Election campaigning enters a fourth phase: the mediatized campaign

Abstract

Thus far, the literature on election campaigns has identified three phases of campaigning: the pre-modern, the modern, and the professionalized phase. With this paper, we suggest that election campaigning has entered a fourth phase, characterized by new applications of communication technologies, quantitative data, immediate communications, a reinvented citizen-politics relationship offering more emotional access and lower barriers for active roles for citizens in campaigns. We root the change in the citizen-media nexus, with changes in media use and production being a major driver of this development. The classification of the ‘Mediatized Campaign’ emphasizes the role of media use and the connected changes in political and social institutions. We suggest that this classification can add coherence to future research on campaigning.

Zusammenfassung

Die wissenschaftliche Literatur unterscheidet bis dato zwischen drei historischen Phasen der Wahlkampfkampagne: die vormoderne, die moderne und die professionalisierte Phase. Mit diesem Aufsatz schlagen wir eine vierte idealtypische Phase der Wahlkampfkampagne vor. Diese vierte Phase setzt sich von den vorherigen Phasen durch die intensivierte Nutzung neuer Kommunikationstechnologien, quantitative Datenanalyse und auf unmittelbarem Austausch basierende Kommunikationsparadigmen ab. Ein weiterer Fokus liegt in dieser Phase auf dem neu entstandenen Verhältnis zwischen Bürgern und Politik, in dem der Zugang zur Kampagnenführung niedrigschwelliger ist und die emotionale Bindung in den Vordergrund gerückt wird. Veränderte Nachfrage, Anwendung und Produktion von Medien stellen die zentralen Entwicklungen in dieser Phase dar. Im Spannungsfeld zwischen Bürger und Medien ordnen wir den Hauptimpuls für den Wandel ein. Die neue Phase, die wir als ‚Mediatisierte Kampagne‘ bezeichnen, hebt die Rolle der Medien hervor und macht auf die damit verbundenen Entwicklungen in politischen und gesellschaftlichen Institutionen aufmerksam. Dieser Ansatz verleiht künftiger Wahlkampfforschung eine kohärente Struktur und weist eine potenzielle Forschungsrichtung aus.

1 Introduction

In the last two decades, a new research field has developed at the intersection of party research, electoral research, and communication research: campaign research. Whereas the first and for a long time last major study on the question of campaigns and campaign effects was conducted in the 1940s by Paul Lazarsfeld and his colleagues (1968) (1944) documented in the seminal work The People’s Choice it took more than 50 years for researchers to focus on the issue of campaigns again and establish a research area which is now flourishing. The reasons for this long negligence have become clear: campaigns had little or no impact on electoral behavior and choice due to a strong party identification, therefore: why study them? It was no surprise that “the study of election campaigns, as opposed to elections, is a major gap in the literature” (Harrop and Miller 1987, p. 240).

This gap has been closed. A growing body of scholarly work now focuses on campaigns, with a specific emphasis on “change” (Schmitt-Beck and Pfetsch 1994; Norris 2000; Farrell and Webb 2002; Römmele 2005; Gibson and McAllister 2011). With the weakening of party loyalties (Gibson 2013; Gibson and McAllister 2013; Mair 2013) and a growing number of undecided and swing voters, campaigns have become more important. With the war room-campaign style of Bill Clinton in 1992, Tony Blair’s New Labour Campaign in 1997 and Gerhard Schröder’s Kampa-campaign in 1998 a new style of campaigning emerged. “Ultimately, campaigns can affect who wins the election. Thus, the strategic decisions of candidates are not merely empty exercises. […] Both the inputs and outputs of campaign processes can be consequential.” (Brady and Johnston 2006, p. 18) But how campaigns go about affecting change differs according to many factors.

A remarkable body of literature emerged emphasizing the change in campaign style and dynamics (Bowler and Farrell 1992; Swanson and Mancini 1996; Norris 2000; Plasser and Plasser 2002; Römmele 2005) also pointing out that campaigns are not isolated events that happens once every four or five years but tells us how citizens and politics interact and engage – “campaigns are the prototype situation of political communication” (Schmitt-Beck and Pfetsch 1994). From a historical perspective, three campaign eras are distinguished in the literature (Norris 2000; Römmele 2005). In a first, or pre-modern era, campaigns were based on the strength of the local party organization and face-to-face contact. This phase is roughly located from the 1920 until the end of WWII. The second wave of campaigning saw a shift from communication via the party organization to mass media communication between parties and voters. The rise of television had tremendous implications for political communication. “During the post-war era, political campaigning has been transformed by the decline of direct linkages between citizens and parties and the rise of mediated linkages” (Norris 1998). With mass media, especially television, parties can communicate to a broader audience. Because party identification and party attachment have declined, parties not only have to mobilize their electorate, they also have to convince the undecided voters of their party program. Within the party organization, more and more power is shifted to the party headquarters, which assumes responsibility for overall campaign strategy.

