MEDIATISATION AND REGIONAL CAMPAIGNING IN A PARTY CENTRED-SYSTEM
HOW AND WHY PARLIAMENTARY CANDIDATES SEEK VISIBILITY
Abstract
Election campaigns are central to political life as well as to the study of political communication and provides much empirical knowledge about the processes of mediatisation and mediation of politics. Most often studies focus on the campaigns featuring the national top politicians. However, most elections campaigns in Western democracies are run by party branches and candidates who rarely make the top headlines in the nationwide media, yet they are also dependent on media attention and agenda-setting to be visible and reach their voters. Relying on several data sets from studies of the Norwegian 2009 parliamentary election campaign, this study asks, first, how regional, mainly "non-celebrity politicians," obtain visibility. We seek to unravel how the media logic works on the regional and local level. Second, we ask why it is important for candidates in a party-centred proportional (PR) system to be visible. Our findings suggest that we should recognise the mediatised and multileveled character of election campaigns in order to understand how media logics work below the nationwide setting.
ELI SKOGERB0 RUNE KARLSEN
Eli Skogerb0 is Professor in Department of Media and Communication, University of Oslo; e-mail: eli.skogerbo@media.uio.no.
Rune Karlsen is Senior Research Fellow at the Institute for Social Research, Oslo; e-mail:
rune.karlsen@socialresearch.no.
rN Cfi
LO IV
d CP
rN
o
o rN
rN
15 >
A 3
a
0
z *
I
(I) 0 c
(6
LD
Introduction
Prior to parliamentary elections, parties and politicians seek and attract attention from nationwide media. Celebrity politicians fill the newspapers and are frequent guests in newscasts and talk shows on television (Van Zoonen 2000; 2005). There is an abundance of studies focusing on how politics is adapted to the media logic, and particularly on the mechanisms of mediatisation in election campaigns (see e.g. Strömbäck and Dimitrova 2006; Davis 2007; Davis 2010; Norris, LeDuc and Niemi 2010; Allern 2011; Young 2011; Aelst, Thorbj0rnsrud and Aalberg 2012; Landerer 2013), however, less knowledge about parliamentary candidates who run local and regional election campaigns and rarely make the headlines of the nationwide media. Their campaigns are carried out in public spaces, mainly but not exclusively made up of news media as well as online and social media, and as such, we would expect these candidates to adapt their campaigns to the media logic, too.
Mediation and Mediatisation
"Mediatisation" refers to the complex and interdependent relationship between the media and other social institutions. The origins of the notion is often credited to Altheide and Snow's (1979) work on "media logics" where the authors argued that the news media "formatted" the way events and messages were shaped and mediated. However, these phenomena have been discussed for decades within different disciplines. Lippmann (1922) observed the difference between mediated and personal communication of news, Adorno and Horkheimer (1997) were absorbed by the power of the mass media during the Nazi period in Germany, and Stein Rokkan (1966) pointed to the media as a channel of influence beside the numerical and corporate (Elmelund-Prœstekœr, Hopmann and N0rgaard 2011). There is no full consensus on the use of these concepts (Couldry and Hepp 2013). Recently, Altheide (2013, 226) emphasised that "mediation" "joins information technology and communication (media) formats with the time and place of activities." Often, however "mediation" refers to the simple fact that messages are conveyed through some kind of media (Strömbäck 2008; Hjarvard 2013). Mediatisation may be studied as processes that have been ongoing through human history (Finnemann 2011) £ whereas other see them as recent developments tied to the expansion of modern news media and, more recently, interactive and digital media (Hepp, Hjarvard and Lundby 2010; Hjarvard 2013).
In order to study processes of mediatisation empirically, we have delimited and operationalised the concept. First, we look at mediatisation of politics, that is, how media logic affects political processes and political outcomes. In Scandinavia,
Hernes (1978) introduced the notion of the "media-twisted society" as a description of how adapting political actions and messages to the formats and timetables of the news media were effective ways for political activists to obtain political influence. Asp (1986) used "medialisation" for techniques used by interest groups to attract attention and set the agenda for in the media as well as for political decision-makers. These early contributions incorporated theories on media power and agenda setting and nourished the strand of thought conceiving "mediatisation" as shifting political power from democratic bodies to the media and non-elected activist group. In this view, mediatisation is inherently negative and detrimental to democracy (Mazzoleni and Schulz 1999).
c >
II
UD IV
In line with others, we do not adhere to this normative view on mediatisation (Strómback 2008; Elmelund-Prastek^r, Hopmann and N0rgaard 2011; Hjarvard 2013). We regard Western democracies and politics as mediated and mediatised, ^ and mediatisation of politics as an empirically observable process of change describing the increasing interdependency between political institutions and actors, such as parties and candidates, and media institutions. By increasing interdepen-dency, we refer to the fact that parties and politicians rely on the media in order to communicate their politics, whereas the media need the parties and politicians as sources, contributors of news and entertainment. There are many and different ways of exemplifying mediatisation. Elmelund-Prastek^r et al. (2011) point to five structural indicators conducive to increasing mediatisation: weak political parties/ decline of class parties; dominance of commercial media; intense competition for media audiences; professional management of parties; and journalistic focus on horse races and not policies. All these are observable in the Norwegian setting, but they may not fully illustrate that mediatisation incorporates an institutional and constructivist approach to politics. Politics is played out inside and outside the media, and political events, such as elections and party conventions, are followed, framed, interpreted and commented on by journalists. Some politicians obtain celebrity status by position or by building up "media capital" by continuous and reproduced media appearances that may be converted into political power (Davis 2010, chs. 5-6). Most citizens experience politics mainly as mediated and mediatised events, and meet top politicians and the political parties only though television, newspapers, blogs or Facebook. Political parties and their candidates strive accordingly to be visible in and gain attention from the media, if not continuously so as an important part of their everyday political life and indeed when campaigning.
