Keywords

1 The Italian Political Communication Before the Social Media Era

Politics and media have always established relations of mutual influence, representing two of the major subsystems on which complex societies are articulated (Luhmann 1995). To outline the specific forms taken in Italy by such systems as scenario of our research, first, we need to understand their role and their peculiarities within the more general processes faced by the media and political communication in advanced Western countries over the past decades. A process mainly characterised by an epoch-making passage, generically indicated as the transition from the “solid modernity” to the “liquid postmodernity”, and connected to technological evolution of media, communication paradigms, social and cultural forms. Such a transition, which has also expressed itself through the progressive decline of “big narratives”, and therefore of the main ideological systems, has profoundly influenced the forms and practices of politics: giving rise, among other things, to the secularisation of the sense of political affiliation, the crisis of traditional parties and militancy, and the growing lack of interest in politics, especially with young people. Furthermore, the development of mass media, has deeply influenced the forms and languages of political communication, and of politics itself.

To summarise these processes, Mazzoleni (2004) identifies three main phases (Blumler and Kavanagh 1999). The first one refers to the post-war period, when parties were the most important element of politics. They were considered to be the link between the political beliefs of the citizens and the institutions that, at that moment of renewal and rebirth, certainly had a crucial role in the life of the country. Leaders were characterised by strong ideas and values, and little attention was given to their personal image and forms of communication. People felt a strong bond with their chosen political parties. In that period, citizens’ political participation expressed itself through militancy in the mass political parties, namely the Italian Communist Party (PCI) and Christian Democracy (DC).

The second phase, starting from the 1960s, corresponds to the period of the affirmation of television as the dominant mass media and the transformation of citizens into “citizen-spectators”. During the 1960s, 1970s and 1980s (Anania 2012), the influence of mass media culture generated well-known phenomena, such as the progressive personalisation of politics and the transformation of political parties and candidates into products to be promoted by marketing activities. The selection of candidates was no longer exclusively in relation to their competence and knowledge, but also for their image on television. Even more relevant is the large proven capability of mass media to influence the political agenda through the effects of “agenda building” and “agenda setting”.

Television allowed parties to reach a bigger part of the population. Politicians tried to adapt their traditional forms of communication to the television language, realising in time the weight their words had when broadcasted to the Nation.

During the same time, political communication leaned more towards marketing styles; for this reason, it was entrusted to professionals of that industry.

For Italians, television became the principal means to gather general news and to be kept informed about politics. In this phase, some peculiarities of the conspiratorial relationship between the Italian political and media system started to develop.

In its early days, the Italian broadcasting system was state-owned and controlled by the Italian government, mainly led by DC. With the arrival of the second RAI channel in 1961, followed by Rai 3 in 1979, the division of the public television networks amongst the main mass parties (DC, PCI, PSI) was accomplished, leading to the logic of “one channel-one main party”.

At this point, a series of new elements characterising the political communication was forming. In Italy, like in the rest of Europe, a strong personalisation and mediatisation of politics took place. New TV formats, such as “politainment”, a term describing the combination of politics and entertainment, came to life. Since the early 1980s, the political language started to abound in claims and slogans, adopting simpler communication phrases, which were designed to reach out to the voters and connect with their feelings and emotions.

This period is referred to as the new era of “Political Advertising” (Jamieson 1992; Trent and Friedenberg 1991) and of “Image Campaigns” (Scammell 1998; Thompson 1995). In those years, the concept of political parties was transformed: it was no more a community to which citizens could belong, but only a product that some political consultant (a new profession created during this period) tried to package and sell. The peak of this evolution and the actual realisation of the so-called “Italian anomaly” took place in 1994, when Berlusconi entered successfully the political arena.

In the Western World, he was the unique case of a Prime Minister, a person able to influence indirectly the media contents of the public networks, that was in fact the owner of the only other subject that composed the Italian television duopoly. In this specific case, political and media systems were combined in the person of the leader.

Furthermore, during the 1994 campaign, the airtime that the media dedicated to political messages, programs, debates and spots increased, with a total pre-campaign airtime of 14.052 min,Footnote 1 demonstrating the transformation of the political campaign into a television and media event.

