Show Less
Open access

Tweets from the Campaign Trail

Researching Candidates’ Use of Twitter During the European Parliamentary Elections

Series:

Edited By Alex Frame, Arnaud Mercier, Gilles Brachotte and Caja Thimm

Hailed by many as a game-changer in political communication, Twitter has made its way into election campaigns all around the world. The European Parliamentary elections, taking place simultaneously in 28 countries, give us a unique comparative vision of the way the tool is used by candidates in different national contexts. This volume is the fruit of a research project bringing together scholars from 6 countries, specialised in communication science, media studies, linguistics and computer science. It seeks to characterise the way Twitter was used during the 2014 European election campaign, providing insights into communication styles and strategies observed in different languages and outlining methodological solutions for collecting and analysing political tweets in an electoral context.

Show Summary Details
Open access

7. Love Britain? Vote UKIP! The Pragmatics of Electoral Tweets during the European Elections 2014 (Albu, Elena)

Elena Albu, Univ. of Geneva

7. Love Britain? Vote UKIP!

The Pragmatics of Electoral Tweets

during the European Elections 2014

Abstract

This chapter discusses the linguistic strategies the UKIP candidates used in “call for action” tweets sent during the 2014 European Elections. Form and content analysis suggests that the tweets are built on the implicit premise that the UKIP is the solution, and a contrastive context of interpretation that generates a conditional type of reasoning is created.

7.1 Introduction

Launched in 2006, Twitter has rapidly become an important communication platform worldwide. If it was initially meant to send real-time news updates, Twitter is now used for numerous reasons in a variety of contexts. The use of Twitter in politics, in general, and during election times, in particular, is one illustration of the multifarious character this micro-blogging service has.

The 2014 European elections are considered to be “the Twitter elections” due to the extensive use of the on-line platform by the political candidates in all 28 member states. The features and constrains of the platform, on the one hand, and the political orientation and ideology, on the other, have led to the emergence of a particular type of political discourse during these European elections. Although the political activity on Twitter has received a lot of attention, the language the politicians use and the impact it has on the electorate’s cognitive environment are still under-explored areas of study. The aim of this paper is to discuss the specificity of the electoral tweets the United Kingdom Independence Party (UKIP) political candidates sent before the elections. The call for action tweets represent a form of strategic and goal-oriented discourse meant to persuade the electorate to vote for them. Particular attention will be paid to the various cognitive and discursive strategies used and to the role they play during the interpretation process. Shedding light on the mechanisms which direct the hearer towards a certain interpretation and the extent to which the linguistic details influence people’s attitudes about political candidates is essential to understanding the process of decision-making. To this end, a corpus-based and a corpus-driven analysis is suggested. Starting from the vote UKIP collocation, a relevant corpus of data has been←145 | 146→ gathered and further analysis using the AntConc software has been conducted in order to discuss the content of the selected tweets. The corpus-based analysis concentrates on a formal analysis of the vote UKIP collocation while the content investigation aims at identifying what linguistic strategies are used and at discussing whether they are productive and how they function at the cognitive level.

The results indicate that the vote UKIP collocation is a very powerful and direct phrase whose role is to persuade the electorate to vote for the mentioned party. The candidates use it alone as a slogan or in different combinations with various operators in order to draw the voters’ attention and to make their tweet visible. A particular function of the hashtags #VoteUKIP and vote#UKIP is also suggested, i.e. to create a contrastive context of interpretation which generates a conditional type of reasoning. The content analysis brings forward two main types of tweets: informative and argumentative. The analysis of the latter reveals that the argumentative tweets are built around the implicit premise UKIP is the solution and generate a contrastive negative campaigning argumentative orientation which allows the political candidates to foster a negative image of the opponents and a positive image of UKIP at the same time.

For the purposes of our argument, the paper is organised as follows. Section (2) discusses the main features of the tweets as particular samples of language use. The discussion is organised around the particularity of the the tweets in comparison with face-to-face communication and with other forms of digital interactions. To set the frame for the analysis of the political tweets during the 2014 European elections, the reasons for which Twitter has been embraced worldwide as an important communication tool during electoral campaigns are indicated. The next section is dedicated to the linguistics of the tweets. After briefly indicating the specificity of the electoral call for action tweets and indicating the corpus and the method of analysis, a formal analysis of the vote UKIP collocation and a content analysis are suggested. The paper end with conclusions.

7.2 The Political Discourse on Twitter

7.2.1 Twitter as a Communication Tool

Initially created as a way to provide personal status updates and to discover the latest news related to subjects one cares about, Twitter has rapidly become an extremely efficient communication medium worldwide. According to the Help Center, Twitter is “an information network made up of 140-character messages←146 | 147→ called Tweets”90. One of the most popular micro-blogging services, it is used for a variety of reasons in various societal and political contexts.

Twitter is a form of computer-mediated communication that shapes a particular type of discourse. In the attempt to compare digital media and face-to-face communication, on the one hand, and to differentiate between different media, on the other hand, Baym (2010) suggests seven key concepts. They are represented by: interactivity, temporal structure, social cues, storage, replicability, reach and mobility (Baym 2010: 7). Considered the main features of the discourse on social media, the seven concepts characterise all forms of computer-mediated communication, such as SMS, chat, email, blogs and others. Nevertheless, there are variations between the different types of digital media, which grant them unique and individual styles.

Displayed in different degrees, interactivity is said to characterize the mode of communication on the internet. The interactivity on Twitter is represented by different types, such as: social, technical and textual. The social interactivity enables social interaction between groups and individuals, the technical interactivity represents the medium’s capability of letting human users manipulate the machine via its interface while the textual interactivity refers to the creative and interpretive interaction between users and texts (Baym 2010: 7–8).

In terms of the temporal structure of the communication medium, Twitter displays features of both synchronous and asynchronous communication. Among the features of synchronous media the following can be mentioned: the rapid transmission of messages, despite the physical distance between the participants, and the sense of placelesness (Baym 2010: 8), which enables virtual closeness between participants. In contrast, the advantage of also having features specific to asynchronous communication is represented by the fact that the tweeters are allowed to manage their self presentation more strategically.

The social cues refer to the context, the meanings of messages and the identities of the people interacting. The Twitter platform displays no physical context as well as no verbal, non-verbal nor para-verbal elements. The next feature is represented by storage and replicability, as the tweets can be stored on the platform, replicated, retrieved and edited prior to sending. The reach refers to the size of the audience Twitter can attain: it has huge potential, depending on the topic, to reach local and global audiences. Twitter is also characterised by increased mobility, as people can send and receive tweets regardless of their location.←147 | 148→

In terms of main features, the tweets are characterised by brevity constraint and the absence of the verbal component and of any prosodic items. As there are no contextual, visual (no body movement, facial expressions nor eye gaze) and auditory (pitch of the voice) cues available, which are important elements in the interpretation process of face-to-face communication, different techniques have been developed that have a great impact on the linguistics of the tweets, such as: the creative use of punctuation, visual interaction, special orthography, abbreviations, acronyms, to name but a few.

