúls and Enguix-Oliver (2016) classify it as “pseudo-political discourse”, arguing that in the networks there is a displacement (Gallardo-Paúls, 2018) regarding what the classic canon has defended as political discourse. This discourse on politics is characterized by accentuating personalism, de-ideologization, and spectacularization (Gallardo-Paúls and Enguix Oliver, 2016), an aspect on which others agree (Vallespín, 2015). On a discursive level that translates into a “logorrheic production, lack of structure, expressive condensation, overinterpretation, hypersignification, hypermilitancy, and affiliate encapsulation” (Gallardo-Paúls, 2018, p. 10); we thus speak of a primacy of the ilocutive function over the referential or propositional one, the absorption of the statement by the enunciation, an ilocutivity based on inference, the expressive condensation or the expressive ilocutivity over the representative one, and a greater incidence on the negative nuances of the speech. This unraveling of the linguistic features that make up the pseudo-political discourse continues along the lines of Mayer’s approach (2002), who warned of the risks that the mediation of politics could infer to political debate. It is the risk of the triumph of theatricality, appearances, and gestures in the face of real political action, to the facts.
The following table Table 1) explains which linguistic features are accentuated, and strengthen personalism, de-ideologization, and spectacularization, framing the different moments of the communicative process.
Table 1. Linguistic features that define the “pseudo-political” discourse according to Gallardo-Paúls (2018).
Source: extracted from Gallardo-Paúls (2018). Self-made.
After reviewing previous research about the discourse generated by the Twitter user during its viewing of television productions, among others, some electoral debates in countries around us, and analyzing what linguistic features define the pseudo-political discourse, we will present this study.
2. Methodology
The object of analysis of our research is the conversation produced on Twitter in the first electoral debate held in Spain because of the general elections of April 28th, 2019. This debate, held on April 22nd, 2019, occurs 26 years after the first debate, the one that took place in 1993 between the candidates of the PSOE and the PP, Felipe González and José María Aznar respectively, in the general elections of June 6th, 1993. The one that is the object of this work was organized by the public television, Radio Televisión Española (RTVE) and transmitted through the channels: La 1, 24h, RNE, in streaming system by RTVE.es, and through international TVE and Radio Exterior, and other televisions such as IB3, Castilla La Mancha Televisión, Canal Extremadura, Trece, Telemadrid, Canal Sur, Aragón TV, and TPA 2. Its audience was of 8,886,000 viewers, which meant a high screen share of a 43.8%, so it was the most seen that day ((Barlovento Comunicación, 04/23/2019).
Taking into account previous approaches to the subject, we wanted to know the interest that the parties could have in intervening and orienting the conversation about the electoral debate in the space of that second screen, and the communicative-linguistic orientation that that debate had between the prosumers that set the trend in that conversation. Therefore, we raise the following research questions:
Did all the parties present on the electoral debate of April 22nd, 2019 intervened in the Twitter conversation?
Were most of the interventions of these parties attacking the adversary?
Could the political discourse created in the conversation of this social network during the debate be properly described as political or rather as pseudo-political?
Did the expressive formulas of this parallel debate have a great tendency towards mockery and joke?
This work is part of the tradition of pragmatic-discursive studies and finds its referents in cognitive linguistics. The discourse analysis guides our research. In it, we use Gallardo-Paúls’s pragmatic base analysis model (2013, 2014) that this author has applied to the analysis of various types of corpora. This group of classic authors such as Austin (1962), Searle (1969) or Wodak (2001), among others, is based on the cognitive linguistics assumptions of Lakoff (1990; 1980) and Fillmore (1985; 1976) and affects the relevance of the frame at the time of the broadcast, in the text itself or statement, and in the recipient. Here we have analyzed the first moment, that of the enunciation. We must also highlight that we agree with Piñuel and Gaitán (1995, p. 516) when they point out that: “it could be asked if what is understood by Textual Analysis or Speech Analysis is only a modality of Content Analysis or vice versa”.