Division of Labour
Our second focus is on the mediatisation of election campaigns. Election campaigns in mediatised democracies can be conceived of as ways of managing and optimising visibility for political parties, their issues and candidates prior to Election Day and are as such particularly spectacular examples of mediatised politics. We argue that mediation and mediatisation take place not only on top-level politics but on the regional and local level, too. Political practices and institutions have increasingly been adapted to the practices of journalism and media institutions. During election campaigns, such practices are observable in the professionalisation of the party organisations, media training of politicians and their advisors, increasing media competence and appearances of politicians and political candidates, and increased use of digital and social media for political purposes. They are constant ingredients of political activity, yet intensified during campaigning periods. In party-centred systems, there is a division of labour between the central party organisation and the party leadership and the local and regional party branches and the local and regional candidates (Karlsen and Skogerb0 2013).
Time and Space
Third, we look at campaigning locally and regionally. The party leaders naturally attract most attention from the nationwide media. These media stage the contest for power of the government, and there is little space for the regional candidates.
Regional candidates have to seek alternative communicative spaces for attracting attention to their candidacies and their parties' politics. We expect that traditional and online local and regional newspapers, radio and television, as well as social media, blogs and websites make up the communicative spaces for these candidates. Both individual and structural factors may influence whether individual candidates succeed or fail in their attempts to attract attention to their campaigns. Individual factors may include candidates' personal communication skills, strategies, and their position on the party list. Structural factors may refer to the parties' historical position in the constituency, the geographical and demographical characteristics of the constituency as well as the local and regional media structure. As a rule, regional candidates cannot draw on a nationwide celebrity status as they are less well known, less exposed on national television, do not hold high positions in the party and accordingly draw less attention to their candidacies. However, they may have accumulated local "media capital" (Davis 2010). Parliamentary candidates are likely to be well versed in mediatised politics, as they often are experienced politicians from local and regional government and parties. They know the local media structure and may take advantage of the competition between the different news media for breaking news, and use online and social media as alternative routes onto the agenda of the news media. Local media structures may be monopolistic or pluralistic depending on the amount of media and communication channels available in the constituency. A pluralistic structure allows for more competition between the media and provides more space for the individual candidates. As the media structure varies between the constituencies in terms of number, popularity, readership and reach of newspapers, broadcasters and online media, it may affect how and to what extent candidates achieve attention.
Local and regional media operate within spatial and temporal frames and have editorial priorities that influence the way they follow and report politics. The main characteristic of local journalism is the localisation of news and stories (Franklin 2006; Mathisen 2010). In the same vein, elections, constituencies, voters, and political candidacies are defined by time and space. Electoral constituencies are rarely identical with the areas covered by either traditional regional and local news media or online media and constituencies do not have identical media landscapes. For political candidates, this means that they may have to localise their messages, too, but still operate within the boundaries of a central campaign. Parties and candidates are likely to adapt, transform and communicate their messages to conform to the journalistic priorities and agendas of local and regional media. Local party branches £ and candidates may front local issues or conflicts that highlight differences and views in the campaign. In the following, we seek to untangle how parliamentary ^ candidates run their campaigns in a media landscape dominated by local journalism.
The Norwegian Setting - Politics and the Media
Q
The study is carried out in Norway and includes candidates who ran for the 2009 national election. Politically, Norway is a stable democracy with a parliamentary government, a multiparty system and well-organised membership parties.
3
a
c >
Although the political parties over time have been weakened by declining membership and reduced party identification, the organisations have remained strong (Heidar and Saglie 2003). The parliamentary constituencies are made up by the
il 00
19 counties, large administrative and geographical units consisting of many different rural and urban municipalities. All major parties put up lists and campaign CT> for representation in each constituency. Political sociologist Stein Rokkan (1967) ^ described the Norwegian political landscape as one of cross-cutting cleavages, and explained the formation of several parties as outcomes of their placement in the conflict structure. Over time, the cleavage structures have been weakened as has the support for the smaller parties. The Left - Right cleavage has remained significant and was in 2009 one of the major conflicts (Aardal 2011a). The party system can be described in terms of how the parties are placed on this continuum (see Table 1).
Table 1: The Norwegian Party System: The Seven Parties Represented in
Parliament 2005-2013
Socialist Left	Labour Party	Centre Party	Christian	Liberal Party	Conservative	Progress
Party (SV)	(Ap)	(Sp)	People's	(V)	Party	Party
			Party		(H)	(Frp)
			(Krf)			
The electoral system consists of direct elections and proportional representation in multi-seat constituencies. The party system is, as is common in Europe, party-centred, as opposed to candidate-centred systems (e.g. USA). The central party organisations draw up campaign strategies that are guiding for the local and regional campaigns run by the party branches. The political parties dominate the nomination of candidates for parliament. The nomination processes are decentralised and the nomination of candidates in ranked order on party lists are made by representative conventions organised by the constituency branches of the parties (Narud, Pedersen and Valen 2002). Once the parties have put together the lists, the voters have formal but in practice no possibility of influencing the ranking order (Aardal 2011b). Consequently, campaigning is directed at mobilising voters for the party, not primarily for candidates. However, in situations where parties compete for mandates, the focus on candidates is likely to increase. Norwegian elections over the past decades have seen increasing shares of volatile, non-voters and undecided voters, meaning that a substantial share may be mobilised to vote or swing their vote until the last days and minutes of the campaign.
