Messy interviews: changing conditions for politicians’ visibility on the web

Abstract

This article provides an updated analysis relating to John B. Thompson’s argument about political visibility and fragility. It does so in light of recent years’ development of communication technologies and the proliferation of nonbroadcasting media organizations producing TV. Instances of a new mediated encounter for politicians is analyzed in detail – the live web interview – produced and streamed by two Swedish tabloids during election campaigning 2014. It is argued that the live web interview is not yet a recognizable ‘communicative activity type’ with an obvious set of norms, rules, and routines. This fact makes politicians more intensely exposed to moments of mediated fragility which may be difficult to control. The most crucial condition that changes how politicians are able to manage their visibility is the constantly rolling ‘non-exclusive’ live camera which does not give the politician any room for error. The tabloids do not seem to mind ‘things going a bit wrong’ while airing; rather, interactional flaws are argued to be part and parcel of the overall web TV performance.

Introduction

The demands for constant media visibility for politicians have only gotten more intense in recent years with the explosion of new communication technologies. Not only the big broadcasting companies expect party leaders to join their news programs and prestigious live primetime debates; newspapers have become ‘multiplatform enterprises’ (Ju et al., 2013) which increasingly produce online video content, partly in the genre of political news (Negredo, 2014). In the Swedish context, the two tabloids Expressen and Aftonbladet proved during the Swedish General Election in 2014 that they have become prominent TV contenders to broadcasting airing a massive amount of material live from their websites as ‘web TV’. Along with regular broadcasting performances, politicians came to feature in innumerable news interview situations set up by the tabloids willingly expanding their visibility to web audiences.

In works from 1995 and onward, sociologist John B. Thompson has traced how the use of communication media creates new forms of interactions with their own distinct properties. These new forms of interactions that are linked to (new) communication media give rise to ‘a new visibility’ which brings about both opportunities and risks for politicians. Primarily basing his argument around analyses of political scandal, Thompson (1995, 2005) describes visibility as a ‘double-edged sword’. On one hand, it provides ample opportunities for communication with the public and chances for politicians to be displayed in a favorable manner in order to gain popularity. On the other hand, it is a visibility that is less controllable than previously and hence also a source of a ‘distinctive kind of fragility’ (Thompson, 2005: 42). The continuous media development leads Thompson (2005: 48) to further characterize the (then) current age as one of ‘high media visibility’. Multiple forms of mediated communication and the proliferation of media organizations are argued to have created an information environment that is more intensive, extensive, and less controllable than ever. The sheer quantity of the flow of information makes it more intense. Moreover, information has the potential to quickly gain global exposure making it more extensive. It is also much harder for political actors to hide actions, influence what material becomes public, and predict the consequences of mediated visibility, that is, the information environment is also less controllable than before (Thompson, 2005).

Thompson’s arguments relating to high media visibility for politicians as being more intense, extensive, and less controllable are still valid as general characteristics of today’s media environment. However, a lot has happened with communication technologies over the last decade, as well as with the media organizations using them. This, we argue, prompts new and updated research.

We approach ‘visibility’ as both a structural component shaping interaction and an accomplishment following from actions of participants in the unfolding interaction (‘management’). Rather than studying visibility in relation to an extraordinary event as the political scandal (cf. Thompson, 1995), we focus on a relatively new mediated encounter for politicians where the management of visibility is central – the live interview – here produced and streamed on the web by Swedish tabloids during election campaigning 2014. Thus, we connect the theme of politicians’ visibility in interviews to the very latest technological developments when it comes to new forms of TV. We believe this to be an innovative take in relation to how political visibility has been studied and discussed so far in existing literature. Our main questions are as follows: what conditions, particularly those deviating from ordinary broadcasting, can be said to shape the web interactions; in what ways do these conditions impact (enable or constrain) the actions of politicians involved in the web interactions; and what kinds of communicative fragilities arise in the web interactions and how are they managed by politicians?

Politicians, elections, and new media

Research that examines politicians’ management of visibility in similar formats as the web interview seems as yet nonexisting. This is hardly surprising as these types of interactions are very new, and also, the ways in which the web is used by (non-)broadcasters for inventing new production formats vary greatly from country to country. We are currently in the midst of a development that has just begun. However, studies that investigate online visibility of parties and politicians in relation to elections have, broadly put, looked at how new communication tools are adopted by individuals and organizations in the competition for voters (e.g. Chen, 2010, 2014; Chen and Walsh, 2010). Some studies pinpoint a particular social medium such as Facebook in order to examine how it is used as a tool for political communication, finding that it is still underused as a vehicle for any real kind of dialogue between citizens and politicians (Ross et al., 2014). Similarly, the use and role of Twitter during elections are the focus of a rising number of studies (e.g. Bruns and Highfield, 2013; Freelon and Karpf, 2015; Johnson, 2012; Jungherr, 2014).