As Ibarretxe Antuñano (2013) explains, cognitive linguistics understands language as part of the cognitive abilities of the human being, among which are also memory, reasoning, categorization, and attention, and acts combined with them. It is a process that affects the two levels of speech indicated by Benveniste (1958): that of the statement and the enunciation, and from a pragmatic perspective at its three levels: the enunciative, the textual, and the interactive (Gallardo-Paúls, 1996).
As we have already pointed out, in this research we wanted to know the cognitive aspects that act in the transmitter and make it frame its messages on Twitter, so we exclusively cared for the enunciative level, leaving aside the textual or interactive levels.
The analysis sheet consists of the categories shown in table 2. First, we analyze the lexical strategy through the analysis of hashtags used in the tweet (if they were connotative or denotative) and responding to whether we needed to use inference or not to understand the tweet. Secondly, we attended to the thematic strategy of tweets. On the one hand, we wondered if the theme was political or not political and, when it was political, what those tweets were about.
For this, we distributed the topics according to several categories. Some were related to the Policy issues as defined by Patterson (1980, in Mazzoleni, 2014, p. 210), that is, we analyzed whether the tweets dealt with: (1) Economic, fiscal and, employment policy; (2) Social policy, welfare State, pensions, and equality; (3) Territorial policy, and (4) Democratic regeneration and post-electoral agreements. This last point, post-electoral agreements, would be part, according to Paterson, of the Political issues, although we analyze it together with the point of “Democratic regeneration” to follow the thematic lines established in the RTVE debate. To these four categories, we add another one called (5) “Opinions about the elections, the candidates, etc.”, to which we link the tweets related to the elections, the polls, the candidates, etc., that is to say, what is called by Patterson as Campaign issues. In the category of “Non-political issues” to which we have referred to earlier, are included issues that, following Patterson (1980, cited by Mazzoleni, 2010) have more to do with “Personal Issues”, that is, with the life and activity of the candidates. Teasing, ironic comments or rhetorical figures hardly classifiable in any other section were also incorporated here.
In the intentional strategy, we tried to see what was the ilocutivity of speech acts produced in the tweet. For this we used the classic classification of Searle (1969), also separating expressive speech acts into two types: positive or negative. Within this strategy, we also analyze the orientation observed in the tweet, which was broken down into types of attack or types of praise (following the proposal used by Benoit (1999) for another type of corpus of an eminently political nature), differentiated according to the political group to which they were directed. Finally, the predictive strategy answered the question of who was the enunciator of the message. At this point the accounts were classified according to two criteria: (1) those that responded to a nickname (“with nickname”) or that included a random selection of letters, and (2) those that we qualify as “identified” and where those belonging to the parties, their main candidates or their local, regional, etc. clusters, those belonging to the media or journalists, and those of users identified by name and surname were included. (Look at Annex 1).
Table 2. Analysis model with categories up to the first level.
Source: self-made.
The corpus is made up of 1,000 tweets randomly selected by the Atlas.ti qualitative analysis program among the most popular in the April 22nd debate. The selection was made taking as reference the hashtags advertised by Televisión Española for the generation of the network conversation, which are #ElDebateEnRTVE and #ELDEBATEeenRTVE. Unlike in other cases (Anstead and O’Loughlin, 2011; Kalsnes et al., 2014; Shamma et al., 2009) and following the example of research such as those of Trilling (2015) we limit the research to the hashtags proposed by the chain, thus avoiding the “false positives” that could have hindered the study and mired our work.
Version 8 of Atlas.ti allows us to import a maximum of 1,000 tweets, which can be selected according to several criteria; in our case, we established the most popular tweets regarding the type of result and, about the data, retweets and images are included in the final corpus.
In 2011 Anstead and O’Loughlin wondered about the significance of a small sample of tweets when the sample universe susceptible to analysis is so extensive. We agree with them and Chadwick (2011) when they point out that: “we also need to be mindful of the possible number of people reading Twitter content, which would multiply the importance of the content by a great magnitude” (Anstead and O’Loughlin, 2011, p. 447). Likewise, we claim the influence that Twitter has today on journalists, the media, and politics as a documentary and news source, which makes it a reference instrument. And it is precisely the most popular tweets the ones that generate public opinion. We understand that the criterion of popularity is significant enough for this to be a valid and reliable corpus.