The last phase that Mazzoleni (2004) identified coincides with the 1990s. Certainly, a fundamental characteristic of that period was the multiplication of media and the fact that the audience started to be a plural-subject: it was no more just an audience, but different audiences with different interests and passions. However, looking back at that moment from today’s perspective, it is possible to recognise in those past decades a phase of consolidation and development of the previous characteristics. The television system was changed from broadcasting to narrowcasting, but the communicational model was still one-to-many, and television remained the focal element in the media communication for the Italian scenario.

This slight evolution of the characteristics of that phase, within the period under analysis, suggests the need to theorize a further step: the “2.0 Era”, characterized by a new communicational paradigm shift.

Starting from the first decade of the new millennium, a new model was affirmed within the global societies. It was a brand-new paradigm characterized by a reticular-model, where the principal communication system shifted from one-to-many to one-to-one. With the birth and the affirmation of Social Media, this period corresponds to what Castells (2009) has called the era of “mass self communication”, in which mediated communication became, for the first time in human history, “self-generated” content, “self-directed” emission, and “self-selected” reception.

This new paradigm has had many effects on the forms and languages of political communication. For the first time, politicians can communicate openly with citizens and electors without the mediation of the journalistic system or of their affiliation party.

In Italy, the adoption of social media was slower than in other European countries, as demonstrated by research developed by the auditing society ISTAT (2009). This study indicated that, during 2009, only 47.3 % of the total Italian population was connected to the Internet.Footnote 2

However, in recent years, the use and adoption of Internet and Social Media from Italian families is still growing at a fast rate. A study done in 2012 shows that 41.30 % of the Italian population has a Facebook profile.Footnote 3 Twitter is a small-scale phenomenon, but its use has increased, starting from 10 December 2011, when a new Italian interface was introduced. Today, there are 4 million Twitter users and the number has doubled since last year.

A recent study conducted in 2013 by News-Italia,Footnote 4 a think-tank studying changes in forms of consumption, shows some interesting data: while general television remains the primary source of information, national newspapers have lost more than 13 % of relevance, and radio listeners have decreased from 57 % to 47 %. However, web usage has increased by 21 %, as a tool to scout for information. This growth is cutting across all age groups, with the 30–49 year-old one growing the fastest.

It is important to stress the steady change and improvement of Italians’ usage of digital technology during this fourth period. Personal computers, the Internet, tablets, and smart phones have become a commodity for Italians, even if they are living in a moment of voluntary austerity.Footnote 5 These tools give people a new feeling of power and a new way to express their point of view to the world.

A common observation in this fourth period is the increase in confidence of peer-to-peer relationships, and a tendency to not being passive or uninformed about new products. The new interest about sharing information, feelings, experiences, and reviews of products is steadily growing. In recent years, the peer-to-peer revolution opens the door to new possibilities for consumers and companies. The importance of the increase in social media, and the communication possibilities cannot be overlooked.

So, how do these changes affect politics? Digital media offer new opportunities that partially revive citizens’ interests in politics, in a moment in which there is a deeply rooted disaffection with politics. This situation is confirmed by the increasingly high percentage of abstainers and undecided citizens. It coincides with a brand-new voters’ need for changing the political class and for direct participation of citizens that is well represented by the electoral success of the Five Star Movement (M5S) headed by Beppe Grillo and can be summarized with a tweet that was collected during our research: “Politicians wearing a tie can’t understand people wearing a sweatshirt”Footnote 6 Different variables contribute to the complex relationship between politicians, communication, and the voters; namely the penetration of digital media, the political culture, and the electoral law.

Indeed, the actual electoral law represents another Italian anomaly: it was named “porcellum” by the very Member of Parliament who proposed it, and it creates a proportional system with blocked lists where the secretaries of parties decide candidates to be voted by Italians (Bentivegna 2012). In this way, the candidates’ communication with the secretaries of the Parties has become more important than the direct communication with the voters, who now do not have the possibility to vote directly for the candidates.

On the other hand, the candidate’s chance to obtain more visibility through the media mainstream is often hampered because of the limited resources, which are almost exclusively reserved for the party leaders. Therefore, social media can offer new opportunities to the less prominent candidates and figures of the various parties to obtain more visibility and a more widely known reputation.

The goal of our research was to observe how candidates of different political parties used Twitter for their political communication during the last 2013 Italian elections.

2 The International Research

The first researches on the use of Twitter in political communication date back to 2009. At that time, the penetration of the use of Twitter in Europe and the United States was relatively high, while in Italy the tool was still almost unknown.