A third phase of campaigning was identified by scholars in the early 1990s. While initially it was referred to in generic terms as an Americanized style of campaigning (Negrine and Papathanassopoulos 1996), it also had more historical or developmental labels applied to it, such as postmodern (Norris 2000), phase 3 (Farrell and Webb 2002), and post-Fordist (Denver and Hands 2002). Despite these differences in nomenclature, there is considerable agreement between these scholars as to the central features of the third era in campaigning (Gibson and Römmele 2001). First, the tools or mechanics of campaigning changed with the adoption of new communication technologies such as the Internet, direct mailing, and phone banks. The innovations at that time occurred alongside an intensification of existing methods of divining voters’ thoughts, such as opinion polls and focus groups. Additionally, there were changes within parties, with a shift of power upward to leaders and outward to external media and public relations consultants. Most fundamentally, perhaps, the overall style of political campaigning is seen to have become more business-like in its approach. Parties continuously interact with voters to market their products, that is, policies (Butler and Collins 1994). In the third phase, voters are seen more as consumers than loyal partisans, to be wooed with sophisticated advertising rather than serious political education.

This paper puts forward the argument that we are observing the emergence of a 4^th phase of campaigning which fundamentally changes the way politics presents itself and citizens engage in it (see Table 1). The fourth phase is characterized strongly by the hybrid communications environment in which it takes place (Chadwick 2013), as well as technological developments enabling widespread use of new communication mediums and a stronger orientation towards the use of data. Contextually, we link this to the process of mediatization, which “refers to a social change process in which media have become increasingly influential in and deeply integrated into different spheres of society.” (Strömbäck and Esser 2014, p. 4) The 4^th phase of campaigning is deeply embedded in the process of mediatization, which in turn is best understood in the way that media and politics relate to one another as institutions (Hjarvard 2008, 2013). Although displaying significant differences from traditional media institutions, significant digital media actors, for example social media platforms, are nonetheless autonomous commercial and competitive actors. The process of content creation in the mediatized campaigns has thus become both more and less professionalized at once. Therefore, this phase, although also heavily influenced by communications and campaign professionals, is also characterized strongly by more recent developments such as citizen-initiated campaigning (Gibson 2013), relying much more on non-professionalized peer-to-peer conglomerations of citizens whose motivations overlap at strategically important moments. The distinction between online and offline aspects has been subsumed by the large-scale permeation of an all-hands-on-deck approach to the total campaign.

Table 1 The 4 phases of campaigning (expanded on from Farrell and Webb 2002)

Our analysis is guided by the notion that political communication is an interactive system (Blumler and Katz 1974) in which changes in one part of the system bring about changes in the other. The major changes in this system which bring the mediatized campaign to prominence stems from the citizen-media nexus, producing a reaction from political actors to the changes in production and consumption of media.

2 The mediatized campaign

The mediatized election campaign is an ideal type (Weber 1949/2011) and thus what we propose in this paper is an analytical tool and not an empirical reality. The characteristics that we suggest make up mediatized campaigns are not necessarily found in any one campaign; the value of the concept lies in the analytical and explanatory power for describing and comparing election campaigns. The concept is not intended to be a rigid classification but a point of departure for further research. It suggests a hypothesis that there is an observable shift in the way election campaigns happen and that there are recognizable commonalities which characterize that shift.

We begin with a characterization of mediatized campaigns, laying out differences but also similarities with former phases. Divisions between the phases will often be blurred, as the shift to mediatized campaigns is an uneven process. The following characterization is described by organizing the changes that lead to the hypothesis of mediatized campaigns into categories attached to the main actor groups involved in the political communication system: Citizens, Media, and Politicians. Accordingly, we address each of these parts of the interactive political communication system in turn, before continuing on with a discussion of what the implications of these changes are for politics and democracy. Table 2 offers an overview of the characterization.