The Internet and social media services are increasingly recognized as important in relation to elections and for political communication as a whole (Jungherr, 2014). However, the importance of social media for accessing, finding, and engaging with news in general is as yet deemed ‘relative’ in the sense that only a ‘significant minority’ incorporates them into their daily news habits (Nielsen and Schrøder, 2014). The exact role and impact of online election material on voters are also debatable. Although not proven to be the game-changer it was initially launched to be during the 2010 UK Election, Gibson et al. (2011) show that increasing numbers of citizens go online to inform themselves on election matters. They argue that this fact alone makes it difficult to write off the Internet as insignificant during election times (Gibson et al., 2011). For the sake of the argument in this article, it is not really relevant whether interview interactions with politicians on the web before Election Day are viewed by large audiences or not. The very fact that they take place at all and are aired live on the web, and that party leaders actually partake in them, is enough to justify scientific interest. It is part of the overall argument in this article that these interviews create new interactional situations for the management of visibility for politicians and that they hold the risk and the potential of massive exposure should a clip from them, for instance, go viral.

‘Web TV’ developments in Sweden

Web TV has become a collective umbrella name in Sweden for (generally) live TV productions on the Internet produced by various media actors, including both print and broadcasting companies. The general definition of web TV includes video-on-demand services as well as traditional broadcasters’ ordinary TV output which is made available as online, on demand programming. Although a rather broad term for Internet-based TV, we refer to it as

• Freely available live video productions that are accessible online through the professional producer’s website or channel, and which are

• Produced in a TV-like setting or manner (e.g. interactions with reporters and/or presenters and guests or co-reporters in or outside of a studio), and most notably are

• Presented under the self-labeled heading of ‘TV’.

Leading the way to the massive amount of live TV web coverage we see in Sweden today has been traditional broadcasters such as Public Service’s Sveriges Television’s (SVT) sports desk which first started airing live exclusive webisodes during the Vancouver Winter Olympics in 2010 before doing ‘the real’ TV broadcast (Kroon Lundell, 2014). Although a minor and experimental web venture, the webisodes caught the interest of audiences and also got other media’s positive attention. Other web TV news trials by SVT have not gone down as well with audiences and have been short-lived (see Ekström et al., 2013).

Although a broad concept referring to various forms of Internet-based TV, web TV specifically started to be regularly used by ‘new’ TV producers such as the tabloids in early 2013 as a self-branding term referring to live video productions staged in TV-like studios. The tabloid Aftonbladet announced that they were expanding their already popular online news site to producing TV in a quest to become a serious contender to the TV output of traditional broadcasters:

For us, it will be more about finding energy than to be technically brilliant. Tabloids have always wanted to be television. We have always tried to find the person behind the headline; we are in the business of feelings and we have always been fixated with providing visuality. Those are fundamental ingredients when it comes to TV. (Editor in Chief, Aftonbladet, 18 February 2013)

Today, most newspapers, national as well as local, produce live web TV material accessible online. During the weeks leading up to Election Day, Aftonbladet aired the first ever online party leader debate with all of the high-profile party leaders participating. The other main tabloid Expressen also held its party leader debate a few days later. Aftonbladet’s debate was co-streamed live on Expressen’s site and vice versa, and afterward, there were plenty of positive reviews also from linear media. It was generally agreed in reviews that the productions had been professional, relaxed but yet with a bite, and something from which the broadcasters themselves could learn.

Broadcast interviews and the canonical political interview

A common form for appearing before audiences in live or recorded filmed sequences is the broadcast interview. Various typologies and taxonomies of the broadcast interview are presented in research with similar characteristics (e.g. Corner, 1999; Femø Nielsen, 2006). The more hybrid character of the news interview has equally been examined (Ekström and Kroon Lundell, 2010; Hutchby, 2011; Kampf and Daskal, 2011; Lauerbach, 2006). Montgomery (2007) distinguishes between the accountability interview, the experiential interview, the expert interview, and the affiliated interview. The accountability interview is the type normally associated with politicians where an interviewer attempts to question what the interviewee (politician) is saying, while the latter tries to justify it (Montgomery, 2011). This kind of political news interview has been the focus of quite a lot of discourse analytic research (e.g. Clayman, 1992; Clayman and Heritage, 2002; Ekström and Patrona, 2011; Greatbatch, 1986; Heritage, 1985; Tolson, 2006).

In what we may call its canonical form, a politician may be interviewed by a reporter or news anchor, most likely a renowned one, from an established media company if it is a high-profile politician. It is an institutional form of talk with a relatively clear frame where the participants engage in asking questions and providing answers from distinct identity positionings, that is, interviewer and interviewee (Clayman and Heritage, 2002). The political interview can be described as a specific communicative activity type (Linell, 2009: 201). A communicative activity type specifies what kind of communicative situation the participants are involved in (for them and others such as audiences) and shape theirs’ and ours’ expectations of what is going on, and what the allowable or disallowable contributions are within the encounter. Communicative activity types can be described as comprehensive and overarching communicative projects (Linell, 2009) where participants are oriented to solving communicative problems or tasks on different levels of abstraction (e.g. the posing of questions and the answering of them).