These researches, mainly conducted using a quantitative and qualitative approach, were focused on the comprehension of the politicians’ use practices of the new media, and on the communicational functions activated with voters. Those developed by Jackson and Lilleker in Great Britain (2011), and by Lassen and Brown in USA (2010) are interesting examples of this kind of research.

The data collected in 2009 by different scholars on the diffusion of Twitter among politicians are pretty similar. In England, the percentage of the Members of Parliament having a Twitter account was about 12 % of the total, in Denmark was 10 %, in Holland was 7 % (Hansard Society 2009). In the U.S.A., the percentage recorded on data gathered in 2010, was slightly higher oscillating between 25 % and 30 % (Goldbeck et al. 2010; Lassen and Brown 2010).

Both the English and the American researches converge on the detected predictive variables. Chi and Yang (2011) demonstrated that politicians more familiar with digital media were also more likely early adopters of the Twitter platform. In particular, they underlined that politicians having a Facebook (Fb) profile tend to open a Twitter page about 6 month before those who did not have an Fb profile (ibidem).

Worthy of note is the fact that, in the U.S.A., the diffusion of Twitter was higher within the Republican Party, whose members are on average older and, of course, more conservative than those of Democratic Party. On the contrary, results obtained in the English research, more similar to the Italian one, showed a higher percentage of early adopters inside progressive and leftist parties. However, although this seems a contradiction, it actually confirms a cross result of many researches, demonstrating how the adoption process of digital media, particularly of Twitter, is not related to few variables but to a complex series of factors, such as: the dimension of the party, the economic resources and, in the U.S.A., the number of the laws promoted by the party and the political composition of the affiliation commission.

From a more qualitative point of view, scholars have mainly focused their attention on the comprehension of how politicians manage this micro blog, within their more general communicative strategy. Using content analyses techniques, Goldbeck et al. (2010) recognise five different functions in the tweet generated by politicians: 1. Direct Communication (tweets containing @ and direct to specific users); 2. Personal Message (emotional tweets); 3. Information; 4. Questing Action; 5. Activities (tweets containing informations about politicians’ and institutional activities) 6. Fundraising.

Moreover, different researches that have analysed the communicational model related to the use of Twitter, show how the tools can be utilised as: (a) a means of direct communication to citizens (following the traditional model one-to-many); (b) a medium of relation and bidirectional communication; (c) an useful tool to activate listening processes and to collect issues and themes from people (Chi and Yang’s 2010; Jackson and Lilleker 2011).

Researches on Anglophone countries converge in demonstrating that politicians seem to be poorly oriented to use Twitter to generate an effective relationship with voters and activate with citizens a dialogue on political issues.

The use of the medium resulted instead in being oriented by an informational communicational model intended for the re-mediatization of contents of other communicational media, mainly useful for informing citizens about candidates’ whereabouts or participation at political events. This kind of use of Twitter is very close to the broadcasting model typical of mass media communication.

3 Social Tweet in Italian Political Communication

As many researchers and academics claim [see, for instance, Valeriani (2011)], in Italy, Twitter, even if is still a medium-scale phenomenon (Bracciale 2012; Cosenza 2012), is now the principal Social Media platform used to create conversational flow about political, diplomatic, or journalistic issues.

The first major study about Twitter use in Italian political communication was developed by Antenore (2012). The research analyses 17,000 tweets produced by Italian politics from 31 of January until 27 of February. At the moment of the detection, only 9.70 % of the total number of politicians were using the platform. From the recorded data, it can be inferred that some accounts were very active but others were completely abandoned.

From a socio-demographic point of view, the results demonstrated a slightly higher presence of female profiles, with a superior level of education, and a prevalence of freelancers. As described in other countries (Chi and Yang’s 2010), in Italy too the age does not seem strongly correlated to the adoption of Twitter. On the other hand, minor parties with a small visibility, are the major users of the social network. Other interesting data are relative to the “seniority within politics”. The presence of Twitter is associated with the 38 % to subjects that are at their first legislature. The longer the political career, the more reduced the presence on the social networks.

The average value of tweets measured for each politician was equal to 185 while the median was 43. Considering this data, Antenore deduced that a small part of the politicians produced about the 80 % of the content. After careful analysis, she claimed that the major parties were the ones that tweeted more. A fact indicating that the cyberspace would move towards a “normalization” (Margolis and Resnick 2000) of the use of the medium. However, focusing the analyses on the single active subjects, the results are inverted: the minor parties’ representatives are those who tweet more to get a greater visibility than that they can obtain in media mainstream.