A tweet is limited to no more than 140 characters in length. A direct consequence of this brevity constraint is that the tweets may not necessarily contain well-formed ideas or a very developed context. Additionally, the body of the tweet is accompanied by four operators: @-mentions, retweets (RT), hashtags (#) and hyperlinks (http://). By using the @-mention the user addresses and mentions other users, the retweet allows the users to re-send other tweets while the hashtag marks the topic of the tweet and makes it searchable. As a result of the constant references to other users and topics, the conversation on Twitter is said to be “multiparty, temporarily fluid and highly intertextual” (Zappavigna 2012: 195). The tweet may also contain hyperlinks sending to various newspaper articles and documents supporting and exemplifying the claims made in the body of the tweet. A functional operator model (Einspanner-Pflock, Anastasiadis and Thimm 2016: 45) has been suggested in order to indicate the impact the operators have in creating the interaction on Twitter.

The tweets represent particular samples of language use pointing towards novel techniques in the process of meaning derivation and interpretation. The length limitation imposes important constrains in creating the tweet. As a consequence, the tweets can be more or less explicit and sometimes the tweeter is expecting his audience to undergo some inferential work in order to reach the intended interpretation. Nevertheless, the background and contextual assumptions the participants share play an important role in guiding the interpretation process.

7.2.2 The Use of Micro-blogging in Political Communication

The Twitter platform was extensively used in the political campaign of Barack Obama in 2008. The main function of the tweets was to send campaign updates and provide the people with the opportunity to volunteer and get involved in the campaign (Abroms and Lefebvre 2009). Ever since Twitter has been embraced by politicians worldwide as it offers real-time access to the public sphere allowing them to create a direct and fruitful relationship and on-line interaction both with←148 | 149→ the citizens and the journalists in a networked environment (Bruns and Burgess 2011, Graham et al. 2013).

The extensive use of Twitter as the channel of choice for diplomacy and communication between world leaders, governments, foreign ministries and diplomats has led to the emergence of digital diplomacy. Twiplomacy is the leading global study of world leaders on Twitter, conducted by leading global public relations and communications firm Burson-Marsteller91. Federica Mogherini, the Vice President of the European Commission, considers Twitter to be “a revolutionary social network even in politics. […] but not all Twitter exchanges are diplomatic, real world differences are playing out on Twitter and sometimes end up in hashtag wars between embassies and foreign ministries”92.

Becoming aware of the huge potential Twitter has, many politicians have used it as an important communication tool during parliamentary, presidential, congressional, federal and local elections in numerous countries. Politicians choose to use Twitter, in general, and during election time, in particular, for several reasons. First, it grants them the opportunity to address a wider and collective audience. They seek to interact with other political members but also with media and journalists and especially with the large mass of voters. Furthermore, a tweet offers the possibility of constant switching between multiple roles and fragmented audiences (Broersma and Graham 2012). This does not only raise the level of participation in public debate (Bruns and Burgess 2011, Graham et al. 2013, Vergeer et al. 2013) but also raises people’s awareness and encourages them to vote.

Second, Twitter allows the politicians to personally get involved in the political campaign. It grants them the opportunity to take individual initiatives and to shape the conversation about elections, to set the news agendas and send their messages to the public directly, being “un-touched by journalists” (Broersma and Graham 2012). The tweets are said to represent important news source, and they are sometimes quoted entirely. A well-chosen utterance can reach the electorate without being mediated by journalists and, therefore, generate a lot of publicity (Thorsen 2014).

7.2.3 Literature Overview

Various investigations were dedicated to how Twitter was used and what purposes it served. For instance, Golbeck, Grimes and Rogers (2010) focused on the Twitter use during the US Congress elections. The analysis revealed that the tweets←149 | 150→ were used rather like “vehicles for self-promotion” (Golbeck, Grimes, and Rogers 2010: 1620), as the members of the Congress tweeted primarily to disseminate information and to report on their daily activities. Conway et al. (2012) looked at the number of tweets posted, the followers gained and the followers added, the occurrence of hashtags, user mentions, hyperlinks and content categories within tweets during the 2012 US presidential campaign. It resulted that the candidates used Twitter to broadcast information about issues they wanted people to know and to self-promote as opposed to transparency.

Although many studies concentrate on the American elections, there are also studies bringing forward how Twitter was used elsewhere. For instance, Gaffney (2010) studied Twitter use during the 2009 Iran elections by tracking the use of the Iran Election hashtag. Tumasjan et al. (2010) studied tweets related to the 2009 German federal election in terms of their content as a reflection of the offline political landscape. Larsson (2012) discussed the Swedish elections in 2010 by presenting a rationale for data collection and analysis of Twitter users with an emphasis on the practice of civic microblogging.

In sum, the accent is placed on how the Twitter networking is created, on the number of tweets posted, the followers gained, the various occurrences of hashtags, user mentions and different content categories within tweets. Nevertheless, little has been said about the content and the language of the tweets, how politicians use Twitter and how efficient their tweets are in communicating their message and persuading the citizens to vote for a particular candidate or party.

2014 European Elections on Twitter

Electoral campaigns are complex and organized political events meant to influence the decision-making process. The campaign teams resort to different channels and techniques to make the message of the campaign visible and to communicate with the electorate. Successfully used in the political campaign of Barack Obama in the USA in 2008, social media, in general, and Twitter, in particular, have become key communication tools during election times, being embraced by politicians across Western democracies. If the main function of the tweets in Obama’s campaign was to send updates and provide the public with the opportunity to volunteer and get involved in the campaign (Abroms and Lefebvre 2009), the 2014 European elections proved to be a turning point which led to the confirmation of what has been called the “institutionalization” of Twitter (Smyrnaios 2014). Many institutions like the European Parliament, the Commission or certain European parties have used Twitter as their main communicational tool during the campaigns by creating official hashtags and having special companies provide analytics and←150 | 151→ promote Twitter accounts and hashtags on other media (Smyrnaios 2014). If the 2009 elections were considered to be the Facebook elections93, the extensive use of Twitter during the 2014 elections by public institutions, political parties and candidates, media and journalists and citizens led to the conclusion that the 2014 elections were “shaping up to be the Twitter elections”.

There are important differences that lead to variations on the linguistics of Twitter as a result of the different contexts in which it is used, the purposes it serves and the audiences it addresses. The context shapes the discourse type and its features and plays an important role in the process of meaning derivation and interpretation. In comparison with other situations where the context in not defined by a particular event, the participants do not share many contextual assumptions and the tweeter has little control on the type and size of the audience, 2014 European elections set a particular social and discourse context. Even if some information may be left implicit due to the brevity constraint, it may be easily recovered due to the contextual assumptions available represented by the approaching elections and existing political landscape. The political candidates engage with large audiences and target mainly their supporters but they are also trying to make other citizens adhere to their ideology and set of beliefs.