Table 2 Major campaign characteristics: the mediatized campaign

2.1 Citizens

Current trends in political engagement show a decrease in long-term types of engagement, such as party or union membership, active participation in religious institutions, or other traditional forms of support based on overarching ideologies (van Biezen et al. 2012; Gordon et al. 2013; Mair 2013). That, however does not imply that political engagement as a whole is decreasing. Non-permanent activities linked to certain issues, such as issue-voting, political consumerism, signing petitions, or short-term volunteering are becoming more prominent (Dalton 2008; Dahlgren 2009; Gibson and McAllister 2013). This requires a shift in the way social institutions interact with each other, not least politics and media. The result for campaigns is a process of learning how to make efficient use of the types of resources available to them now (e. g. cognitive surpluses). This means making even sporadic and limited engagement productive. Capitalizing on ad hoc and non-committal forms of engagement is certainly a characteristic of the mediatized campaign. The reluctance to permanently associate oneself with a political party results in a political landscape where citizens involve themselves in politics in a way that resembles interest oriented clustering (Bennett 1998; Castells 2008). Thus campaigns also focus on garnering the support of ad hoc affiliates while still maintaining a fading base. The result are election campaigns that resemble more and more issue-based campaigns common in civil society, or of social movements.

A further significant change for citizens is the amplified opportunity to be a prosumer, an active producer as well as consumer of campaign content. Becoming an active producer for the campaign goals is not limited to postings on social media, but refers to any ways citizens are activated to take over a part of the campaign effort. The role of journalism is also necessarily transformed due to the prosumer function of citizens; citizen journalists and citizen campaigners are enabled to create and channel content for insertion into the public sphere. This need not imply that the barriers to being heard in the public sphere are lower, but that they function in a different way. It is no longer a question of access in terms of the traditional skill-set, connections, and finances necessary to make a voice heard, but rather how narratives and storytelling are woven into the fabric of the broader campaign. This continues to rely on elite networks, financial resources, and expert abilities. In part because of the accessible and collective nature of phase 4 campaigns, the policy issues form the backdrop for what is going on at the center, namely including citizens in the narrative of the campaign and its potential success. This often entails the public discourse being about citizens and their relations to the campaign and the campaign story, sometimes at the cost of the specific policy details associated with a campaign.

Citizen involvement in mediatized campaigns can be motivated less by an extrinsic obligation or sense of duty and more by an intrinsic desire to shape responses to perceived social problems. Politics becomes hyperlocal, it becomes personal. This can be observed in many contexts where a traditional right-left organization of political allegiances is supplemented or sometimes even superseded by values which are absolute and thus cannot be placed on a traditional left-right continuum (Voltmer 2012).

As regards the agency of citizens in mediatized campaigns, it can be said that they tend to be more enabled, but not necessarily more empowered. The campaign can profit from citizen involvement even when citizens choose how and when to engage in campaigning activities, but the parameters are clearly set as to what results from their engagement. Formats and content prepared by professionals close to the campaign core are made to determine the output from volunteers further from the command center. Thus, although transformed, agency of citizens in mediatized campaigns need not increase.

2.2 Media

Mediatized campaigns exist in the context of the Hybrid Media System (Chadwick 2013). Hybridity implies that there is no one clear media logic or channel. Rather, actors in the hybrid media system use whatever medium or channel deemed most appropriate (limiting the usefulness of an online vs. offline distinction). The system is multimodal, built upon interactions among newer and older media, resembling a process of integration. In the hybrid context, not only does the media as a system blur boundaries with various mediums and logics merging and becoming intertwined, hybrid genres also offer new potential (i. e. comedic news, semi-fictional mockumentaries, docu-soaps, game-docs etc.). Actors have complex and evolving power relations based upon adaption and interdependence. Campaign success is thus also contingent on the act of prioritizing and juggling many mediums across the contemporary mediascape.

Furthermore, the contemporary media context enables candidates to ‘directly’ interact with more voters via digital media than was ever before possible. Receiving a response to one’s own question, or a message of appreciation for action taken on the candidate’s behalf, or any type of personal acknowledgement enables certain citizens to feel connected to candidates on an emotional level (Lee and Oh 2012). This possibility was limited in previous campaign models, where physical and temporal co-presence was required for this kind of bond. The immediacy of digital media interactions can strengthen the feeling of personal connection to the candidate, a stronger bond than the one-way support characterizing previous campaign phases. Furthermore, these types of interaction can produce feedback for campaigns, qualitatively as well as quantitatively, which feed into their efforts supported by data (more on this below in the section on Politics).