To point out the use practises of politicians, a content analysis on a sample of 1400 tweets was conducted, using the categories of Goldbeck et al. (2010) previously quoted. 49 % of tweets were of information type, 13 % were of reporting about participation on the media, 11 % asked for participation of citizens/electors, while 9 % are personal tweets. These tweets are part of the macro-category Activities that Antenore adapted to the Italian contest creating the subcategories of Official Business, Location, and Media.

Considering these data it is possible to say that, during 2011, the “Direct communication” (containing @ and addressed to specific person) appeared very limited, while the “Internal Communication” (tweets addressed to other politicians) is totally absent. Then, 6 % represent political activities for the local government, and 4 % the institutional activities.

Another important item of information is that 85 % of the “Informative” tweets contain links to blogs, to Parliamentary, or party sites. This indicates a use of the social media as a press review and underlines how politicians were not interested in being more transparent with the voters through the use of the medium.

To conclude, it is meaningful to notice that only 5 Members of Parliament from a total of 92 used the retweet function, and only 279 tweets of 1400 contained a hashtag, revealing a basic use of the tool and a difficult appropriation of its specific grammar by politicians. Moreover, the communication activated on the medium appeared to be totally unidirectional and poorly addressed to involve or confront voters.

The growing interest in politics and social media in Italy is demonstrated not only by the increase of academic researches, but also by the birth of many web sites and analytic tools trying to explain the new phenomenon of the first social Italian election. The majority of such tools are user-generated. Some of them are similar to news websites; instead, others are instruments for bloggers, journalists, and researchers who can use them to extract useful data. Some examples of the first type are sites such as Italia2013.me and Bachecapolitica.it.

Italia2013 is a web site of this kind that presents itself with the following quote: “Participate, share, vote” and offers a collection of useful information for the voters: such as informative articles, photos about political events and links to “social events” like a movement on Facebook, or hashtags that become case-history because of their massive use.

The new political Italian environment, characterized by a strong intolerance toward politicians’ false promises, has created a new desire of transparency. To respond to the citizens’ needs, many sites have been developed about politics, “fact checking”, and reliability of politicians’ statements. For instance, in Bachecapolitica.it, every user can contribute to verify the truthfulness of what a politician has said: he/she can post a comment, say like or don’t like on a post, and he/she can insert a politician’s promise. Indeed the site is entirely composed by user generated posts reporting promises made by politicians.

On the other hand, the diffusion of social media marketing has stimulated companies and private subjects to work on these themes to apply them for business and personal promotion. Therefore, in this particular conjuncture, many scientists and analysts start researches aimed to intercept buzz on the Internet and to measure sentiment about the different political lists with the intention to predict electoral results.

For example “Voices from the Blog”Footnote 7 has adapted an equation proposed by Hopkins and King (2010) by which it is possible to analyse the “sentiment” of the web on different relevant arguments. Their work is just at the beginning but, with the mix of the mathematic formula and the team of manual analyses, they obtained data about last Italian and American election, affected by an error lower than 2 % (Ferrazza 2013).

Another important role in this field of analysis is played by BlogMeter, that has developed software for the semantic analysis of the buzz related to political topics in Italian Facebook and Twitter. Specifically for the last election they developed “PolisMeter”,Footnote 8 a web-based tool where Italians can follow the “social” communications produced during the 2013 political campaign.

4 Twitter Uses and Communicational Functions Before and After the 2013 Political Election

We can say that, in Italy, the 2013 political election was the first “Social Media Italian Election”; a phenomenon, even if on a smaller scale, similar to the one won by Obama in 2008 in the USA (Brown 2012). This election represented the starting point of a “Twitter (R) Evolution” in Italian political communication.

However, if it is possible to identify this last election as the first social election, could we say that Italian politicians are really aware of the “social” use of Twitter? Have politicians understood the voters’ need for one-to-one communication in the social media age?

To answer these questions, we analysed the activity on Twitter of a sample of Italian political candidates in two defined ranges of time. The first period was the “campaign period”, from 25 January 2013 until 26 February 2013; the second one was the “post-election” period from 27 February 2013 until 27 April 2013. During this long-lasting phase, Italy remained without a government and without a President of the Republic.