Within each constituency, the candidates compete for a fixed set of seats. As most constituencies are geographically rather large and have more or less clear sub-regions, some parties, typically the large ones may divide the counties into several local campaigning grounds. As candidates normally live in localities within the constituencies, they also belong to different coverage areas of the local media. For the candidates, the existence of local newspapers and radio stations that cover their place of residence may be of particular importance for their ability to attract journalistic interest and use the local media as communicative platforms for promoting their candidacies.
The four selected constituencies provide somewhat different structural communicative conditions for the election campaigns. Hallin and Mancini (2004, 11) regarded the Norwegian media system as a typical example of the Democratic-Cor-poratist Model as the state is active in designing public media policy including public service broadcasting and press subsidies and simultaneously there are strong
legal and institutional barriers against interference in the editorial freedom. There is a historical coexistence of commercial media and political parallelism, meaning that many media, typically newspapers, have a past as being party press or tied to civic organisations. Over the past decades, the media dealigned themselves from the political parties, yet there are remnants of the system. Norway, as other countries, also has several distinct characteristics indicating that the Democratic Corporatist Model is weakened or perhaps never fitted that well (Herkman 2009; Humphreys 2012; 0stbye and Aalberg 2008).
Important for understanding the roles of the media in regional campaigns are two structural traits: first, the ubiquity of media all over the country, traditional and online. Approximately 225 newspapers, mostly local, many only issued on print a few days a week, cover the entire country. Two popular tabloid newspapers have nationwide coverage and a few regional newspapers cover larger geographical areas (H0st 2013). The number of local newspapers has increased over the past decades, and local journalism occupies more space in the regional newspapers. Newspaper readership is high, although declining on print. Broadcasting and newspapers are significant and important information sources for voters (Karlsen 2011). The public broadcaster, NRK, offer national and regional radio and television as well as online services in all counties. The largest private TV channel, TV2, provides online services and nationwide broadcasts on several channels, too. Broadband services and social media are widely used, although the share of users varies largely between different services such as e.g. Facebook, Twitter and YouTube.
Research Questions
The core question running through our analyses focuses these issues: How do regional, mainly "non-celebrity politicians," obtain visibility? They have to compete both with other candidates from their own constituency, with other sources and with the news agenda in general.
Second, we ask why it is important for candidates in a party-centred proportional y (PR) system to be visible. A party-centred PR system where the ranking order on the party lists is fixed, is conducive to a party-centred campaign-style with a rather limited role for candidate-centred campaigning (Karlsen and Narud 2013). Candidates who cannot use the campaign to change the ranking order of the party list and thereby improve their own chances to be elected, logically would not need to be visible and mobilise voters to support their own candidacies. Such reasoning unfortunately removes the fact that politics and election campaigns are mediatised processes, they take place on mediated arenas and follows the media logic. Increased personalisation is one of the aspects of this. With increasing personalisation, voters recognise parties not only by ideology and issues but also by candidates. Parties personalise their images by fronting their party leaders and top candidates and the voters recognise and relate to celebrity politicians in the party leaderships as well as to individual politicians running in their home constituencies (Van Zoonen and Holte-Bacha 2000). Voters experience, learn about and make choices about politics from many sources, however, their experiences with and conceptions of parties and politicians will more often originate in images and representations in the media than in personal meetings. We do not argue that there is a simplistic in-qq fluence from media experiences to voting, we simply point to the fact that citizens,
0
c ¥
I
c >
as individuals and members of social collectives, usually experience, discuss, assess and act politically by way of their experience with politics as mediated events. Election campaigns cannot be imagined without mediation, they are designed to be carried out in public spaces and on all kinds of media platforms. Although not all campaign activities happen through or in the media, mediation is essential to reach large groups of voters. This is why we hypothesise that visibility for individual candidates is essential in campaigning in mediatised democracies.
Data and Research Methods
We analyse three unique sets of data, one quantitative and two qualitative, collected in the aftermath of the Norwegian 2009 parliamentary campaign. The first is the 2009 Norwegian Candidate Survey that was sent to all candidates running for election for the seven major parties. The original sample was 1972. The response rate was 52 percent, which left 1015 candidates. All parties and top and lower placed candidates were about equally represented. The survey data were analysed by means of computerised statistical measures, including descriptive as well as analytical statistics. We have run analyses controlling for socio-demographic background variables as well as variables concerning campaigning and mediatisation. The survey complemented, contrasted and triangulated our findings from the qualitative data. By triangulating, we seek to increase the relevance, validity and reliability of the empirical findings, as well as strengthening the hypotheses that are generated from the study.
The second data set consisted of qualitative interviews with 29 top candidates from the seven parties. This included candidates placed as no. 1 or no. 2 on the party lists in four constituencies; Buskerud (pop. 269,000) in east Norway; Ro-galand (pop. 452,000) on the south-west coast; Sogn and Fjordane (pop. 108,000), situated in the west; and Troms (pop. 160,000) in north Norway. The four case constituencies belong to regions where the traditional cleavage structure was widely different, meaning that the parties had different historical starting points. Third, we included interviews with nine editors in local and regional media in the same four constituencies. These focused on the editorial and journalistic priorities of the local and regional media in covering the general election in 2009. The two sets can be classified as "elite interviews" and were analysed systematically with the intention of adding information and meaning (Kvale and Brinkmann 2009). Taken together, the three data sets produce unique, statistical, reliable and to some extent generalisable findings as well as exemplary and illustrative insights into the campaigning efforts and media strategies of regional candidates. Although the data were collected in Norway and vary in generalisability, our findings provide insights into and generate hypotheses about the priorities and strategies of rank-and-file politicians and local journalism that have relevance in other national settings, too.