The nature of the contemporary communication context is dynamic and inconsistent, due in part to rapid developments in communication technology. Garrett et al. (2012) have dubbed these developments sociotechnical changes, which carry with them profound potential: “New ICTs [information and communication technologies] have the power to alter the way politics are done. From political discussion to political deliberation, and ultimately, to political engagement, technologies have altered the nature of democratic discourse.” (Garrett et al. 2012, p. 218) These alterations remain a potentiality until they can be observed. This does not imply technological determinism. It means that from phase 3 to phase 4 of campaigning, practices have not necessarily changed due to technological progress, but because actors employ (sometimes new) technologies for campaigning purposes differently.

Unevenness characterizes application of digital media in mediatized campaigns. Nielsen (2015) refers to the “prominence of social media in different policy processes” as “deeply uneven.” (Nielsen 2015, p. 2) Some contexts may have high penetration of internet services, while in others this is not the case. Some countries have legislative contexts that prohibit the amassing of data about individuals, in other countries regulations may not be so encompassing. Some campaigning contexts strictly regulate the use of mediums like television or newspapers for campaigning purposes, while others have only financial rather than legislative restrictions (Hallin and Mancini 2004). Technology development often exceeds the speed of regulation (Jonhson et al. 2014), resulting in a grey area for what can be used or done in election campaigning. mediatized campaigns are often characterized by a realm of campaigning communication which is unregulated or unclear regarding what is and is not allowed (Baishya 2015).

Professionalized (phase 3) campaigning had a strong emphasis on traditional, one-way media – the message was controlled by party headquarters and the overall communication goal was being and remaining “on message.” New ICTs were used, but as a one-way communication tool. In phase 4, there is a strong emphasis on interactive digital media and the overall approach to web based communication is different. In the hybrid media context, direct communication is not only an ends in itself, but also allows new possibilities to leverage traditional media. Tweeting in time for the evening news, or ‘leaking’ information to active supporters via SMS before making a press release about it are examples of the complex and multifaceted nature communication possibilities in the hybrid context.

Low barriers to participate invite citizens to be an active part of the campaign. This represents a major departure from the former professionalized or phase 3 campaigning models: citizens and voters are no longer merely the object of campaign strategies, in mediatized campaigns they have become the subjects as well. The current phase 4 campaigning instrumentalizes voters. As Cormode and Krishnamurthy (2008) suggest, “the democratic nature of Web 2.0 is exemplified by creations of large number of niche groups (collections of friends) who can exchange content of any kind (text, audio, video) and tag, comment, and link to both intra-group and extra-group pages.” The resulting change for campaigns lies in the way information can be produced and shared (Gibson 2013; Copeland and Römmele 2014). User ability to share political information through social media makes passive or incidental exposure to political information through social media possible; unlike in the Web 1.0 era, people no longer have to actively pursue political information to encounter it (Copeland and Römmele 2014).

2.3 Politics

The political context of mediatized campaigns has become more dynamic and reactive. Politicians are constantly interacting with their constituents, justifying actions, seeking support, or determining constituent interests and policy preferences. There is necessarily a campaigning element in all political communication in the 4^th phase; campaigning is a permanent status. The notion of a secure mandate due to an attachment to an overarching ideology is losing ground. Trust becomes transformed. The ease of information transfer has to some extent eroded the need for trust; citizens can check up on candidate’s voting records, press releases, and a whole host of further information at a very low cost, if and when they are interested. This implies a change in the function of media as an institution, as the way the public interacts with politics.

Regardless of the extent to which digital communication technologies are employed by campaigners, their potential cannot be negated. New political and technological methods allow for political parties to “scan the public sphere for concerns that their candidates for office should address or avoid” using methods like data and text mining (Krippendorff 2013, p. 221). Data on citizen’s online activities can similarly provide campaigners with insights into what voters care about, not only that, but where, when, and to what extent.

2.3.1 Personalization and branding

In phase 4, citizens are active agents, carrying out crucial aspects of the campaign. This decentralization activates citizens by giving them ownership of the processes and content, but not more agency in terms of influencing the content of the campaign. The message is still tightly controlled by the candidate and her team. Well-funded and managed campaigns in the 4^th phase devote considerable resources to micro-managing the messages in a wide variety of communication arenas. They do this by feeding volunteers specific messages and talking points, emphasizing what needs to be said and where. In some instances large teams of formal employees can be devoted to managing and steering conversations online, along with the help of volunteers or individuals with weaker ties to the campaign.