The party lists chosen for the research were: the Democratic Party (PD), the People of Freedom (PDL), Future and Freedom (FLI), Left Ecology Freedom (SEL), Civic Choice-Monti, Act to Stop the Decline (FiD), Five Star Movement (M5S), CasaPound Italy (CPI), Italian Civil Revolution, and The Union of the Center (UDC). These lists represent the general situation and the big parties, but also a more distinctive situation—the new one that emerged during this election. Indeed M5S, FiD, Civic Choice, and CasaPound represent an alternative dimension of Italian politics and a new possibility for the constituency.

While analysing these two dimensions, it is possible to highlight the complexity of the Italian political environment, recognise new rules, and modus operandi that transform these parties in national case histories.

From these lists we studied, 41 specific candidates were selected by means of two different criteria. For each list, there was the leader and three other exponents chosen if they were among the first three most popular inside the regional electoral lists.Footnote 9 The actual sample was, finally, made up of 33 politicians because only the active profiles were considered in the study, and six politicians no longer have a Twitter page.

We analyzed the Twitter profile of each politician selected in two main ways: quantitatively and qualitatively.

For the quantitative part, we observed that the extracted data underlined if the communication was one-to-one or one to many, if the number of tweets per day was reasonable or not, if the grammar of Twitter has been understood or not.

Starting from the quantitative analyses, a strong heterogeneity within the total number of tweets generated by the different candidates was shown. Candidates presented a maximum of 1230 tweets and a minimum of 5 in the first period; and a maximum of 335 and a minimum of zero in the second period. The average for each candidate was about 109 tweets in the first period (with a median of 49) and 30 for the second one (with a median of 30).

Other macroscopic data are the differences between the tweets generated in the campaign and those in the post-election period. Data shows a transition from 3,592 tweets in the first period, to 2,020 tweets in the second one. They decreased by about 44 %. Until now, these data suggest an actual strategic use of the tool. It appears to be mainly finalised to obtain more visibility and a better reputation during the campaign.

In support of this hypothesis, we can consider that some politicians opened their account in the pre-campaign period, while others disappeared from the channel immediately after the election day. Indeed, about nine politicians, including the Senator and ex Prime Minister Mario Monti, stopped tweeting or reduced their Twitter use drastically (for instance Mario Monti tweeted 273 times during the pre-campaign, and only three times in the following period).

A cross result founded in the research was the total absence of shared strategies of Twitter use among members of the same party. Effectively, the variable “party affiliation” did not generate any statistically significant difference in the obtained data. This results is consistent with the possibility offered by Twitter as a “personal medium”, which allowed candidates to bypass the traditional logic of mediation carried out by the parties between them, the media, and citizens.

For the qualitative part of research, all the tweets tweeted by the sampled candidates has been collected and 654 of them, extracted from a total of 1,230, were analysed. For each politician, tweets were selected following this sampling pattern: if the tweets were less than fifty, they were all analysed; if they were more, only one tweet out of three has been analysed.

Tweets were studied using different variables: Jakobson’s functions of language (Jakobson 1963), the presence of the @ symbol, the presence of the # symbol, issues, and slogans.

Jakobson divided the functions of language into six: the referential, the emotive, the conative, the poetic, the phatic and the meta-linguistic one. The phatic function is the one with which it is possible to activate, maintain or terminate a conversation; this function is connected with the emission channel. The conative is addressed to the recipient. It is useful to express what can be defined as a “call to action”. The poetic one is focused on the message: its language is complex, rich, and rhetorical. The meta-linguistic function is completely interested in the construction of the message and in its language features. The referential one is related to the context and is useful when it is necessary to describe actions and behaviours in a specific space and time. The last one is the emotive, which is focused on one’s personality, feelings, and desires.

Through the analysis of Jakobson’s language functions, the net prevalence of tweets with referential function emerges both in the campaign (50 % of the total) and in the post-campaign phase (40 % of the total). This function is mainly generated by tweets that gave information about candidates’ presence on a TV show or at electoral events, or that indicated articles or links connected with their sites or blogs. This practice demonstrates a mainly unidirectional and top-down use of Twitter, finalized to narrate the politicians' agenda and to construct the press review of their activities. The emotive function is the second most present in the messages produced during the two periods. Respectively, they were 30 % and 40 % of the tweets generated. These interesting data leave room for two different ways of interpretation.