The use of Twitter during the 2014 European elections has led to the emergence of a particular political discourse, shaped in accordance with the characteristics and constraints imposed by the online platform. The specificity of the electoral tweets is also influenced by the ideology and beliefs of the political party the candidates are part of. They play an important role in choosing different linguistic strategies meant to emphasize their political views, on the one hand, and to make the electorate take action, on the other hand.

7.3 The Linguistics of the Call for Action Tweets

7.3.1 General Description

Tweets can be classified in relation to the Election Day in tweets sent before and after the elections. In the first category the following types of tweets are included: (a.) informative tweets used in order to broadcast information about the politicians’ daily activities and their media appearance, (b.) messages as position taking, (c.) direct replies to other tweets, (d.) call for action tweets, i.e. tweets usually inviting people to vote, (e.) tweets promoting negative campaigning and←151 | 152→ (f.) positive campaigning. In the second category the following types can be mentioned: (g.) thanking their supporters, (h.) posting the official results, (i.) making reference to the efforts of the campaign, (j.) acknowledging their political role.

In this paper particular attention is paid to the call for action tweets, i.e. to the way in which the UKIP political candidates reach out to the audience and try to persuade them in order to get their vote. The language politicians use plays an important role in the processes of persuasion and decision-making. Depending on what linguistic strategies are used, the tweets can have different cognitive impacts and, therefore, be more or less efficient in convincing the electorate to vote for them.

The common elements of any political campaign, regardless of the election type, are usually represented by information and news sharing, information about the campaign, location and activity sharing, direct interaction with citizens, personal messages and messages targeting campaigns and mobilization support (Ikiz et al. 2014). Political campaigns are complex events with important stakes consisting of more phases relative to particular key moments. This leads to a dynamics of the tweets content, as the political candidates resort to different linguistic strategies that serve their communicative purposes in different moments during the campaign. If at the beginning of the campaign the politicians are often engaged in conversations with the electorate, answering their questions and addressing them directly, towards the end of the campaign and around the election day the discourse becomes asymmetrical, a one-way transfer of information, from the politicians to the audience. The call for action tweets under investigation in this paper characterize the end of the political campaign.

7.3.2 Corpus and Method of Analysis

The corpus of data is represented by the tweets sent by the UKIP political candidates within the time span of two weeks prior to the 2014 European elections. The analysis is partly corpus-based as the word search represented the analytical starting point and partly corpus-driven as the findings were used for further analysis, i.e. to establish more categories of tweets sent by politicians in order to reach out to the audience and persuade them to vote for them.

The analysis consists of more steps. First, the tweets have been selected by word search: vote UKIP, including the combination of the collocation with hashtags and @ mentions. Second, the tweets have been coded by hand and only the relevant tweets have been selected, i.e. those concerning the elections and expressing the politicians’ intentions towards the approaching vote. No retweets have been included in the analysis as the interest lies in the candidates’ way of approaching the←152 | 153→ electorate in their attempt to create a connection with them. The software used for concordancing and text analysis is AntConc (Anthony 2014).

As a result of the corpus-based analysis, a formal analysis of the vote Ukip collocation is suggested while as a result of the corpus-driven analysis a content investigation is put forward.

7.3.3 Vote UKIP

In terms of strategies, the vote UKIP structure, with the variation vote for UKIP or voting UKIP, is by far the most frequently used. The message it conveys is simple, clear, direct and does not require any inferential work in order to be understood by the audience. It is a memorable structure which quickly draws the voters’ attention. These features have an immediate effect on the voters’ cognitive environment, making this linguistic strategy a very efficient one.

As expected, most of the vote UKIP tweets were sent closer to the election day. The collocation appears 107 times in the investigated corpus of data under all its forms: used alone, in combination with the operators (32 instances of vote #ukip and 4 instances of vote @ukip), in upper or in lower case.

The orthography of this collocation can vary from one tweet to another. An interplay between upper and lower case has been noticed: either the verb to vote or the name of the party is emphasized. There are also instances when the first letter of the verb vote is written in upper case followed by the name of the party in capital letters (Vote UKIP), even if it is inserted in the body of the tweet:

(1) Alan Stevens: Remember to Vote UKIP tomorrow. Polls are open from 7am until 10pm. http://t.co/EBGWJddXxA

This message was tweeted under the exact form by more politicians from different regions. The hyperlink at the end of the tweet sends the reader to the UKIP’s website where the photo of Nigel Farage together with some other party members is displayed. The picture is accompanied by the following message: Vote UKIP tomorrow. Polls are open from 7am until 10pm. The insertion of this message in the body of the tweets explains the upper case used for the verb vote even if it is not signalled accordingly.

There are also instances when only the first letter of the name of the party is in upper case (vote Ukip) or the entire name of the party is written in lower case as part of the informal style that characterizes sometimes Twitter (vote ukip). As there are no prosodic elements available, the orthographic interplay creates a visual impact important in drawing people’s attention and making the message memorable.←153 | 154→

Vote UKIP can be followed by different adverbs with an informative purpose like today, tonight or tomorrow, depending on the time frame of the tweeter and the date of the elections. Sometimes there are semantic alternatives used for the Vote UKIP phrase, such as: look for the ukip logo, look out for our famous pound logo, Look for the “£” logo!:

(2) Jill Seymour: Everyone remember to go out and vote @UKIP Look out for our famous pound logo

The vote UKIP structure can occupy different places in the tweet which leads to different cognitive and emotional impacts. It is not used on the first position at the beginning of the tweet very frequently but rather at the end. It is usually part of the body of the tweet when the tweet is informative as shown in (1). If the tweet is argumentative it can follow an argument offered by the politician, being directly related to its content or it is simply added at the end of the tweet in order to categorize it and to draw attention to it. In the latter situation, the Vote UKIP collocation functions like a slogan having a greater emotional impact on people’s cognitive environment, as shown in (3) and (4):

(3) Bill Etheridge: Stand up to the mindless thugs and chants of the hard left. Stand up for freedom. Vote ukip!

(4) Michael Heaver; UKIP; East of England; 4; Make history. Vote UKIP.

The vote UKIP structure is used in different semantic combinations with the same role of recalling the vote and encouraging the electorate to give them their vote. Different semantic categories have been identified, ranging from telling the citizens to urging them to vote for their party in particular: remember to vote, don’t forget to vote, please vote, please remember to vote. These tweets have mainly an informative role, being used to broadcast general information about the vote, position on the ballot paper, the location and time of the vote, etc.:

(5) Jonathan Arnott;UKIP;UKIP;North East;1;UKIP supporters are the most certain to vote, so I’m confident that a bit of rain won’t hurt our chances. Don’t forget to vote #UKIP today!