The analysis universe of the parallel discourse diffused on Twitter because of this debate was n=107,915 tweets, so our sample represents only 0.93%. To know the analysis universe, we used a script programmed in Python “Get Old Tweets Programmatically” (Jefferson-Henrique, 2016).
The coding of the tweets was carried out by a researcher in Atlas.ti and co-occurrence tables were used to extract the results, which were also processed and analyzed in Microsoft Excel.
3. Results
One of the most visible results of the analysis is that up to 70.6% (n=706) of the sample were retweets, that is, they were not contributions written ex profeso by the prosumer (Toffler, 1980), own reflections, aphorisms or original thoughts, but were messages of foreign authorship and re-diffused by the user from its account, which shows the high “virality” of these contents, in the sense already pointed out by Roncallo-Dow, Córdoba-Hernández, and Durán Camero (2019, p. 129) or del-Fresno-García and Daly (2019, p. 70). In turn, this points to the limited creative proactivity of these users.
The first question we try to answer is whether all the parties present in the televised debate intervened in that other conversation on Twitter, which the analysis confirmed. In a first approximation, Atlas.ti pointed out the existence of up to 44 retweeted interventions from the @populares account, 39 from the @PSOE, 30 from @ahorapodemos, and 29 from the @CiudadanosCs account, thus showing more activity the first-mentioned party. However, their real interventions, that is, not only those from the official accounts were higher. To know the scope of the parties’ intervention in the debate on the second screen, we linked other users close to the political organizations (candidates, regional or local organizations, etc.) with their organization; for example, the interventions of @cayetanaAT (Cayetana Álvarez de Toledo’s account) were incorporated into the PP. To know the predicative structure, we focused our attention on the tweet’s author origin (in the cases of retweet it was analyzed as authorship of the “second speaker”). Thus, we found that parties intervened in the debate on the second screen, but in relatively low percentages regarding the total sample, since none of them exceeded 10%.
The most present party in this debate on networks was Unidas Podemos, which through its account and others associated intervened with 8.9% (n=63); but they all participated to a greater or lesser extent, for example, the accounts linked to the PP, with 5.4% (n=38); to the PSOE, with 4.8%, (n=34), and to Cs with 4.5%, (n=32).
All these accounts linked to parties were analyzed under the category of “Identified accounts”, where, as we have already seen, the accounts of anonymous users who on Twitter were identified by name and surname, and journalists or media accounts (n=80; 11.4%) were also included. Together, most of the interventions were based on these “Identified accounts” (70.4%, n=704), of which the majority came from users not assigned to any party or media. See Graph 1.
Source: self-made.
Graph 1. Tweets according to the type of analyzed accounts.
Secondly, we wanted to know if the parties’ interventions on Twitter were mainly an attack on the adversary, that is if, in their attempt to lead the conversation on Twitter about the debate, the attack on the adversary had been their main goal. We know that studies on political communication like those of Benoit (1999, p. 20) have analyzed political discourse according to the acclamations, attacks or defenses to the candidate or party that it proposes. Taking into account that model, and within the intentional strategy, we tried to know what the orientation of the tweet had been. As we have previously seen, there were three options: the attack on the adversary (specifying which adversary it was talking about), the praise or acclamation (with similar distinction), and the absence of clear orientation. The results of this network debate show that the attack on the adversary was not the preferred option for the accounts linked to the parties on Twitter except in one case: that of Ciudadanos.
The accounts linked to PP, PSOE, and UP spread messages about praise, acclamation, and support to their candidate, with different gradation, because in the accounts linked to PP that support was very significant, up to 85% (n=34), somewhat lower in those linked to PSOE (70.2%, n=33) and did not exceed 50% in the case of those linked to Unidas Podemos (41.9%, n=31).