Findings
In the candidate survey, we investigated the communicative platforms of the candidates by asking the following question: On a scale from 1 to 5 where 1 indicates unimportant and 5 indicates very important, how important were the following media for you in your campaign effort? Table 2 reports the results.
Table 2: The Importance of Media for Candidate Campaign Communication 2009 ("On a scale from 1 to 5, where 1 indicates unimportant and 5 indicate very important, how important were the following media for you in your campaign effort?"; in percent and arithmetic mean; source: Candidate Survey 2009, N=1015)
	Not important			Very important			
	1	2	3	4	5	Mean	N
Local newspapers	6	4	12	26	53	4.16	973
Regional newspapers	16	8	16	27	34	3.56	963
Local radio	23	13	21	22	21	3.06	965
Regional television	32	11	18	21	19	2.84	965
National newspapers	42	13	14	15	17	2.54	966
Social Internet media	36	13	22	18	11	2.54	947
Nation-wide television	46	10	11	12	21	2.53	965
Nation-wide radio	41	13	17	15	14	2.46	958
Personal website	47	14	19	12	7	2.17	932
Table 2 shows that there was a distinct hierarchy in the candidates' media preferences. Local newspapers were the most important communication channels. Regional newspapers and local radio followed next. Nationwide television was ranked as third from the bottom of the nine communication channels. 46 percent regarded nationwide television unimportant for their campaigning whereas 47 percent considered personal websites of no value. We found few differences between the candidates concerning their background, such as age, gender, and party affiliation. Age was relevant for the assessment of social media, and the candidates representing parties to the right on the political spectrum valued television somewhat more than other media but overall similarities were more striking than differences.
Concerning regional differences, nationwide TV and radio were very important only for the top candidates campaigning in the capital, Oslo. Several of the top candidates in the capital were party leaders, so this finding illustrate that the party elites and the regional candidates operate on different media arenas. Further, the candidates regarded online and social media as less important for their campaigning than traditional media. Only 11 percent of the candidates who answered the question regarded online media as very important. The survey data also showed that position on the party lists influenced the assessment of the media: the top candidates assessed traditional media as more important than candidates placed lower on the lists.
The qualitative interviews with individual candidates supported the findings in table 2. The interviewees unanimously listed local media, in particular the local and regional newspapers, as fundamental to their campaigns. They also provided reasons why they put the local media on top of the hierarchy. One candidate, a high-profiled and well-known MP who was not part of the party leadership, remarked, "There are limitations on the number of Labour MPs who have access to the nationwide media" (Candidate interview 7\12\2009), while another detailed the importance of local media:
£ 3
a
0
£ *
1 *
tf) 0 c
B
fN 00
The regional and local media are extremely important, and increasingly so in every election campaign. For us who are not in the party leadership they are the	CO
most essential communication channels. It is there we present our candidacies	^^
and the party programme (Candidate interview 01\12\09).
Knowing that the focus in the nationwide media was on the party leadership, the candidates considered the regional and local media as their main arenas for displaying their views and face to the voters. Contrary to the many premonitions of the demise of the local media (Franklin 2006), some candidates argued that the local and regional media were becoming more important for the candidates over the years: "The local and regional media have improved, are more critical, and follow us in ways they formerly did not" (Candidate interview 1\12\09).
Further, candidates confirmed the division of labour and the different conditions for access inside and outside campaign periods:
During election campaign, the main media focus is on the party leader and the party elite. In this period, I concentrated on the local media, although I did try to get through to the national media once or twice. Between elections, it is "both please," - it is easier to get access in the national media. As a representative for the party, I must have a strategy to show my face in the newspapers and on TV (Candidate interview 22\9\09).
Visibility
Having a "strategy to show my face" was the essence of the campaign efforts in the local constituencies, if we are to believe the candidates:
The election campaign is very much about visibility. People may agree or disagree with your message but being visible has a value in itself. Accordingly, the [local and regional] newspapers and radio programmes are very important (Candidate interview 02\11\09).
The candidates emphasised that their main objective in the local campaigns was to be visible. For these candidates the local and regional newspapers provided the largest, most attractive and most efficient arena for reaching the voters. The local newspapers were "read by everybody" as one candidate claimed. Another argued that it was most "effective" to be in the local media. Others pointed out that getting coverage in local and regional newspapers was more valuable than being quoted in a major quality newspaper, while some measured the success of the campaign in front pages:
The regional media are important for the home market and for issues that are regional and local. The goal in the election campaign was to get as many articles and front pages as possible in the leading regional newspaper. If we got three or four front pages, it was a successful election campaign (Candidate interview 12\11\09).
These viewpoints were common for the candidates, regardless of party, position in the party and place on the lists. The editors of local and regional media confirmed that their media were important stages:
Election campaigns are about getting known, make people know who you are. ... The local newspaper is a very important arena in this (Interview, editor-in-chief, 08\12\09).
Put simply, the main differences between the national and local campaigns seem to be the stages on which the politicians strove to be visible. For the candidates, the ability and need to obtain visibility varied not only with their personal capacity for attracting journalistic attention, but also with their position on the lists and the "security" of their mandates (table 1).