Nonetheless, the broad involvement and inclusion of citizens in campaigns, as well as the transfer of ownership in this sense also serves as a legitimating function, putting the campaign in the hands of the people. This requires careful management so that it is successful, and it should not be taken to imply completely flat hierarchies and liquid messaging. The involvement of citizens as drivers of a campaign has a considerable connection to the candidate as a person. A firm branding of a candidate makes room for slightly less control of messaging, as a personal image can accommodate minor discrepancies more aptly than can a clear set of ideas.

Not surprisingly then, mediatized campaigns often foreground the candidate. This allows for a perception of a political candidate as more than just the sum of their political positions. Candidates have always played an important role in the overall campaign concept, literally embodying what the campaign stands for, but in phase 4, the role of candidates as a brand is emphasized more than ever before. As Scammell (2007) points out, branding is now a permanent aspect of campaigning. Anholt and Hildreth (2004) define a brand as “nothing more […] than the good name of something.” Originally from the marketing literature, branding is seen as shortcut to consumer choice enabling differentiation between consumer products. This becomes more and more relevant as democracies shift from party democracies to audience democracies (Pfetsch 2014).

The original model of the permanent campaign was characterized inter alia by continuous polling, intense news management, and constant attention to media images. This model now is outdated (Needham 2005; Scammell 2007). Whereas the permanent campaign emphasizes the instruments of media politics, the brand concept covers the underlying strategic concerns of efforts to maintain voter loyalty through communication designed to provide reassurance, uniqueness, consistency of values, and emotional connection with voters’ values and visions of the good life (Scammell 2007, p. 188). The emphasis in branding very much is on the image as well as the emotional connection. It is largely defined by the undecided voters – a striking difference from commercial branding where much of the effort is directed at retaining the loyalty of existing consumers. Prominent examples of this type of branding include the Obama campaign’s success in linking the candidate to the captions of ‘hope’ and ‘change’ in 2008, Angela Merkel’s widely recognized hand gesture featured prominently on her party’s advertisements in 2013, or Alastair Campbell and Tony Blair’s rebranding of Britain’s Labour party as ‘New Labour.’

The ability to custom-make the face of a campaign to fit in with the concerns and priorities of individuals is also a defining aspect of phase 4, and data based strategies make it possible to know which issues to emphasize, when, and to whom. This also fits together with the role of the citizen in mediatized campaigns, who need not simply observe, but can also engage and shape mediated discourses.

2.3.2 Consultants

The role of professional campaigners and consultants has also been broadened due to the use of big data. Formerly, a key characteristic of consultants was a long lasting and close relationship with the candidate. As campaigns have developed, so the role of consultants has become professionalized and a whole consultancy industry has developed. One of the indicators for this trend is the formation of professional associations (like degepol¹ in Germany, the American Association of Political Consultants², The Brazilian Association of Political Consultants³, the Association of Professional Political Consultants in the UK⁴, The International Association of Political Consultants⁵, and the European Association of Political Consultants⁶) which we see in numerous countries and regions. Journals have emerged that concentrate on campaigns and the consultancy industry (Campaigns and Elections, Zeitschrift für Politikberatung, Journal of Political Marketing) and universities offer programs that educate and train young consultants-to-be, for example the well-known MA in political management from George Washington University⁷ in the USA.

But what has changed as regards the role of consultants if we juxtapose phase 3 and phase 4-campaigning? In phase 3, pollsters and media analysts played a crucial role. Some of the key advisors in campaigns had a leading role in media before getting involved in the campaign.⁸ In addition to all their merits, the consultants of phase 3 also had to have one central credential: they had to be ‘political animals’ with an instinct of what works and what does not work. Issenberg calls these consultancy types Gurus. “Gurus were the celebrated political wise men whose practices had become the industry default, thanks to their success serving up a cocktail of lore and myth, anecdote and inertia, able to so thoroughly intoxicate the candidates who paid their bills.” (Issenberg 2012, p. 147).

In phase 4 we see a new garde of consultants. Along with the ability to gather, store, and process big data came the need for human capital to carry out these tasks. This changes the nature of political consulting as well as the people involved, generating and feeding a demand for quantitatively oriented campaign consultants (Nickerson and Rogers 2014). Micro-targeting becoming prevalent leads to questions campaigners often pose to be formulated more specifically like in A/B testing. The ability to analyze data to pinpoint which messages should be used when, where, and for how long requires quantitative, evidence-based arguments for campaign action. Data was used in former phases of campaigning as well, like the results of focus groups or polls used in the 3^rd phase. In mediatized campaigns, quantitative data, often proprietary data maintained by companies, campaigns, and/or parties is what drives campaigns. Consultants with a quantitative and methodological orientation replace to some extent the spin doctors of former phases.