The Twitter strategy of the accounts linked to Cs was mainly of attack toward PSOE (54.8%, n=17) before the self-praise (35.5%, n=11) although, when observing the percentages, it is seen greater variation and that the results are not so resounding.
Although the attack was not the most used strategy by the parties, it did characterize the general conversation on that second screen of #ElDebateenRTVE, since more than half of the interventions (52.1%, n =578) of the prosumers were of attack, either to parties, journalists or in general, as can be seen in Graph 2.
Source: self-made.
Graph 2. The intentional strategy of the tweet according to its Orientation.
The analysis of ilocutivity points to different nuances as we talk about accounts linked to the Partido Popular, in which interventions of a “compromising” nature (47%; n=18) take precedence, of representative ilocutivity in the case of Unidas Podemos (51%; n=32), and, in the cases of Ciudadanos and PSOE, the preferred ilocutivity is the “Negative expressive” (Cs: 63%, n=20; PSOE: 47%, n=16).
In the case of the PP, this compromising ilocutivity is exemplified in this tweet (Image 1), in which hyperbole is used such as “the greatest fiscal revolution in history”, transmitted through the first person of the singular.
Source: Twitter screenshot made on October 1st, 2019.
Image 1.
The candidate of Cs also preferred to use that verbal person, thus activating strategies for the personification of his speech.
The representative ilocutivity of the accounts of Unidas Podemos can be explained by the numerous occasions in which they use the speech referred to in this other network conversation, as in the example that is seen below (Image 2).
Source: Twitter screenshot made on October 1st, 2019.
Image 2.
Returning to the general corpus, the negative expressivity tweets are those that prevail on the second screen, although a short distance from the representative ilocutivity, fruit, as we have just seen, of the direct quotation of the issuer’s speech on television.
To answer the third question, that is, to know if the type of discourse about the debate on Twitter was far from the canons that have defined political discourse as such (van Dijk, 2003; Charaudeau, 2005), we focused our attention on variables such as the use of the hashtag and inference (which build the lexical strategy), the thematic agenda of the tweets and the prominence of the party (corresponding to the thematic strategy), and the ilocutivity and orientation of those few characters, often completed with links, images, and/or videos (that shaped the intentional strategy).
The use of hashtags was low, confirming something pointed out by previous studies (Boyd, Golder, and Lotan, 2010); they were only used 24.4% of the time (n=244). When these hashtags appear they are fundamentally connotative (18%, n=180), which indicates persuasive intentionality. A large part of these connotative tags responded to campaign hashtags or those chosen by the parties to haul their candidates in the debate, that is #GanaPedro, #VamosAlbert, #ValorSeguro, # LaHistoriaLaEscribesTú. These shared space with others such as #DebateCTXT with which some journalists or media companies wanted to continue a textual conversation making it transmedia. Of course, the use of hashtags not only offers us a framework for interpreting the message but its use implies participation in a community with similar interests (Bentivegna and Marchetti, 2015; Zappavigna, 2011).
The inferences were very present in the corpus, as more than half of the tweets required prior knowledge of the national political context to give a logical sense to the text. Sometimes the graphic help in the form of video or image favors the interpretation of the tweet, but other times it makes it more complex and fuller of nuances.
These tweets loaded with inferences and presuppositions occupy 55.2% of the sample, a percentage somewhat lower than that of other research (Vergeer and Franses, 2016) but significant, which makes us think that this double screen debate dialogues over the original debate, creating a textual framework that only the formal elements (hashtag, threads, etc.) and the context itself allow unraveling.
To know if the thematic agenda present in #ElDebateEnRTVE was related to that of the televised debate or if that other debate adopted another differentiated agenda, we analyzed the tweets of the debate on the second screen according to its relationship with any of the four topics in which the public chain debate was divided into, of which we have already spoken.
The network conversation turned out to be more focused on issues secondary to the political issue and citizen interest policies than on those issues discussed in the televised debate; in fact, the largest number of tweets dealt with “Opinions about elections, candidates, etc.”, (36.5%) and strictly non-political issues (25.2%). (See Graph 3).