(6) Paul Nuttall;UKIP; North West;1; Don’t forget to vote @UKIP today for the European and Local council elections. Every vote matters. Let’s cause a political earthquake.

(7) Louise Bours;UKIP;UKIP;North West;2; Today is the day where we can cause this political earthquake! Thanks for your continued support. Please vote #UKIP!

(8) Roger Helmer;UKIP;UKIP;East Midlands;1; Tomorrow: Please remember to vote #UKIP. Bottom of the ballot paper. Top of the poll!

While many tweets are directive speech acts, meant to recall the vote and invite the citizens to vote for them, the vote UKIP structure can also be part of a rhetorical←154 | 155→ question. In this case the accent is not placed on the interrogative part, as the politician is not seeking for information, but his tweet functions like a confirmation, being followed by details regarding the ballot paper and the place the party occupies on it. By using the imperatives make sure and complain he draws the people’s attention regarding the good functioning of the election:

(9) Paul Oakley; UKIP; London;2; Voting UKIP tonight? Make sure your ballot paper’s unfolded. We’re at the bottom. Complain to staff if you can’t see us.

There are also tweets where a mix of strategies can be noticed. In the following example the candidate used a combination of do not forget, followed by the location on the ballot paper and the semantic alternative for the name of the party Look for the “£” logo:

(10) Louise Bours;UKIP;UKIP;North West;2; Do NOT forget that #UKIP is at the bottom of the ballot paper. Vote very carefully! Look for the “£” logo! We can do this #Chester!

The vote UKIP structure can be inserted in the body of the tweet or can be used in combination with different operators, such as hashtags or @ mentions. Different combinations between the vote UKIP structure and the operators have been identified in the corpus of data. For instance, the vote UKIP can be used as a hashtag (#VoteUKIP) or the operators, the hashtag or the @ mention, can be used before the name of the party (vote #UKIP, vote @UKIP, vote for #UKIP, voting #UKIP94). The latter situation is the predominant one in our corpus of data, as shown in the following examples:

(11) Jonathan Arnott; UKIP; North East; 1; UKIP supporters are the most certain to vote, so I’m confident that a bit of rain won’t hurt our chances. Don’t forget to vote #UKIP today!

(12) Jill Seymour; UKIP; WestMidlands; 1; Everyone remember to go out and vote @UKIP Look out for our famous pound logo.

The analysis revealed that there is a correlation between the place of the hashtag in the tweet and the function it has. The hashtag can be integrated in the body of the tweet and be used on middle or final position as in (13) or can be unrelated to the body of the tweet, being used at the end, as in (14):

(13) Donna Edmunds; UKIP; UKIP; South East; 5; @State_Control maybe by ‘our country’ he means ‘Europe’. Anyway, he can’t hear their voices either. If he could, he’d #voteUKIP @Ed_Miliband←155 | 156→

(14) Gawain Towler; UKIP; UKIP; South West; 3; for years the legacy parties and media have demonised a the ‘feckless young’ to justify mass immigration. It’s a disgrace #VoteUKIP

While in (13), the hashtag is part of the conditional clause replacing the predicate and the direct object, in (14) it is added at the end of the tweet, without any apparent connection to the content. In both cases, the hashtags are used to make the tweet visible, to organise the election conversation, to draw attention to it and make it user-friendly.

Apart from their main functions, the #VoteUKIP and vote #UKIP hashtags serve other purposes, contributing differently to the process of meaning derivation and interpretation. In (14), the hashtag goes beyond search functionality: without the hashtag, the content would have been incomplete and fragmentary and the reader would not have been able to reach the intended interpretation. With respect to it, Scott considers the hashtags to be “highlighting devices”, as they guide the hearer towards the intended overall interpretation (Scott 2015: 14). In other words, the hashtags create a particular context of interpretation by making certain background and contextual assumptions highly accessible to the readers. For instance, the #VoteUKIP and vote #UKIP hashtags activate the background assumptions which situate the tweet in the context of the approaching elections and function like reminders of the party running for the European elections.

The role the hashtag plays is even more complex. Without the hashtag, the tweet seems to express the politician’s attitude towards a description of a long lasting situation. The political candidate criticizes and expresses his disagreement to the way in which the mass immigration problem has been dealt with. By using the hashtag, he suggests that the party he is representing has different views and beliefs and is, therefore, the solution to the mentioned problem.

Scott (2015) identifies more categories of hashtags that serve different functions in the discourse and play different roles in the process of meaning derivation and interpretation. In relevance-theoretic terms (Sperber and Wilson 1995), the hashtags can contribute to the derivation of the proposition expressed by the tweet, i.e. explicature and higher-order explicature, or to the implicitly communicated meaning, i.e. implicated premises and implicated conclusions.

By activating certain contextual assumptions the hashtag creates a contrastive context of interpretation which facilitates the following conditional type of reasoning: if you vote UKIP, the mass immigration problem will be solved. In other words, the intended interpretation is arrived at inferentially, i.e. the reader has to take an inferential route in order to derive the implicated premise: UKIP is the solution to the mentioned problem. In sum, the hashtag is used to create a←156 | 157→ contrast between the negative image of the political opponents and the solution represented by the mentioned party. By using the hashtag, the politician obeys the brevity constraint and manages to send a powerful message meant to discredit his opponents at the same time.

There are cases when more operators are used, such as: the hashtag #VoteUKIP, the mention @UKIP and the hashtag #UKIP, in the attempt to make their tweets visible and easy to find:

(15) Jason Smith; UKIP; UKIP; Yorkshire and the Humber; 5; Immigrants do lots of jobs in public sector because they’re cheap labour – less jobs for Brits and drive down wages #VoteUKIP @UKIP #UKIP

(16) Jason Smith;UKIP;UKIP;Yorkshire and the Humber;5; Tories support Turkey joining the EU!! Tories are EU puppets #VoteUKIP @UKIP #UKIP

At first sight, this overuse of operators may seem redundant, all of them serving the same purpose: making the tweet searchable and highlighting a trending topic. In line with the suggested analysis, the #VoteUKIP hashtag is more complex, playing an important role in guiding the reader’s inferential process. The #VoteUKIP creates a contrastive context of interpretation, in the attempt to discredit the political opponents and promote their own image. Furthermore, voting their party is equalled with the solution of the mentioned problems as they have the exactly opposite values and beliefs.

On balance, depending on their position and whether they are part of the body of the tweet, the hashtags fulfil distinct roles, leading to a reinterpretation of the hashtags in terms of the roles and functions they perform. Apart from the search functionality, as a consequence of the brevity constraint, the hashtags are used to create a particular context of interpretation, i.e. they make certain contextual and background assumptions available to the audience. The #VoteUKIP hashtags contribute to the derivation of the implicitly communicated meaning and generate a conditional type of reasoning in a contrastive context of interpretation: if you vote UKIP, the problem, whatever that is, will be solved.