3 Characteristics of mediatized campaigns and implications of for democracy and politics

3.1 The rapid-response campaign

The speed and immediacy of communication affects more than just the speed of the political process. Candidates can directly react to evolving circumstances, such as refuting statements made by their competitors, or taking a position on new developments. Uncertainty is significant because it can undermine the feeling of citizens that the political actor they support will act in line with their expectations. This development may be linked to the transformation of citizen trust in politics in that if trust is less exist, uncertainty can be more damaging the perception of political actors, leading to a demand for more immediate reactions. For these reasons, political campaigners may see great benefit in minimizing the amount of time the public needs to linger in uncertainty (formerly more closely equated with trust).

The normalization of developments in political practices is cemented when politicians take them up in their campaigning practices. Campaigning practices represent the avant-garde of practices in political communication; they are not yet bound by the restrictions of a public position (except incumbents), and thus have more freedom to try things out. Mediatized campaigns thus foreshadow at least some of the communication practices which can be expected to become common practice in the near future. A term as an elected official cannot be decoupled from the experience of running a successful campaign, especially when the campaigning just begins anew once an election is over.

The complicity of politicians (especially during campaigns) with the speed demanded by citizens and normalized by media practices indicates the importance of mediatization for campaigning and the political process (Blumler 2014). Along with the quickening of political communication and its accompanying appeals to emotional reaction come skepticism that politics can be appropriately understood and reflected in such a dynamic and high speed context (Virilio 2006; Dahlgren 2009).

3.2 The emotional campaign: politics gets personal

An important characteristic of mediatized campaigns is the recognition that the emotional basis for decision-making works in tandem with rational decision-making, rather than replacing it. The immediacy of communications enabled by digital mediums pander to emotion, allowing reactions to be demanded and delivered without time for reflection or even before events fully unfold. The convergence of online and offline similarly implies a blurring of boundaries between public and private, as many individuals are both publically and privately active on an interwoven jumble of online mediums.

This leads to the happenstance political encounters as when citizens are confronted with political news or appeals by nature of their connections bringing them to the fore on individualized social media accounts or ‘personal public’ (Schmidt 2014). Thus, when one is engaged in social activities which are non-political in nature or motivation, the political finds its way in. How much time and effort is put into pursuing these political twists of fate is contingent on many factors, and will differ according to individual context. However, even brief exposure can be relevant for politics. These types of exposure are central to mediatized campaigns; and it is undoubtedly a response to the difficulties faced by political actors in reaching and engaging citizens via more traditional communication routes.

Due to these happenstance political moments and the broader context of digital political communication, politics must become accessible and presentable in its new environment. Politics is facing pressure on many fronts to become soundbyteable, social-mediatable, or voting advice applicable. The medium increasingly demands brief bursts of information: a pregnant quote that lends itself to conversion to a tweet or a short youtube video, a facebook status update, or a yes/no response to a complicated question on a voting advice application (VAA). These may have political meaning and persuasion power in and of themselves, enabling an effect even if the political episode only begins or ends with a brief or coincidental message. This type of encounter with the political has been successful in many recent campaigns (Baishya 2015; Harder et al. 2016), and is thus an important element of the mediatized campaign, and increasingly of politics.

Additionally, due in part to the segmentation of the public (Prior 2007), there is no overarching authority to pass definitive judgement on truth claims (Harsin 2015). This means that the emotional becomes a more common basis for political decisions; an especially effective situation in which to exploit emotional political decision making with partisan ‘authorities’ available to confirm and legitimize feelings. While this may emphasize fact-checking duties attributed to the media (see Graves 2016; Graves and Cherubini 2016), false information, especially in high-partisan contexts can still have substantial effects on perceptions of politicians (Thorson 2016).

3.3 Citizen participation: the ease of access and ad-hoc campaign

The constant communication between political actors and citizens allows for immediate feedback for actions. Campaigns can judge the response to their activities immediately through digital feedback (both in the form of big data and more qualitative approaches). This allows campaigners to legitimize positions as being justified by outpourings of support via real-time digital arenas, or it can indicate missteps with (sometimes) enough time to correct them. Easy access to interaction with candidates and campaigns is expected, indeed political institutions are recognizing the demand for this and while there is no clear recipe for a response, many are being tried out. Due to the normalizing power of campaigns, it is not beyond reasonable expectation that this will soon become a requirement in all realms of everyday politics, not just connected to campaigns.