Source: self-made.
Graph 3. Thematic agenda of the debate on Twitter in percentage.
Of the four major thematic blocks of the debate, the one that generated the greatest number of interventions on Twitter was that of Economic, fiscal, and employment policy (25.5%), a great distance away from the second topic, Social policy, welfare State, pensions, and equality: 7.1%, of the Political regeneration and post-electoral agreements: 3.3%; or Territorial Policies (2.5%); these last three with almost anecdotal results. We observe how the content of the debate on Twitter did not revolve precisely around the topics discussed in the original debate because, although ours is not a contrastive study, the fact that the percentage of tweets related to the thematic agenda treated in the televised debate is so small, it forces us to ask whether or not Twitter really served as an instrument of deliberation and constructive discussion, to which Trilling (2015) pointed at. In fact, this author points out the difficulty that a topic treated by a candidate in the televised debate be brought to the conversation on the net in the positive or negative way the candidate wishes it to be considered, thus underlining the distance between both conversations and the practical impossibility of any candidate influencing the conversation on this social network (Trilling, 2015, p. 272).
If we try to see the thematic agenda raised by the accounts linked to each of the parties, the results are different. While the accounts related to Pablo Casado’s party tweeted mainly on economic, fiscal, and employment policy (95%), those related to Ciudadanos were distributed and, although they devoted quite a lot of attention to that topic (41%), they also largely wrote about Democratic regeneration and post-electoral agreements (34%) and gave their opinions on elections and candidates (22%). The accounts linked to the party of Pablo Iglesias devoted a lot of space to the issue of “Economic, fiscal, and employment policy” (70%), and somewhat less to the secondary theme of “Opinions on elections, candidates, etc.” (13%) and of Social policy, welfare State, pensions, and equality” (11%). And precisely the latter was the most treated by the accounts linked to the PSOE (44%), above their interest in the “Opinions on elections, candidates, etc.” (31%) and the Economic, fiscal, and employment policy that caused so many interventions in the PP and Unidas Podemos accounts.
A part of the news that the next day summarized the electoral debate of April 22nd in newspapers, radios, and televisions presented it through an image: the memes that had emerged in that conversation (ABC, 23/04/2019; FCINCO, 23/04/2019; G.C., 22/04/2019; El Heraldo, 23/04/2019; Justo, 23/04/2019; LOM, 23/04/2019; Redacción Yo Tele, 23/04/2018; Verne, 23/04/2019). The anecdote contained in the memes, in the tweets with ironic or burlesque content, occupied pages in practically all the media of the country. Surely because, as Cortázar Rodríguez indicates, today: “Internet memes are considered fragments of culture that contain ideas and that influence people as they spread (sometimes exaggerating the immediate influence they might have about people’s attitudes and behaviors)” (2014, p. 204).
Given this, the last of the hypotheses of the work was to verify that, certainly, the social network created by Jeff Dorsey dedicated much of its texts to this type of ironic or burlesque texts, as some previous studies had pointed out (Anstead and O’Loughlin, 2011; Freelon and Karpf, 2015; Trilling, 2015). In fact, in some of these researches the prevalence was very high: “In the case of Twitter reactions to the first and third 2012 Presidential debates, political humor was quite prevalent. Fifty-eight per cent of the most-duplicated tweets across both debates contained humor” (Freelon and Karpf, 2015, p. 402).
However, our analysis indicates that these types of interventions in which mockery, joke, and humor are used in shaping memes were not excessively high. These humorous textual formulas only constitute 16.1% (n=161) of the sample, and this considering an essential aspect in its construction, and that is that it rescued the 1000 most popular interventions, those where, in the opinion of Freelon and Karpf (2015), more often appear. After these results, we turn our attention to the traditional media, which were so interested and talked about the memes created ex profeso during the analysis of the debate on Twitter and we asked ourselves about the criteria they use in generating their agenda, by their criteria of newsworthiness and their role as information gatekeepers.