7.3.4 Content analysis

In this section particular attention is paid to the content of the call for action tweets. Although all the tweets under investigation have the same topic, i.e. the approaching elections, and are part of the same category, i.e. call for action tweets, further analysis is necessary in order to discuss the functions they perform within this category and to identify what linguistic strategies are used by the political candidates in order to reach their communicative goals. Furthermore, particular attention will be paid to whether the linguistic strategies are productive, on the←157 | 158→ one hand, and to the way in which they function at the cognitive level, on the other. A content analysis of the call for action tweets is important to see how the speaker’s meaning is created and how the intended interpretation is reached by the hearer.

Two main types of the call for action tweets predominate in the corpus of data: informative tweets and argumentative tweets. The first category is meant to offer general information about the vote, such as the date and the location of the vote, as well as the position UKIP has on the ballot paper. It is a limited category, the tweets being used mostly to recall the vote and to invite the electorate to vote for their party, as indicated in the previous section. The political candidates adopt different styles, ranging from impersonal messages, as shown in (17) and (18) to personal tweets in which they kindly ask the electorate to vote for their party, as shown in (19) and (20). The message is usually direct and the interpretation is easily reached, as no information seems to be left implicit:

(17) Bill Etheridge;UKIP;UKIP;West Midlands;3; Voting for the euro and local elections is on Thursday may 22nd. Please look for the ukip logo on your ballot… http://t.co/WMNcuKfaUs

(18) Alan Stevens;UKIP;UKIP;South East;8;Remember to Vote UKIP tomorrow. Polls are open from 7am until 10pm. http://t.co/EBGWJddXxA

(19) Margot Parker;UKIP;UKIP;East Midlands;2; #UKIP Team East Mids MEP candidates in Newark campaigning today. Please Vote#UKIP May 22nd

(20) Andrew McNeilis;UKIP;UKIP;London;6; Please vote on Thursday. And I would be honoured if you voted for us http://t.co/8gyItp9Q7L

While in the first three examples, the audience is addressed collectively, the last tweet brings forward the politician’s voice. Apart from the interjection please, used to politely ask the electorate for their vote, the first-person personal pronoun I indicates that the speaker is committed and assumes what he is saying. Furthermore, the candidate grants more expressive force to his message by choosing to express his attitude towards voting for their party (I would be honoured), referred to by the inclusive first-person plural pronoun us.

The second category is represented by the argumentative tweets. In comparison with the informative tweets, this category is broader, the accent being placed on answering the question why the electorate should vote UKIP. The analysis of the tweets indicates that the answers provided are built around general arguments and anti-EU arguments, which shape a negative campaigning argumentative orientation to the politicians’ discourse.

Considered to be an important part of a politician’s campaign arsenal, negative campaigning focuses on fostering a negative image of the opponents with reference to their personality, record and opinions. Negativity is considered to←158 | 159→ be valuable to voters, as well. Mattes and Redlawsk (2014) suggest a theoretical model and carry out experiments which show that voters need more than the arguments provided by the candidates for why they should be elected. It appears that negativity assures more attention from voters, on the one hand, and the voters are more likely to incorporate the negative information into their evaluation, on the other (Mattes and Redlawsk 2014: 4–5).

The politicians can choose to attack their opponents directly, without mentioning any of their their views and actions, or they can compare and contrast themselves with their opponents. In the latter situation, they usually provide positive information about their own record while the information about the opponents is negative. The call for action tweets under investigation in this paper fall into the second category, i.e. the contrastive negative campaigning argumentative orientation characterizes the discourse of the UKIP political candidates. They choose to always juxtapose themselves with their opponents as it allows them to create a negative image of the opponents and a positive image of UKIP at the same time. In what follows, particular attention is paid to the way in which the contrastive negative campaigning argumentative orientation is realised and to the linguistic strategies used by politicians in contrasting and comparing themselves with their opponents. To this end, the following aspects will be discussed: how the question why UKIP should be voted for is addressed, how the premise (UKIP is the solution to the problem x) is expressed, how the negativity is expressed and what type of reasoning is activated. All the tweets under investigation are meant to answer the question why the electorate should vote UKIP and they are all based on the same premise UKIP is the solution. However, what is different about them is the way in which the question is addressed and what linguistic strategies are used. They can range from being explicitly expressed to being left implicit for the reader to infer, situation in which more cognitive effort is required.

Based on the combination between how the question is addressed and how the premise is expressed, different categories of tweets have been identified: (a.) the question is expressed directly and the premise is identified with a minimum of cognitive effort; (b.) the question is indirectly addressed and identifying the premise does not require a great deal of cognitive effort; (c.) the question is in the background and the premise is left implicit and therefore more cognitive effort is required in order to reach the intended interpretation.

(a.) The first category of tweets is represented by the situation when the question is directly addressed and the premise follows naturally, as in the following examples:

(21) Michael Heaver;UKIP;UKIP;East of England;4; Why the City should vote UKIP in the EU elections http://t.co/Vf13XsLgqE←159 | 160→

(22) Donna Edmunds;UKIP;UKIP;South East;5; “Ukip hold dear the true and honest views of the Great British people”. Why are you voting UKIP on May 22nd? http://t.co/cC5p7WT4ho

(23) Paul Nuttall; UKIP; UKIP; North West; 1; EU migrant figures up again with 7 % increase in foreign workers. Another reason to vote @UKIP today. http://t.co/bYoTruNJmw

In all three cases, the arguments are given in the newspaper articles the hyperlinks at the end of the tweet are pointing to. The body of the tweet can be interpreted to be an introduction to the arguments provided at length in the newspaper articles and are marked at the linguistic level accordingly. The second tweet is represented by the actual background question: Why are you voting UKIP on May 22nd? while the first one is a report of the same question. The difference between the two, apart from the direct vs. reported style, is to be found at the way in which the electorate is addressed: the City vs. you, respectively. The second person pronoun you adds more expressive force to the message, by addressing the electorate directly in comparison with the general, impersonal the city. If the first two tweets offered general arguments, the last one focuses on a particular one, represented by the mass immigration, explicitly introduced by another reason to vote @UKIP today. Although at first sight the tweets seems to address the question without expressing any negativity, at a closer look it can be noticed that the negativity is found in the newspaper articles the hyperlinks send to and not in the body of the tweets.