Table 3: Proportion of Candidates who were Regularly Interviewed in Local and National Media during the Parliamentary Election Campaign by Perception of Winning Chances 2009 (in percent; source: Candidate Survey 2009; N=1015)
	Local newspapers	National newspapers	Local TV	National TV	(N)
Could not win	60	8	27	6	656
Could hardly win	85	17	62	16	99
It was an open race	93	44	78	36	58
I could hardly lose	88	30	66	29	54
I could not lose	93	70	89	59	38
Regularly interviewed = used more than 1 hour every week on interviews.
Position and Party
Table 2 shows differences between candidates with different perception of their chances to win a mandate and their access to the media. Two findings are very clear: the local newspapers were important media arenas for all candidates. Even the group with no winning chances were regularly interviewed locally whereas they had insignificant access to national media. Further, the candidates in the best positions, those who could not lose their seat, together with those who had to fight hard to win a mandate, won the attention of the media. This is hardly surprising, given the media logic: the secure positions on the lists were likely to be occupied by local "celebrity politicians" who can count on newsworthiness by their sheer presence, or following Davis (2010), by their accumulated "media capital." The interviews shed further light on these findings. Several of the candidates with secure positions on the party lists, did not, according to themselves and the editors, need ^ to do very much to get attention. One editor confirmed this:
Q	We have one of the Labour Party veterans here. He is not very visible. Still, he
is the classic politician, the one who is always present at party meetings and always works for local projects and gets credit for it, too. He is probably the politician that ordinary people vote for although he does not appear in the news very much. However, he has his footing in the working class, if anything like that still exists. They are his people. When he comes up with something, we put him on because we know he is good and he knows what is important (Interview,
0 _____________________________
editor-in-chief 04\12\2009).
The editor here describes a "local working-class hero," a politician whose views and stories will interest the local readers, thereby fitting the demands of local journalism. Equally expected is that the political struggles for power and positions qq between candidates fighting for insecure seats attract more media attention than
candidates who do not stand a chance. The editor pointed to the difference between the candidate who held a secure top position for the Labour Party and the top LH candidate from the Centre Party that had not taken a mandate in the constituency ^ for decades. "He knew that he would win whereas the Centre Party candidate had to do something exceptional to get a seat in Parliament" (Interview, editor-in-chief 04\12\09). By pointing to the differences in their positions, the holder of a safe seat versus the candidate running in an open race, the editor also drew attention to the substantial differences between campaigning for a large and a small party. Driving a successful campaign from the underdog's position required quite another effort for entering the news. When asked why one particular candidate from a small party attracted so much attention, the editor pointed to her media capital:
She was extremely energetic and contacted us on SMS, telephone and everything else. She succeeded in breaking the barriers. Moreover, she was trained in getting what she wanted. She continuously announced newsworthy issues and events that we simply had to report. She was always present and often together with someone from the top party leadership or the Government, promising to solve or support a local issue. She arranged meetings with local party leaders and if we could not send a photographer, she even fixed pictures. She was very keen on winning a seat in Parliament and she succeeded, too (Interview, editor-in-chief 04\12\2009).
The editor described a candidate well versed in mediatised politics. When interviewed, she confirmed that the campaigning efforts were part of a systematic communication strategy aimed at attracting local media attention and mobilising local support:
We systematically worked the media and we worked closely with our local party representatives, too. When I travelled around the constituency, I did so together with the local representatives from the places I visited, e.g. the mayor, a group leader, the leader of the local party branch. They were always involved in the campaign and the issues that we fronted (Candidate interview 10\11\09).
Localising the Campaign
The interviews illustrated the mechanisms of mediatised politics in practice: candidates played not only the media logic but also the local media structure. Focusing local issues, cooperating with the local party representatives, showing that also the "big issues" had a local angle, were among the techniques employed to attract attention from local and regional media. In areas with a multitude of different local media, they adapted their messages to ultra-local, local and regional media, as well as observing differences between online and traditional publishing. As expected, differences in the media structure opened for different and individualised media strategies. The candidates in one constituency praised the ubiquity of local media and described a competitive media situation that benefitted the candidates. They expressed more control and autonomy over their communication strategies than candidates in constituencies where the media structure was centralised or nearly monopolised. In a situation with local news competition, the candidates experienced that their value as sources was higher than in a near-monopolised situation where the dominant medium could choose from an abundance of sources. This finding
supports the argument that media concentration lead to less diversity and more difficult access for views and individuals (Baker 2007), and, equally important, that diversity of media channels may also mean diversity of voices (Karppinen 2012).
There were some differences between the candidates concerning how they localised issues. As may be expected in a party-centred system, candidates from the small parties with few secure seats focused on specific local issues, sometimes promising financial support for local projects. Candidates from the large parties seemed to adhere to their parties' central campaign strategies arguing that their main issues, e.g. securing kindergartens or good schools everywhere, did not need specific localisation as they were important and relevant for their local voters, too (Karlsen and Skogerb0 2013).
The editors supported that localising issues and messages were in line with their editorial philosophies. In order to be journalistically attractive, the candidates had to be relevant for the local audiences:
For us it was important to cover the election campaign by focusing the local issues. The local campaigns are often overshadowed by the national election campaign. ... Many candidates were not prepared for anchoring their messages locally. ... We had a criterion for coverage of e.g. a Minister's visit to the constituency that he or she had to contribute to a local issue (Interview, editor-in-chief 14\10\2009).