First, the considerable weight of emotive tweets can be read as an appropriate use of the medium that users can utilise to communicate with the outside in a non-mediated way, showing their subjectivity and their emotional experiences with openness, and authenticity.

Secondly, it appears as a further testimony to the personalisation process in politics that is also reproduced through Twitter. This consists of a communicational logic of self-centeredness and self-reflection, which penalises against the opportunities that Twitter offers to create an active audience and a direct communication with the voter. The phatic tweets were only generated during the campaign period, and with a very small percentage compared to what we could have expected. Only 7 % of the total tweets were of this type. This percentage, which was reduced to zero in the post-election phase, shows the politicians’ low interest in the actual process of communication that Twitter can achieve, and to its ability to directly connect the profile’s owner with his/her followers.

In both periods, the conative function was very low within the analysed tweets, and was limited to stimulating the participation to events.

In both periods, the lowest Jakobson’s functions embodied in the politicians’ tweets were: the poetic (4.40 %) and the meta-linguistic one (0.02 %).

One of the most important items of information we collected was the reduced presence of political issues: only 40 % of the tweets created in the two phases of analysis connected with issues that had a political relevance, and this applies equally to the various parties. Particularly, the economy was the most quoted topic present in 31 % of the tweets, followed by the themes of justice (2 %), welfare (3 %) and work (2 %). The changes recorded in the percentage of tweets connected with the different themes were minimal in both periods and not significant.

This result, coherent with those resulting from other Italian researchesFootnote 10 is in line with the styles of the political communication on all media that characterizes Italy. Indeed, they confirm how in the “3P System”—“Policy, Political and Personal Issues”—the Policy area is, in Italy, the most under-represented in political discourse where discussions prevail about skirmishes between parties and about the principal and more mediatised politicians.

In Twitter language, the hashtag (#) identifies a theme that converges all the tweets that contain it. It is useful when creating a flux of conversation that allows debates among users about specific issues. The analysis of the hashtags most used by the studied candidates allowed us to rebuild the themes debated during the election. The study confirms how politicians tend to use the medium in accordance with the self-centeredness typical of the Italian political communication. Self-centeredness is merely transported from the traditional media to Twitter. About the total of 1,899 hashtags collected, only 156 (about 8 %) are correlated to policy issues, 4 % referred to television programmes or other media, 6 % to institutional issues, 10 % to places of significance for the election campaign, and about 63 % (equal to 1,189 hashtags) correlated to the political candidates themselves. Among these, the communicational anomaly of Berlusconi emerges. His hashtags (#berlusconi, #consilvio) are present in a large majority of the messages produced by the account activated by the ex-Premier during the 2013 electoral campaign (@berlusconi2013) and they represent about 37 % of the total hashtags used in the two analysed periods. It is important to indicate that the hashtag #noprodi refers to Romano Prodi (PD), which encompassed 60 messages generated in the post-election period. It is connotative of an oppositional conversation against the election as President of the Republic of the ex-Prime Minister Romano Prodi. Such hashtags accompanied and facilitated the failure of his candidature that negatively influenced the subsequent Italian political affairs.

The textual analysis conducted of the most retweeted tweets during the campaign period offers interesting insights about the virtual conversation between candidates and citizens. Among these retweets, the emotive ones are largely prevalent (they account for 47 % of the most retweeted, compared to 35 % of the total examined tweets). This result shows that, even in political communication, the messages’ passionate components are those that easily generate engagement and viral effect. The most retweeted messages are those that, largely expressed in an emotive way, contain critical essays of other politicians (mainly regarding Berlusconi, Grillo, and Bersani). Further evidence of how, even on Twitter, the interest of voter-followers is mainly captured by content relative to clash between leaders and other personal or political issues. To summarize, the logic of the political spectacle, typical of mass media and mainly of television, seemed also be passed through the 140-character social media.

The centrality of the leaders in the relationship’s dynamics between politicians and followers is also confirmed for M5S. The findings show that despite its characteristics of a political movement that could result in more horizontal and equal relations among members, the most retweeted tweets were exclusively those of the leader Beppe Grillo, while those of the other candidates of the same party generated very little interest.

5 The “Mass-Mediatization” of Twitter

The 2013 Italian election marked the arrival of Twitter as a popular communication tool between politicians and citizens. Its use reached a significant level of importance to the point of influencing the general dynamics of political communication.

The research here presented was addressed to investigate the use of Twitter by Italian politicians and candidates and the related communicational functions.