Another example of a direct way of presenting the reasons why the electorate should vote UKIP is represented by the tweets sent by Jonathan Arnott. Unlike his colleagues, he compiled a list of reasons he labelled 7 reasons to vote UKIP on Thursday and presented them the week before the elections. Similarly to the previous category, the negativity is expressed contrastively in the hyperlinks:

(24) Jonathan Arnott;UKIP;UKIP;North East;1; In the week up to the election, I’m doing ‘7 reasons to vote UKIP on Thursday’. To kick things off, reason #7 here: http://t.co/WpTd6RoKIt

(25) Jonathan Arnott;UKIP;UKIP;North East;1; The fourth in my week-long series of reasons to vote UKIP on Thursday. Reason #4: EU money is OUR money http://t.co/MYimQDdah1

The structure and the argumentative pattern of this politician’s tweets are similar: an introductory remark about his series of reasons, the argument itself, followed usually by a hyperlink sending to a newspaper article, which provides more information to support his opinion. In spite of the length of the tweets, the arguments are concise and presented more as campaign messages. This makes them memorable, directly influencing the citizens’ evaluations and the process of decision-making.←160 | 161→

The hashtag #WhyIvotedUKIP was also created in order to address the background question and indicate why anyone should vote UKIP. Its structure consists of the wh- word why, the first-person pronoun I, the past tense of the verb to vote and the name of the party. It is based on the interplay between lower and upper case, meant to create a strong visual impact, which makes it easy to remember.

In terms of content, it is inherently argumentative. The wh- word why forms a correlative pair with because, as part of the questions – answer pattern, the tweeter being expected to give arguments to support his statement. Mainly used by Barry Cooper, the hashtag is placed on initial position in order to categorize the tweet and make it visible. It is usually followed by I want and I think, because being mostly left implicit:

(26) Barry Cooper;UKIP;UKIP;South East;10;#WhyIvotedUKIP I think it is ridiculous that we spend more on foreign aid than we do on policing our streets and keeping communities safe

(27) Barry Cooper;UKIP;UKIP;South East;10;#WhyIvotedUKIP I want politics to be more about politicians asking people what they want and not telling them want they need

(28) Barry Cooper;UKIP;UKIP;South East;10;#WhyIvotedUKIP I want our armed forces properly supported, our veterans cared for and the military covenant taken seriously

By using the first-person singular pronoun I both in the hashtag and in the body of the tweet, the candidate expresses high commitment to what he is stating and personal involvement in the issues he is describing. It also helps the politician to distance himself from the other political candidates, fostering a positive image of himself, as being someone who has principles, takes responsibility and who is not afraid to take action when necessary.

Apart from the first-person singular pronoun I to mark his own voice, the inclusive plural form we/us is also used in the attempt to present himself as being part of the same group as the electorate and having shared values. By establishing an us vs. them dichotomy the speaker creates an image of the group he belongs to in a positive way and the other group in a negative way. At the same time, he is making his opponents responsible for all the issues he is mentioning.

A pretty similar hashtag is also used: #WhyImVotingUKIP. Apart from the aspectual difference, i.e. perfective (voted) vs. imperfective (am voting), #WhyImVotingUKIP is used mostly on final position and the tweet usually begins with because. In comparison with the subjective stance adopted by the tweeter in the case of the previous hashtag, there is a combination between a subjective approach, marked by the use of the first-person I, as in (29), and an impersonal and general tone, depicting a particular situation, as in (30):←161 | 162→

(29) Donna Edmunds;UKIP;UKIP;South East;5; Because I want my daughter to grow up in a free country #WhyImVotingUkip

(30) Donna Edmunds;UKIP;UKIP;South East;5; Because democracy only works when our representatives are directly accountable to us. End the democratic deficit. #WhyImVotingUkip

(b.) The second category of tweets is represented by instances where the question why the electorate should vote UKIP and the premise UKIP is the solution are addressed indirectly but they are easily inferred by the reader and therefore little cognitive effort is required in reaching the intended interpretation. In contrast, UKIP’s beliefs and position towards particular problems are directly expressed (Only UKIP is unequivocally opposed to them and that’s what UKIP voters believe in and vote for), as in the following examples:

(31) Roger Helmer;UKIP;UKIP;East Midlands;1; The Telegraph reports there are now 30,000 on-shore wind turbines in the UK. Only UKIP is unequivocally opposed to them. Vote #UKIP.

(32) Michael Heaver;UKIP;UKIP;East of England;4; Getting out of the EU, proper border controls, creating new grammar schools: that’s what UKIP voters believe in and vote for.

Although similar in the message they convey, the argumentative patterns of the two tweets are different. The first pattern is built, as follows: the problem is highlighted, Ukip’s position to the problem is stated (Only UKIP is unequivocally opposed to them) and the solution to the problem is offered (Vote #UKIP). In terms of argumentation, the following set of explicit and implicit premises can be identified: the UK has a lot of wind turbines; the wind turbines are bad (implicit); UKIP is against them; UKIP wants, can or will do something about them (implicit). And the reasoning behind this pattern is represented by the following inference: if you want less // no on-shore wind turbines in the UK, then vote UKIP. UKIP is, therefore, implicitly presented as being the solution to the on-shore wind turbines problem.

In the second pattern the solutions to more problems are first offered, followed by an emphasis (that’s what UKIP voters believe in and vote for) on the implicit premise represented by UKIP is the solution to the mentioned problem. In this case, the interpretation is based on the following reasoning: if you want these problems solved, then vote UKIP. In sum, an explicit image of UKIP as having opposite values, of being different from the other parties is created in both cases. The negativity is explicitly represented by the dichotomy us vs. them, i.e. UKIP vs. the opponents.

The reasoning behind the tweets is conditional and has the following form: if you want the problem x solved, then vote UKIP. The interpretation is reached←162 | 163→ by undergoing the same type of conditional reasoning that was activated in the interpretation of the hashtags, discussed in the previous section. In contrast with the consequent which is always the same (vote UKIP), the antecedent of the if…then construction varies corresponding to different factual aspects. Therefore, the interpretation of the if…then construction is not based on truth-conditions but the accent is rather placed on how persuasion is arrived at. To this end, the context and people’s knowledge are important factors that influence the derivation of of this structure.

Based on pragmatic inference, the conditional reasoning is a rich source of context and content effects. Unlike deductive inferences, where the conclusion is a logical result, always derived from the stated premises, pragmatic inferences yield probable conclusions which are only likely to be true. Additional premises that represent relevant factual information may be added to the initial premises which increase the semantic information. Looking at the tweets, it can be noticed that the conditional reasoning is activated by the vote UKIP collocation used at the end of the tweet. In other words, the vote UKIP structure plays an important role in the interpretation process as it helps the reader to retrieve the implicit premise.