The attempts of local party branches and candidates at enhancing the visibility of their campaigns by inviting national celebrity politicians illustrate the dynamics between the national and the local campaigns and between the media logics of the nationwide and local media. Hosting a visit from the top party leadership shows local citizens that local candidates and votes are important to the central party. It may attract attention from the nationwide media to local issues and the region. Yet, if we believe the editors, it may not enhance the candidates' local news value, as they step out of the local media logic and into the national media arena. Commu-y nicating politics thus not only demands that candidates have considerable insight into processes of mediatisation, the local candidates need substantial skills for translating big politics into issues that reflect on voters' everyday life in different geographical and medial settings, too.
Still, the local and regional media were not the only platforms. Several candidates across party lines emphasised personal contact with voters through home visits and other forms of face-to-face communication as important and believed that such activities would increase in coming elections. In contrast, political meetings and debates tended to be deemed as unimportant for reaching new voters. Among the arenas that a number of the candidates used, were social and online media.
Online and Social Media Were Add-ons
Q
Online media and social media were indeed gaining importance for the candidates. Social media, e.g. Facebook, were considered more important than the more static personal websites. In 2009, quite a few candidates did not employ social media in their campaigning effort. Those who used them regarded social media highly and for the youngest candidates they were the second most important channels. qq Both the survey and the interviews indicated that online and social media com-
c >
plemented and did not replace other media exposure (Figure 2). Candidates who assessed social media as important valued all other media, too. The underlying explanation may be that the already "media savvy" candidates benefitted most ^ from online media.
Figure 1: Social Media are Add-ons: The Importance of Different Media for Campaign Communication for Candidates Who Say that Social Media Are Important, and Candidates Who Indicate that Social Media Are not Important
Social media important: 4 and 5 on the scale from 1-5. N: Important=268 Not important=667-676.
All parties encouraged their candidates to be present online. The Labour Party, for instance, sought to lower the threshold for online participation in order "to be present everywhere," to quote one Labour candidate. None of the interviewees was unaware of online and social media. Everyone had a personal website provided for them by the party and all used electronic media for information and communication. Not all candidates had a personal online profile, but all were aware that many voters, in particular the young, expected them to. Some were active on several platforms, many preferred Facebook, and others did not prioritise social media at all. The following quotation is typical of the latter group:
I am not very active on Twitter. I started but fell out again. I wanted to be there myself and not have other people write for me. ... However, I did not manage (Candidate interview 29\01\10).
This candidate was a party leader, member of the Cabinet and enjoyed the corresponding celebrity, media capital and access to all media while running for election in her home constituency. Still, she saw it as a problem that she failed to be personally active on social media. Her example illustrates the strong claim for authenticity that prevailed among the candidates. With few exceptions, they all
regarded personal authorship on the web as essential, thereby also emphasising that social media presence was part of personal image-building (Enli and Skogerb0 2013).
The candidates also distinguished clearly between different online media, using blogs, Facebook and Twitter for different purposes:
My goal is to blog a couple of times a week. I want to get my political message out on the web so that those surfing the net can pick it up and discuss it. I want to create two-way communication with the voters concerning concrete political issues. I use Twitter primarily to draw attention to my blog, and Facebook, too, but Facebook lies in the intersection between my private and public roles (Candidate interview 01\12\09).
Only one of the interviewed candidates had the Internet as her main platform, and she used online presence as a substitute for lacking access to other media. As a candidate without winning chances, she had few opportunities for attracting attention from the traditional news media.
Importance of the Nationwide and the Local Campaigns
As much as the candidates emphasised the importance of local and regional media in the regional campaigns, a different pattern emerged when they assessed the salience of different media in the national election campaign (Figure 3).
Figure 2: Important Media in the Election Campaign: Percentage of the Candidates
Who Placed the Medium as Most Important, Second Most Important and Third
Most Important (Source: Candidate Survey 2009; N=1015)
00 00
Whereas only candidates campaigning in the capital identified nationwide television and newspapers as important for their own campaigns, eight out of ten candidates identified nationwide television channels as the most influential in the national campaign. Over 70 percent regarded the public broadcaster NRK most
significant. These figures point to the perceived importance of the media in securing visibility for the party elites, highlighting the parties' core issues, and setting CT> the agenda of the nationwide election campaign. The interviews confirmed that ^ the candidates regarded nationwide TV and newspapers as essential for setting the agenda of the national campaign. Some even questioned whether the local campaigns made a difference but the most common assessment was that local campaigns might have a limited local effect on the outcome, but not change the general trends of the campaign:
Those who say that the local election campaigns are decisive are simply wrong. They may be important in order to draw one or two percent of the voters, but the 12, 14, 15 percent, the so-called core voters are recruited nationally. It can clearly be seen from the fact that the Conservative Party had 10 percent on the opinion polls when the election campaign started and ended up with 17 percent of the votes cast. It was not the good job done in the counties that caused that, it was caused by the formidable job done by the party leader on TV (Candidate interview 12\11\2009).
The editors shared their views, as the following, typical, quotation illustrates:
The campaign is dominated by the nationwide agenda and we have marginal influence. We introduce and make known the local candidates to the local voters but we do not have much influence on the nationwide agenda or the outcome of the election on the national level (Interview editor-in-chief 6\11\2009).
Discussion
Our study opened by asking how candidates running for election in their home constituencies obtain visibility, taking as a starting hypothesis that election campaigning in the regional and local constituencies are mediated and mediatised processes, just as the national campaigns. Our findings show, clearly and unequivocally, that this is the case. The media channels are different; the centrality of mediatised politics is not.