This reflects a dynamic political atmosphere where politicians must constantly campaign to garner support from citizens, whose short-term political commitments require ongoing effort from a campaign to involve them. Maintaining a structure where support and engagement can be immediately employed productively thus becomes a large part of a campaign’s task, and this remains the case once the candidate is in office. In this way, citizens are inserted into an active role in politics as subjects; in opposition to former phases of campaigning where they were the targets of a campaign as objects. This can imply democratizing aspects, because the distance between citizens and politics can be smaller. If the individual makes the choice to engage in a campaign, it is possible to begin at very little cost to the citizen (or the campaign). This is the case in the 4^th phase phenomenon of citizen-initiated campaigning (Gibson 2013; Copeland and Römmele 2014). The responsibility for a successful campaign depends, as before on citizens. Mediatized campaigns, however, want not only their vote, but as much engagement as they can give.

The open-door style of mediatized campaigns enables engagement from citizens on an engagement a la carte model. Citizens can engage, when, where, and if they choose to. Provided that citizens have the resources required to engage in a campaign, access is easier than in past phases of professionalized campaigning.

Many citizens can be brought to rally around the campaign goal, becoming active during election times and resulting groups of active citizens bound by a common objective: for their preferred candidate to win the election. Upon winning or losing an election however, much of this ad-hoc group dissolves, leaving the candidate with the task of transforming the left-over resources for use in achieving political goals or for reactivation come next election.

3.4 The data-driven campaign

Use of big data in campaigning, a clear 4^th phase trait, remains a problem and opportunity. While the use of strategies based on big data can contribute greatly to the effectiveness and efficiency of campaigns, the ethical use of big data for campaigning purposes will be dependent on the regulatory context and the culture (Dalton 2016). Great differences in the way big data and digital media is applied by campaigners can already be observed between high-income democracies, for example the nano-campaigning tactics on the rise in the USA are unthinkable in much of Northern Europe.

Collecting and integrating data on voters from all areas of a campaign (volunteers, donations, voting records, accounts of activism, support, geographical, consumer, and demographic data, etc.) contribute to a campaign’s efficiency (Issenberg 2012), namely producing the most votes with the least amount of investment. The implications of increasing campaign efficiency based on big data, while it can be very effective, also has side effects. Targeting only citizens who are likely to lend support or change their opinion on a candidate can change the essence of political representation (Hersh and Schaffner 2013; Hersh 2015). If campaigns continue to increase efficiency by using big data, segments of society, for example, who have not voted in the past may become completely neglected from campaign communication efforts in the future. This can pose a threat to social cohesion and could strengthen gaps in society, distancing citizens from political processes. After all, targeting is also a choice about who not to talk to.

4 4^th phase democracy: campaigning is doing politics

The initial stand-alone message represents the top of the pyramid for most who encounter it, but it must be supported with a solid foundation for maximum efficacy. While the political encounter often remains shallow and brief, there will be a minority of citizens who will initiate a follow-up. They can take a number of forms, from issue-voters with a high interest in a specific issue, to need for cognition individuals who require deeper convincing (Lee et al. 2008). These types of citizens are the exceptions to the trend of shallow political communication which characterizes much of the 4^th phase of campaigning. They also impart the requirement for depth of campaigns, because they will be those who will take notice and draw attention if and when any information seems to be being withheld, a significant risk. In mediatized campaigns, these can be citizens or journalists. Therefore it is the networked mass public that creates the demand for high levels of transparency which leads to the type of communication model indicative of mediatized campaigns. Much like the campaigns themselves, the communication of information in the context of mediatized campaigns resembles an open door, allowing and encouraging engagement whenever possible as easily as possible, no prerequisites required. This type of interface is arguably the surface of the campaign, which should ideally encourage citizens to enter and become a part of the campaign. It is this type of surface, or interface, that we can expect to characterize politics, not only campaigns.

That being said, it seems unlikely that the mediatized campaigning is related to the empowerment of citizens. Even with the manifold and easily accessible opportunities for participation, all of the large digital media platforms are commercial entities which accumulate profit from user activities. Citizen political engagement on digital media platforms is thus commodified. The overarching campaign strategy of extracting the most votes with fewest resources aided by market-oriented tactics like big data and candidate branding dampen optimism regarding the normative implications of mediatized campaigns. Efficiency is certainly a major consideration in these campaigns, but is not one which is commonly associated with a healthy democracy.