(c.) The third category of tweets is represented by the instances where the premise UKIP is the solution is left implicit. It is a complex category, the tweets focusing on complex arguments built mainly around anti-Europe. Therefore, UKIP is presented as being and standing for the opposite values and policies in comparison with the unfavourable depicted reality. The question why the electorate should vote UKIP is usually left implicit and the degree of negativity in these tweets is high. A dichotomy between an unfavourable, negative reality and the solution to those problems expressed by vote UKIP is created. The following examples are representative for this category:

(33) Andrew McNeilis;UKIP;UKIP;London;6; Check out low EMPLOYMENT rates of EU economic miracle. Spain, Italy Romania would you want that for uk! Vote UKIP http://t.co/ZR1C3SeuTy

(34) Jonathan Arnott;UKIP;UKIP;North East;1; 10,000 EU officials earn more than our Prime Minister. What a waste of our taxes. Vote #UKIP today http://t.co/HntjJ8p2qu

The same pattern can be identified in these tweets: a fact is presented, followed by a personal comment and the solution (Vote UKIP) is presented at the end of the tweet. Although personal, the comment is presented under a general form. Without explicitly committing and assuming his remarks as there is no first-person singular pronoun used, by presenting negative information the political candidate appeals to the electorate’s emotions in the attempt to capture their attention. Moreover, negative items are said to be remembered with more detail←163 | 164→ and for longer periods in comparison with positive and neutral items (Mickley and Kensinger 2008).

The vote UKIP collocation used at the end of the tweet is rich in content as it fosters a positive image of the party the politician is representing, on the one hand, and guides the readers towards the intended interpretation by encouraging them to undergo a conditional reasoning, on the other.

Although at first sight the following tweet seems fragmentary and incomplete, its logical form and argumentative orientation can be easily retraced:

(35) Michael Heaver;UKIP;UKIP;East of England;4; Of course, you can only control immigration by leaving the EU. So vote UKIP today.

The reasoning behind the tweet suggesting the solution to the immigration problem seems to be based on the following premises: the UK has an immigration problem; the immigration problem can be controlled contrary to what others say; leaving the EU solves the immigration problem; Ukip wants to leave the EU; which lead to the following conclusion: if you want the immigration problem solved, then vote for Ukip.

Some inferential work is needed in order to reach the intended interpretation. There are some discursive clues which guide the interpretation process. The discursive markers of course and so play key roles in providing the necessary contextual information. Of course is highly conversational and emphasizes the fact that the tweet was sent in response to a previous message debating the immigration problem and the UK leaving the EU. Therefore, this tweet is part of a larger conversation built around some key topics addressed during the 2014 European elections, such as the immigration problem and the UK leaving the EU. The discursive marker also emphasizes the existence of more opposing points of view, the tweet representing a counter argument in the debate regarding immigration. The second discursive marker so, functions as a conclusion: So vote UKIP today. In other words the solution to the immigration problem is represented by leaving the EU, action that is fully supported by UKIP.

The same type of conditional reasoning can be identified in the following examples as well:

(36) Andrew McNeilis;UKIP;UKIP;London;6; More from the ministry of fear – don’t forget the govt funds this think tank. Love Britain? Vote UKIP http://t.co/Py0jBNp38g

(37) Andrew McNeilis;UKIP;UKIP;London;6; Ok. Imagine the media over last 5 weeks played the ball (issue) not the man (Farage etc) the truth is do you love the uk? Then vote UKIP←164 | 165→

The if…then structure has a different form in these tweets. While the consequent is represented by the vote UKIP collocation, the antecedent is represented by the question: Love Britain? which leads to the following recreated structure: if you love Britain, then vote UKIP. The same pattern with a small variation at the level of the antecedent can be noticed in the next example:

(38) Tim Aker;UKIP;UKIP;East of England;3;Tory change in the EU = £55 m a day to the EU, more EU immigration, more EU laws. Love Britain, Vote #UKIP

The content of the tweet makes reference to the newspaper article which considers the UK leaving the EU not to be an appropriate measure. The political candidate holds different views and accuses the party in office of supporting the issues discussed in the article. By calling the party in office ministry of fear, the political candidate attempts to create a negative image of his opponents and, implicitly, promotes his own image and that of his party.

At first sight, the reasoning line is represented by the following premises: UKIP disagrees with the current opinion of the government; UKIP thinks they are playing EU’s interests and not Britain’s which lead to the conclusion: If you love Britain, then vote UKIP. Therefore, in the attempt to recreate the premises of this argument, a transition between political aspects to loving UK correlated with voting UKIP can be noticed. In other words, the antecedent built around loving UK encompasses more layers of meaning: being against the current government and political aspects, on the one hand, and supporting UKIP which represents a change in the political landscape, on the other. At a closer look, based on the following premises: leaving the EU means loving Britain; UKIP wants Britain to leave the EU; UKIP loves Britain more embedded if…then constructions can be identified, as follows: if you love Britain, then you want the UK to leave the EU; if UKIP is elected, then Britain will leave the EU; to culminate with: if you love Britain, then vote UKIP.

Looking at the argumentative line, a strong appeal to emotion can be noticed in comparison to a logical argumentation. The lexical markers focusing on feelings (ministry of) fear and love, accompanied by the urging advice don’t forget are meant to manipulate the electorate’s emotions, influencing them in the process of decision making. It is a fallacious argumentation, based on argumentum ad passiones, because inciting emotions in people does not serve as evidence for a claim. The same pattern and the same argumentative line can be noticed in (37) as well. The consequent is explicitly expressed by then vote UKIP and the antecedent is represented by the same question do you love the uk?. In this example, UKIP is equalled to the combination expressed by values of truth and love. The appeal←165 | 166→ to the electorate’s patriotism and feelings of belongingness leads to a fallacious type of argumentation.

In conclusion, all three categories of tweets are built on the same cognitive pattern, based on the premise according to which UKIP is presented as being the solution to the problems depicted more or less explicitly. The conclusion is reached inferentially, only some clues being given that lead to the intended interpretation. Negativity is an inherent property of all the call for action tweets under investigation. It is displayed in various degrees and can be found in the grammatical forms used, in the hashtags and in the hyperlinks accompanying the tweets. To reach the intended speaker’s meaning, the reader has to undergo an inferential route represented by the conditional type of reasoning, derived in a contrastive context of interpretation.

Conclusions

This paper aimed at discussing the specificity of the call for action tweets sent by the UKIP political candidates during the 2014 European Elections. The call for action tweets are the political messages sent usually towards the end of the electoral campaign when the discourse becomes asymmetrical, as a one-way transfer of information from the politicians to the electorate. Based on the premise that they have different impacts on the people’s cognitive environment and they can influence the electorate’s opinions about certain candidates and political parties, particular attention is paid to the linguistic strategies and grammatical forms the politicians use.

A general description based on Baym’s seven key concepts is suggested in order to highlight Twitter’s individuality in comparison with face-to-face communication, on the one hand, and with other forms of computer-mediated communication, on the other. As a result of the main linguistic features, such the brevity constraint and the absence of any verbal, non-verbal and para-verbal components, the tweets represent particular samples of language use whose derivation and interpretation led to the development of novel techniques.