Our findings confirmed, first, the interdependency between local media and local parties and candidates. Elections were important news that "had to be covered." Making candidates and political alternatives known to the voters is prioritised by local journalism. The journalistic newsworthiness of candidates, parties and issues were measured by the relevance for local audiences and their adaptability to local news criteria. Campaign strategies, media capital, and the party organisation were resources drawn upon by the candidates. Their journalistic attractiveness varied with media competence, status, position on the party lists and chances of winning a seat. Local editors published what they saw relevant for their local audiences and the local media logics were well known by the candidates. In a media economy where fragmentation of attention and diversification of media products are strong trends, increased localisation of parliamentary campaigns seem to be an emerging hypothesis. Some candidates indicated that they tailored their media performances to different local media. Such strategies might mean that they also fragmented and tailored their political arguments; however, our findings also indicate that central campaign strategies prevailed, preventing a clear conclusion. Our interviews suggested that the candidates perceived a dilemma between being responsive to local
demands during the campaign and accountability as representatives for a large party that was likely to be in government after the election.
There were differences shaped by the media landscapes. Within local media, the candidates had ample room for driving their own campaign. Interestingly, whereas the candidates in the survey described their communicative spaces as hierarchical and ranked, the interviews pointed to benefits of campaigning in constituencies where the media landscape was diverse and different media covered different parts of the constituencies. For these candidates the media diversity worked as a network of media spaces that offered many outlets. Small ultra-local newspapers, online and print editions, party web sites and social network sites like Facebook and Twitter were described as complementary if not of equal importance to the nationwide media.
This description challenges the media hierarchy that emerged when candidates ranked channels according to perceived importance, and may point towards a shift in the power play between the media and politicians. The more media accessible to them, the more spaces in principle will be open for meetings between politicians and their voters. This observation does not necessarily rock the fundamentals of the mediatisation framework but it needs to be re-described in a situation of "communicative abundance" (Karppinen 2012). Our interviews as well as other studies of how politicians use social media and other channels to avoid the gatekeepers and market their views and images suggest that this is a likely interpretation (Enli 2007; Skovsgaard and Van Dalen 2013).
Our second main question was why visibility is essential for candidates in party-centred systems. Neither local candidates nor local media claimed to influence the agenda of the nationwide election campaign, suggesting a contradiction between the importance that the candidates attributed to being visible and their assessment of their own campaign efforts to have little bearing on the nationwide campaign. Instead, they pointed to the party leaders' performances on television to explain success or failure of the party on Election Day. So, why did the candidates put so much effort into their regional campaigns if they actually believed that most of their efforts did not make a difference? We suggest that the explanation is that they conceive of campaign communication as multi-layered and hierarchical communication processes, following the same organisation and division of labour as the ^ political system and the media structure. Local campaigns matter locally but do not outweigh the nationwide campaigns. Yet, were they not to campaign locally, the candidates risk disappearing from the voters' eyes and thereby lose their votes. If we regard their assessment of the different levels of campaigning from this angle, the contradiction disappears.
^ Conclusion
Our findings provide new insights into the complexity of mediatised politics. First, even in party-centred systems parliamentary election campaigns are multi-level, mediatised political communication processes. Candidates obtain visibility through a network of mediated spaces, locally or nationwide. Second, personal visibility is important also in settings where the candidates cannot change their own winning chances. Large numbers of voters decide if and what party to vote for shortly before Election Day, and they know parties mainly through the media
0 £
>
(6
o
^ appearances of their candidates. Visibility for candidates means visibility for parties.
Acknowledgement
-T—I
The research projects reported in this article were funded by the Norwegian q^ Research Council and the University of Oslo.
References:
Aardal, Bernt. 2011a. Detpolitiskelandskap: en studie avstortingsvalget 2009. Oslo: Cappelen Damm. Aardal, Bernt. 2011b. The Norwegian Electoral System and its Political Consequences. World
Political Science Review 7, 1, 1-31. Adorno, Theodor W. and Max Horkheimer. 1997. Dialectic of Enlightenment. London: Verso. Aelst, Peter van, Kjersti Thorbjornsrud, and Toril Aalberg. 2012. The Political Information
Environment during Election Campaigns. In T. Aalberg and J. Curran (eds.), How Media Inform Democracy: A Comparative Approach, 50-63. London: Routledge. Allern, Sigurd. 2011. Fjernsynsvalgkampen. Program, deltakere og maktkamp 1961-2009. Oslo: Pax. Altheide, David L. 2013. Media Logic, Social Control, and Fear. Communication Theory 23, 3, 223-38.
Altheide, David L. and Robert P. Snow. 1979. Media Logic. London: Sage. Asp, Kent. 1986. Mäktiga massmedier: studieripolitiskopinionsbildning. Stockholm: Akademilitteratur.
Baker, C. Edwin. 2007. Media Concentration and Democracy. Why Ownership Matters. New York:
Cambridge University Press. Couldry, Nick and Andreas Hepp. 2013. Conceptualizing Mediatization: Contexts, Traditions,
Arguments. Communication Theory 23, 3, 191-202. Davis, Aeron. 2007. The Mediation of Power: A Critical Introduction. London: Routledge. Davis, Aeron. 2010. Political Communication and Social Theory. Hoboken: Taylor & Francis. Elmelund-Pr»stek»r, Christian, David Nicolas Hopmann, and Asbjorn Sonne Norgaard. 2011. Does Mediatization Change MP-Media Interaction and MP Attitudes toward the Media? Evidence from a Longitudinal Study of Danish MPs. The International Journal of Press/Politics 16, 3, 382-403.
Enli, Gunn Sara. 2007. Gate-keeping in the New Media Age: A Case Study of the Selection of
Textmessages In a Current Affairs Programme. Javnost - The Public 14, 2, 47-62. Enli, Gunn Sara and Eli Skogerbo. 2013. Personalized Campaigns in Party-centred Politics.