4. Discussion and conclusions
Two were the electoral debates that the candidates of the main political parties with a presence in the General Courts of the XII Legislature staged: the one broadcasted by RTVE that took place on April 22nd, 2019, and that of Atresmedia a day later. The conversation held on the social network Twitter by parties and users and collected under official tags was the object of our research. The analyzed corpus is thus part of a “filiation environment”, and a “conversation that can be sought” in the sense indicated by Zappavigna (2011), because thanks to the use of tags a geographically dispersed audience participated in a conversation that transcended the immediate and allowed the audience of the debate - and others who did not follow it - to follow a conversation in which the television screen was the reference.
The debate on the second screen that we have analyzed had a high number of retweets, along the lines pointed out by other studies (Bentivegna and Marchetti, 2015; Guerrero-Solé and Mas-Manchón, 2017). However, other works point to different results; this is what happens with the study of Vergeer and Franses (2016), which perhaps we can interpret as a bias of our research that, remember, worked on the most popular tweets, or may be related to the cultural differences between Holland and Spain. In any case, our data confirm the great presence of the practice of retweeting also in the commentary and visualization of an eminently electoral and political program such as the debate. Through this practice, the user expresses, participates, and builds its virtual identity, but not as actively, creatively, and thoughtfully as it would from generating its own textual products.
As we assumed, the political parties present in the RTVE set also wanted to be present and transmit their message in that other parallel virtual conversation on Twitter and participated. They did it through accounts of prominent members, of the party itself, and its regional and/or local organizations. Their participation in this discourse on the second screen allowed them to transfer their message in this social network, in which they implemented classic strategies of political communication such as the attack on the adversary or the acclamation of the candidate himself, to try to get their message to reach all social groups, also to those who were not watching the electoral debate on screen but did follow it through the networks. In this way, they achieved disintermediation in the information flow (Bentivegna and Marchetti, 2015), and were established in authorized voices while confusing with the collective. However, their participation in this conversation did not exceed 10%. It would be interesting to know if this presence maintains these characteristics in future electoral debates.
However, interventions from user accounts linked to PP, PSOE, UP, and Cs were much less than those of citizens without known political and/or journalistic links (64.9%), which coincides with Hosch-Dayican et al. (2014) and López-Meri (2017).
The intentional strategy proposed by the political parties in their discursive framing on Twitter was quite similar in the cases of PP, PSOE, and UP but different in the case of Ciudadanos. As we have already pointed out, the first ones chose to praise their candidate through their linked accounts to different degrees, since the accounts linked to PP and PSOE did ample work of personalization and demand of the leadership of Pablo Casado and Pedro Sánchez, respectively. On the other hand, the priority action of the accounts of Ciudadanos was the attack on the adversary, especially the attack on the government party and its candidate, Pedro Sánchez. It would be interesting at this point to know if the strategy was organized by the campaign teams or a product of the personal initiative of members and political groups, but this requires another type of methodological approach that was not the object of this study. However, while that was the strategy of the parties, the attack was the priority textual option of the network debate as a whole, which coincides with the study by Hosch-Dayican et al. (2014) about citizen behavior on Twitter during the Dutch campaign for the 2012 elections.
The debate on the second screen showed that beyond the hashtag that served to convey the conversation (#ElDebateEnRTVE) the use of tags was not very high (Villar-Hernández, 2019) although when they appear they do so mostly to activate an evaluation framework, and with a clear persuasive goal: to evaluate positively or negatively any of the candidates. The topics of this debate had a thematic distribution that we know did not keep parallels with the debate on the television set, even without having analyzed the televised debate. The reason is clear, and that is that only one of the issues addressed by politicians, Economic Policy, gained significant relevance in the debate on Twitter. As we saw, most of the conversation on the second screen dealt with opinion issues (about elections and candidates) and non-political issues, so the programmatic policies of the parties, the commitments that were being enunciated in the other space, had barely any weight on Twitter. We agree with Trilling that part of the Twitter content was influenced by issues that in the televised debate went unnoticed (Trilling, 2015). On the other hand, this thematic treatment of the debate does not distance itself so much from that of the classic media, according to García Marín (2015), who pointed to technological support as an explanatory variable of that biased behavior, which is an obvious conditioner of the discourse in a social network like Twitter.