The collected data reconstruct an extremely differentiated and contradictory scenario. On one hand, this environment shows the typical characteristics of every adoption phase of communication technology. During this phase, early adopters, who, by definition, are not yet accustomed to a new medium, tend to reproduce on it the communicational logics and the grammars of previous media.

On the other hand, however, it is possible to identify some features that negatively characterise the Italian political communication by analysing the communicational model and the content of tweets.

It was in that period that candidates became aware of the strategic importance of Twitter as a powerful political tool, as its penetration in Italy increased significantly, particularly within the most advanced segment of society, wherefrom the majority of its users are. Such awareness can be demonstrated by the large number of tweets produced during the pre-election phase and by the fact that many politicians, including Prime Minister Mario Monti, and On. Silvio Berlusconi, opened a Twitter account after the electoral campaign had already started.

Nevertheless, the Italian politicians’ awareness did not match their actual knowledge of the medium and therefore their ability to understand its communicational logics. The majority of the politicians, which this study takes into account, used Twitter in a tactical way, mainly centred on focusing attention on their electoral campaign. Indeed, it is not coincidental that a drastic decrease of tweets by politicians occurred in the weeks after the election, with some accounts even becoming idle. Among the communication functions generated by micro-blogging, namely information, relation, and listening of the followers, the candidates analysed by this study have mostly focused on the first one, thus replicating a communication model typical of mass-media. In fact, by analysing the communication functions (Jakobson 1963) on Twitter, it has emerged that the majority of tweets were of the referential type, aimed to give out information to the public rather than to attempt to create relationships, build a dialogue and stimulate feedback from the users, which is exactly what Twitter can help achieve when used properly.

With regards to content, a very small number of tweets discussed policy issues, whereas tweets on political and personal matters were largely prevalent, similarly to what occurred on mainstream media. Emotional tweets, often containing critical or direct attacks against other candidates (particularly on main leaders like Berlusconi, Bersani, or Grillo), were those that obtained the major viral effect and engagement, resulting as the most retweeted.

A textual analysis of such tweets has demonstrated that politicians reproduced on Tweeter pretty much the same traditional communication grammar used in mass media. Elements that allow us to talk about a mass-mediatization of Twitter on 2013 Italian elections can be found in the nature of many of those tweets, such as the self-referential quality of the outbound communication. Most of the hashtags used by the sample referred to the politicians themselves, and mostly to the party leaders, creating a reference system in which politicians are, at the same time, objects and subjects of the overall generated discourse. This trend reproduced on Twitter, represents a long-time identified characteristic of the Italian political communication (Mazzoleni 2004).

A second element is the high quantity of tweets generated by some of the already most known politicians. These data confirm the process of normalisation of the medium used, as already described by other researches (Margolis and Resnick 2000). Indeed, Tweeter, which could represent a new possibility for smaller party candidates to reach out to citizens and connect with them, seems to be used in a massive way, and with greater efficacy in terms of followership, in particular by already well-known politicians and leaders. In essence, those politicians and parties who are able to mobilise the necessary financial resources to outsource the communication to PR agencies or consultants, according to a typical mass media model.

Finally, an additional element is given by the explicit and strong “orientation towards the mass media” of the content and the release timing of tweets produced by the monitored candidates, orientation, which is recognisable on at least two levels. First, the mainstream media are directly present in a significant number of tweets and hashtags in the form of information concerning, for example, television shows in which the tweeting candidate has participated or will participate. In the analysed tweets, Indeed, the Media System appears to be the true recipient of messages. In fact, it is mainly the mainstream media that candidates (and/or their staff) seem to think when they write their own tweets, not so much to get in touch directly with the citizens, yet rather to attract the attention of the media system and rely on the re-launching effect it is able to ensure. This explains the frequent use of tweets by candidates containing “strong”, therefore “easily expendable” content, using the tool to create attention or generate conflict and release previews and statements in order to make sure to get noticed by the media, thus influencing their agenda.

To conclude, it is possible to say that the 2013 Italian election campaign was the first to see Twitter as an important channel of communication, but italian politicians have used improperly this media; more like a mass media rather than a social media. From here our definition of a “mass-mediatizated” use of Twitter. Probably also as a result of this improper use, Twitter has played a very ancillary and peripheral role in the political communication of 2013 election campaign, at least if compared to classic media and, in particular, to television.