Information, Communication & Society 16, 5, 757-74. Finnemann, Niels Ole. 2011. Mediatization Theory and Digital Media. Communications 36, 1, 67-89. Franklin, Bob. 2006. Local Journalism and Local Media: Making the Local News. London, New York:
Routledge, Kindle edition. Heidar, Knut and Jo Saglie. 2003. Predestined Parties? Organizational Change in Norwegian
Political Parties. Party Politics 9, 2, 219-39. Hepp, Andreas, Stig Hjarvard, and Knut Lundby. 2010. Mediatization - Empirical Perspectives: An
Introduction to a Special Issue. Communications 35, 3, 223-28. Herkman, Juha. 2009. The Structural Transformation of the Democratic Corporatist Model: The
Case of Finland. Javnost - The Public 16, 4, 73-90. Hernes, Gudmund. 1978. Det mediavridde samfunn. In Hernes. G. (ed.), Forhandlingsokonomi og
blandingsadministrasjon, 181-95. Bergen: Universitetsforlaget. Hjarvard, Stig. 2013. The Mediatization of Culture and Society. Milton Park: Routledge. Humphreys, Peter. 2012. A Political Scientist's Contribution to to the Comparative Study of Media Systems in Europe: A Response to Hallin and Mancini. In N. Just and M. Puppis (eds.), Trends in Communication Policy Reseach: New Theories, Methods and Subjects, 157-176. Bristol: Intellect. Host, Sigurd. 2013. Avisäret 2012. Volda: Moreforsking.
Karlsen, Rune. 2011. Velgernes valgkamp. In B. Aardal (ed.), Det politiske landskap. En studie av
stortingsvalget2009, 41-46. Oslo: Cappelen Damm. Karlsen, Rune and Hanne Marthe Narud. 2013. Nominations, Campaigning and Representation. How the Secret Garden of Politics Determines the Style of Campaigning and Roles of
Representation. In Esaiasson, P. and H. M. Narud (eds.), Between-Election Democracy, 77-101. Colchester: ECPR Press.
Karlsen, Rune and Eli Skogerbo. 2013. Candidate Campaigning in Parliamentary Systems: Individualized vs. Localized Campaigning. Party Politics, published online on June 7.
Karppinen, Kari. 2012. Rethinking Media Pluralism. New York: Fordham University Press.
Kvale, Steinar and Svend Brinkmann. 2009. Interviews. Learning the craft of Qualitative Research Interviewing. Los Angeles: Sage.
Landerer, Nino. 2013. Rethinking the Logics: A Conceptual Framework for the Mediatization of Politics. Communication Theory 23, 3, 239-58.
Lippmann, Walter. 1922. PublicOpinion. New York: Harcourt, Brace.
Mathisen, Birgit Roe. 2010. Lokaljournalistikk: Blindpatriotisme ellerkritiskkorrektiv? Kristiansand: IJ-forlaget.
Mazzoleni, Gianpietro and Winfried Schulz. 1999. "Mediatization" of Politics: A Challenge for Democracy? Political Communication 16, 3, 247-61.
Narud, Hanne Marthe, Mogens N. Pedersen, and Henry Valen. 2002. Nominations for Parliament Elections: The Case of Denmark, Finland, Iceland and Norway. Odense: Odense University Press.
Norris, Pippa, Lawrence LeDuc, and Richard G. Niemi. 2010. Comparing Democracies 3: Elections and Voting in the 21st century. Los Angeles: Sage Publications.
Rokkan, Stein. 1966. Norway: Numerical Democracy and Corporate Pluralism. In R.A. Dahl (ed.) Political Opposition in Western Democracies. New Haven: Yale University Press.
Rokkan, Stein. 1967. Geography, Religion and Social Class: Crosscutting Cleavages in Norwegian Politics. In S.M. Lipset and S. Rokkan (eds.) Party Systems and Voter Alignments. Cross-National Perspectives. New York: The Free Press.
Skovsgaard, Morten and Arjen Van Dalen. 2013. Dodging the Gatekeepers? Social Media in the Campaign Mix during the 2011 Danish Elections. Information, Communication & Society 16, 5, 737-756.
Strömbäck, Jesper. 2008. Four Phases of Mediatization: An Analysis of the Mediatization of Politics. International Journal of Press-Politics 13, 3, 228-46.
Strömbäck, Jesper and Daniela V. Dimitrova. 2006. Political and Media Systems Matter. A Comparison of Election News Coverage in Sweden and the United States. The Harvard International Journal of Press/Politics 11, 4, 131-47.
Van Zoonen, Liesbet and Christina Holtz-Bacha. 2000. Personalisation in Dutch and German Politics: The Case of Talk Show. Javnost - The Public 7, 2, 45-56.
ÜVan Zoonen, Liesbet. 2000. Popular Culture as Political Communication - An Introduction. Javnost - The Public 7, 2, 5-17.
Van Zoonen, Liesbet 2005. Entertaining the Citizen: When Politics and Popular Culture Converge.
^	Lanham, Md.: Rowman & Littlefield.
Young, Sally Ann. 2011. How Australia Decides: Election Reporting and the Media. Cambridge: Cambridge University Press.
0stbye, Helge and Toril Aalberg. 2008. Media and Politics in Norway. In J. Strömbäck, M. 0rsten and T. Aalberg (eds.), Communicating Politics: Political Communication in the Nordic Countries, 83-102. Göteborg: Nordicom.
fN Ch