The high need to contextualize the 280 characters of the message on Twitter to get a right understanding of the text as a whole was another feature of the conversation, which makes us claim the label of “pseudo-political discourse” in the definition of this concrete communicative situation.
The fact that the tags were mostly of a connotative, together with the high need to contextualize tweets to follow the speech and that the conversation dealt with more secondary issues than those of the debate, allows us to affirm that the debate in the network was pseudo-political, although this reality is not as evident as its analysis in other types of corpus poses (Gallardo-Paúls, 2017). However, some features that point to the existence of personalization (high incidence of expressive acts), de-ideologization (presence of abundant non-political issues, and expressive speech acts), and spectacularization (high prevalence of inference and the presence of irony, mockery, and sarcasm) in the general set of the analyzed corpus.
This need to constantly apply inference for text decoding leads us to think about the ease of a message being interpreted in different ways, according to the reader’s nodal points. This, in turn, facilitates interpretive confusion and de-ideologization, since it empties the text of absolute meaning.
The presence of memes and humor, so prevalent in other latitudes (Anstead and O’Loughlin, 2011; Freelon and Karpf, 2015) was not especially significant in our corpus, although it echoed in most of the traditional media the day after its existence.
The present research starts a line of work poorly developed in the Spanish context: the empirical study of the interaction on Twitter about the electoral debate in the context of a general election to the Presidency of the Spanish Government, in particular the one produced on April 22nd, 2019. It has been described that the four parties present in the television debate participated, although not in a very relevant way, in that one that took place on the second screen, and that their interventions were directed above all to praise the performance and policies of their candidates with one exception, Ciudadanos, a party that opted for the attack on the PSOE and its candidate. With this study, a line of work begins, that pretends to know the communicative strategies of the different parties around the parallel debate created in the social network Twitter.
As for the group of prosumers who participated in this other debate, their interventions were mostly of attack, as studies in countries around us have pointed out (e.g. Holland). On the other hand, much of that speech could be described as pseudo-political. However, memes, humor, and humorism were not an essential and central part of it. A longitudinal continuity of this study would be interesting to see if the results obtained in this work are maintained or variations occur, but with the necessary study of alternative research techniques that allow a more agile coding of a higher sample of the tweet population.
The unstable Spanish political reality, to which today we must add new elections in 2019 (November 10th) has meant that at this time we can add new electoral debates among candidates for the Presidency of the Government, thus consolidating this television phenomenon and its textual correlate in social networks. Twitter has become a tool for political visibility, another platform for distribution of content with its own and differentiated features where a different game is played, but a game to win or play after all.
Bibliographic references
AUTHOR
Paz Villar-Hernández. Paz Villar-Hernández has a Ph.D. from the University of Valencia, a Degree in Journalism (Information Sciences) from the Polytechnic University of Valencia, and a Master in Education and ICT (Universitat Oberta de Catalunya). She currently works as a technician at the Servei de Formació Permanent i Innovació Educativa of the University of Valencia. Among her research interests are political discourse, audiovisual communication, media, scientific publishing, and Higher Education. She is editor of the scientific journal, Research in Education and Learning Innovation Archives (REALIA): https//www.uv.es/realia and is part of the Editorial Committees of the Journal of New Approaches of Educational Research (NAER) and RELIEVE (Electronic Magazine of Research and Educational Evaluation) among others. She has worked at Duke University (United States) and has made stays at Northampton University and K.U. Leuven She has participated in the PRODISNET 01 and PRODISNET 02 research projects.
Paz.Villar@uv.es
H-index: 2
Orcid ID: http://orcid.org/0000-0001-9617-1688
Google Scholar: https://scholar.google.es/citations?user=RGGhsHgAAAAJ&hl=es