Twitter has been used in politics on a daily basis, as a channel of choice in diplomacy and during election times. The politicians choose Twitter as an important communication tool during the electoral campaign for several reasons: they can address wider and collective audiences, it offers them real-time access to the public sphere, it grants a new dimension to the public debate, and it allows the politicians to get personally involved and take individual initiatives as the message is un-mediated by the media. Due to the extensive use of the platform in all 28 member states, the 2104 EE are considered the “Twitter elections”.←166 | 167→

The analysis is two-fold, organised around the vote UKIP collocation and around the content of the tweets. A formal analysis of the vote UKIP collocation is provided with an emphasis on the following aspects: the orthography, the importance of the different positions it occupies in the tweet, the semantic alternatives used for the collocation as well as the combination between vote UKIP and different semantic categories. The vote UKIP collocation is also used in combination with the # hashtag and @-mention operators, as follows: #VoteUKIP, vote #UKIP, vote @UKIP, vote for #UKIP, voting #UKIP. A correlation between the place of the hashtag in the tweet and the role it fulfils has been identified: when it is added at the end of the tweet without any apparent relation with the body of the tweet, it functions like a slogan and has a strong cognitive impact on the electorate.

Special emphasis is put on the functions the hashtags have. Apart from making the tweet visible, organising the election conversation, drawing attention to it and making it user-friendly, the hashtags have developed a special function which plays an important role in the process of meaning derivation and interpretation. The #VoteUKIP hashtag creates a contrastive context of interpretation which generates a conditional if…then type of reasoning.

The content analysis concentrates on two main types of tweets: informative and argumentative. While the informative ones are meant to recall the vote and to invite the electorate to vote for their party, the argumentative ones seek to provide various arguments to the question why the electorate should vote UKIP. The analysis reveals that the argumentative tweets are characterised by a contrastive negative campaigning argumentative orientation, as the political candidate choose to always compare and contrast themselves with the opponents. Consequently, a constant dichotomy between the negative image of the opponents and the positive image of UKIP presented as being the solution to the mentioned problems is created.

In the attempt to indicate how the contrastive negative campaigning argumentative orientation is realised a set of criteria has been suggested based on the following points of discussion: how the question why UKIP should be voted for is addressed, how the premise (UKIP is the solution to the problem x) is expressed, how the negativity is expressed and what type of reasoning is activated.

Based on these criteria, three categories of tweets have been identified, as follows: (a.) the question is expressed directly and the premise is identified with a minimum of cognitive effort; (b.) the question is indirectly addressed and identifying the premise does not require a great deal of cognitive effort; (c.) the question is in the background and the premise is left implicit and therefore more cognitive effort is required in order to reach the intended interpretation. All the tweets←167 | 168→ are built on the implicit premise Ukip is the solution and a contrastive context of interpretation that generates a conditional type of reasoning is created.

The results cannot be taken as representative of the entire behaviour of Twitter during the 2014 European elections and are best regarded as a selective snapshot of the political activity of the UKIP political candidates. Further comparative analysis between other parties in the UK but also between different European countries is required in order to have a more comprehensive picture of the Twitter use during the 2014 European elections.

References

Abroms L., and Lefebvre C. (2009). Obama’s Wired Campaign: Lessons for Public Health Communication. Journal of Health Communication, vol. 14, p. 415–423.

Anthony L. (2014). AntConc (Version 3.4.3) [Computer Software]. Tokyo, Japan: Waseda University. Available from http://www.laurenceanthony.net/.

Baym N. (2010). Personal Connections in the Digital Age, Polity Press, Cambridge.

Broersma M. and Graham T. (2012). Social media as beat: tweets as news source during the 2010 British and Dutch elections. Journalism Practice, vol. 6(3), p. 403–419.

Bruns A. and Burgess J. (2011). #ausvotes: how Twitter covered the 2010 Australian federal election. Communication, Politics & Culture,vol. 44(2), p. 37–56.

Conway et al. (2012). Twitter Use by Presidential Primary Candidates During the 2012 Campaign. American Behavioral Scientist, vol. 57(11), p. 1596–1610.

Einspanner-Pflock J., Anastasiadis M., and Thimm C. (2016). Ad Hoc Mini-Publics on Twitter: Citizen Participation or Politicical Communication? Examples from the German National Election 2013. Citizen Participation and Political Communication in a Digital World, New York, Routledge, p. 42–59.

Gaffney D. (2010). #iranElection: quantifying online activism. Paper presented at WebSci10: Extending the Frontiers of Society On-Line, Raleigh, NC: US.

Golbeck J., Grimes J.M. and Rogers A. (2010). Twitter Use by the US Congress. Journal of the American Society for Information Science and Technology, vol. 61(8), p. 1612–1621.

Graham T., Broersma M. and Hazelhoff K. (2013). Closing the gap? Twitter as an instrument for connected representation. The Media, Political Participation and Empowerment. Routledge, London, p. 71–88.

Ikiz et al. (2014). Political Use of Twitter: The Case of Metropolitan Mayor Candidates in 2014 Local Elections in Turkey. ICEGOV ‘14 Proceedings of the←168 | 169→ 8th International Conference on Theory and Practice of Electronic Governance, ACM, New York, p. 41–50.

Larsson, O. (2012). Studying political microblogging: Twitter users in the 2010 Swedish election campaign. New Media & Society, vol.4, p. 729–747.

Mattes K. and Redlawsk D. (2014) The Positive Case for Negative Campaigning, University of Chicago Press, London and Chicago.

Smyrnaios N. (2014): #EP2014: the European election through the lens of Twitter, http://ephemeron.eu/1219.

Scott K. (2015). The pragmatics of hashtags: Inference and conversational style on Twitter. Journal of Pragmatics, vol. 81, p. 8–20.

Sperber D. and Wilson D. (1995). Relevance: communication and cognition, Wiley- Blackwell, Oxford.

Thorsen E. (2014). Online Reporting of Elections. Routledge, New York.

Tumasjan et al. (2010). Predicting Elections with Twitter: What 140 Characters Reveal about Political Sentiment. Paper presented at the Fourth International AAAI Conference on Weblogs and Social Media, George Washington University, Washington, DC.

Vergeer M., Hermans L. and Sams S. (2013). Online social networks and micro-blogging in political campaigning: the exploration of a new campaign tool and a new campaign style. Party Politics, vol. 19(3), p. 477–501.

Zappavigna M. (2012). Ambient affiliation: A linguistic perspective on Twitter. SAGE Internet Research Methods, Vol. 4: Research ‘On’ and ‘In’ the Internet – Investigating the Online World. London SAGE Publications Ltd, p. 193–215.←169 | 170→ ←170 | 171→


90 https://support.twitter.com/articles/215585?lang=en

91 http://twiplomacy.com/blog/twiplomacy-study-2015/#section-22

92 http://twiplomacy.com/blog/twiplomacy-study-2015/

93 Tony Lockett’s blog European Union 2.0: https://tonylbxl.wordpress.com/2014/02/16/will-2014-be-europes-twitter-election/

94 Although only the second part represents a proper hashtag, vote #UKIP will be analysed like a collocation, as it seems to form a complex unit in the corpus of data.