Overall, we can speak of a convergence of campaigning and politics itself. Although this process began well beforehand, mediatized campaigns represent the arrival of campaigning as the major activity of politics. As with campaigns, political decisions must now be presented to a public which increasingly constitutes campaigns. Thus, the selling of ideas to a public is crucial if one’s own campaigners will be drawn from the ranks of the public on a larger scale than before. Political action may need to become more palatable to a wider variety of political understandings. There may be less room for risky or headstrong political action without marketing it to the broader public, while at the same time less room for compromise.

5 The mediatized campaign: coherence in the research agenda

There has been significant work done already looking at campaigning in the 4^th phase. Some studies show significant variance in the adoption of mediatized campaigning practices within a single country or campaign, which has been the case in Europe, where younger democracies emphasize Facebook more than other mediums than more established democracies (Lilleker et al. 2015). In Norway, some candidates focused more than others on communication via social media (Gunn and Skogerbo 2013). In Germany, instances of online citizen-initiated campaigning are beginning to show (Copeland and Römmele 2014). In other situations mediatized campaigning tactics have been adopted on a larger scale, such as in the USA, where the Obama campaigns in 2008 and 2012 leveraged media, alternately using media institutions and direct contact with citizens, integrated citizens into the campaign, and employed strategies driven by data (Issenberg 2012; Kreiss 2012; Bimber 2014; Hersh 2015). The elements of rapid responses and emotional devotion overriding rational appeals in a highly partisan political situation were effective for Donald Trump in 2016. The fragmentation of the public via a high choice polarized media environment as described by Prior (2007) were highly relevant in that case, allowing perceptions to be influenced by unsubstantiated or false truth-claims as described by Thorson (2016). Yet other cases have revealed strategies akin to running fully fledged mediatized campaigns in parallel with more traditional campaigning methods, as with Modi in India (Baishya 2015; Price 2015; Neyazi et al. 2016). Relevant also for the rate and readiness of application of 4^th phase practices are also conceptions of trust in political institutions and media organizations (see Hallin and Mancini 2004; Nielsen 2015), the levels of which will impact which practices are employed and how.

The concept of the mediatized campaign is a useful tool for understanding a shift that is taking place in campaigning but also in politics as a whole. Limits to the concept’s applicability lie in the diversity and lack of uniformity of campaign practices, rendering a wholesale application of the mediatized campaign for understanding every campaign everywhere unrealistic. There will certainly be nuances and variance between those campaigns assigned the mediatized label, but the concept itself groups a number of characteristics enabling a common foundation for further research into how mediatized campaigns differ and why. The next steps in researching mediatized campaigns are thorough empirical evaluations of contemporary campaigns paired with their contexts. Specific accounting of when and where aspects of the mediatized campaign can be found can help understand the spread of campaigning practices and their transformation, and strengthen the value of the mediatized campaign as a concept.

We have suggested that campaigning has entered a phase distinct from previous phases. We thereby suggest a research agenda around the hypothesis that there is an observable shift in the way that election campaigns happen. This shift may be centered in election campaigns in Europe and the United States, but is not exclusive to these cases. What binds the wide variety of campaigns in their various contexts is the presence of some, but not all of the characteristics that make up the ideal type. To understand the degree of homo- and heterogeneity among mediatized campaigns and what the main drivers of the shift are, identification of characteristics and their prominence across contexts is necessary. Our proposal for a starting point are the following characteristics:

• 1. Immediacy. The response time of campaigns when reacting to events or news is faster than ever before. How much of this is due to the ability to react quickly due to technological possibilities (posting a tweet takes less time than organizing a press conference), and how much is due to the expectation of citizens for immediate responses (mainstreaming of the 24-hour news cycle)?

• 2. Active appeals to personalized emotions and the transformation of trust. In how far is it necessary to gain the trust of citizens/voters who can immediately check truth claims with the wide array of immediately available online resources? Has trust become a less important characteristic for candidates? What role do individual emotional connections play in connecting citizens to campaigns?

• 3. Ease of access to active roles in the campaign and the ad-hoc campaign. What are the opportunity costs of becoming an active member of the campaign (time, resources, etc.)? How much is the campaign and its human members are part of a permanent organization such as a political party? How much of the campaign machinery remains after the election is over? How likely are those involved in one campaign to be involved in the next?

• 4. Use of quantitative data as a basis for important decisions. To what extent is quantitative data used to inform strategic decisions of the campaign?