Political broadcast interviews are familiar communicative activity types to party leaders, reporters, and audiences alike where visibility needs to be carefully managed. Their overall communicative conditions are well known to the participants. This makes it fairly easy to make sense of what is going on and how one should solve a particular communicative task, for example, in the form of a challenging question with regard to one’s party’s politics. Politicians and their press people rely on their prior knowledge and experience about the concrete situation, its participants, and the communicative activity type’s properties as described above. It is thus easier to plan and predict how to solve communicative tasks and thus to manage one’s visibility in a controlled, reliant, and recognizable manner.

However, interview types and formats are not static. As Montgomery (2011) reflects in his study on accountability interviews and the occurrence of micro-arguments within them, the accountability interview (and thus the political news interview as such) is not a form that simply reproduces itself canonically. Rather, it is unstable and changing and harbors both residual and emergent interactional norms. As for all situated interaction, participants in political news interviews are involved in the shaping of that particular encounter in a certain time and place, drawing on both discursive and nondiscursive resources and context-related knowledge as they do so. Simultaneously, they engage in the altering of long-term situation-transcending practices on which the structure, organization, and content of their interactions emerge and rely (cf. Linell, 2009: 52). A substantial contextual factor of relevance shaping interaction is also the development of new technologies of mediation (cf. Thompson, 1995, 2005) which alter the interactional conditions for individuals engaging in talk, allowing or disallowing certain contributions (Hutchby, 2001). Due to these factors, it can be assumed that substantial communicative changes will occur in the transit between broadcast and web interviewing.

Web interviews constitute a fairly new social encounter. It shares some of the interactional properties of an ordinary broadcast interview of the more generic character as described above. It is still an institutional encounter where the other participant who invites you to interact can be expected to be a representative of a media company but quite likely not one of the well-known broadcasters. The talk that goes on is also based on a turn-taking with questions and answers, and there are familiar recording techniques used (cameras and microphones). However, there are also other kinds of communicative conditions that do not apply for most TV broadcasts that render the situation more unstable. Two of these conditions that are easily recognizable as different are changes in settings where the interviews occur and changes in the interviewers’ institutional affiliation. We assume and expect that these altered ‘external’ conditions will impact on how the interactions in the web interviews unfold, and it is the task of the forthcoming analysis to examine exactly how this is manifested.

Data

The data consist of a corpus of recordings of live streamed videos from the two leading Swedish tabloids’ (Expressen and Aftonbladet) websites during the week leading up to Election Day in September 2014. Both tabloids had a vast number of scheduled programs but also aired more ‘spontaneous’, unscheduled live shows from different events containing interviews, debates, journalistic commentaries, edited news stories, and vox pop interviews.

The overall ambition with the data collection was to get what could be seen as a representative sample of the various program formats on offer, meaning that we recorded at least one program from each tabloid out of all the different formats that occurred during Election Week. In total, 80 videos or clips were collected, totaling nearly 16 hours (see Table 1).

Table 1.

Data overview.

Medium	No. of clips or videos	Total length
Aftonbladet	40	7 hours 48 minutes
Expressen	40	8 hours 55 minutes

For the purpose of this article, the data were first examined in their entirety in order to get a picture about the overall context(s) in which interviews appear. Thereafter, we strategically chose to focus on live interviews that took place after the live primetime broadcast party leader debates produced by the traditional TV companies. These ‘post-interviews’ are typical of the kind of live interview that the tabloids conduct with politicians for streaming on the web. They are normally done in backstage-looking settings usually less seen in regular news coverage, often against a somewhat disorderly background of other ongoing interviews and people on-the-go.

In this article, we present two examples of these ‘post-interviews’ in order to characterize the general conditions of the web interviews in relation to the canonical broadcast interview. The examples also contain ‘disturbances’ in so far as things occur that alter the communicative conditions within the ongoing interactions, and the ways in which the politicians’ manage these disturbances expose how the altered conditions for the interview may result in risks or opportunities when it comes to visual fragility for the politicians involved.

Analytical approach

The nature of live streamed web TV interviews is explored using the type of discourse analysis often described as the media talk approach (see Hutchby, 2006; Tolson, 2006) and the related research on traditional broadcast interviews (Clayman, 1992; Clayman and Heritage, 2002; Ekström and Patrona, 2011; Greatbatch, 1986; Heritage, 1985; Montgomery, 2007). This strand of research starts from a simple observation that media talk has a performative character (Tolson, 2006). It is produced and designed with the audience in mind in line with institutionalized conventions (Heritage, 1985). The key task is to detect such conventions and demonstrate how talk is designed in order to ‘relate to their audiences in specific, inclusive, and cooperative ways’ (Hutchby, 2006: 11). For instance, in a broadcast news interview, the performative and institutional character of talk is reflected in how interviewers avoid taking a personal stance and strive for a position of ‘formal neutrality’ while posing questions (Clayman, 1992; Clayman and Heritage, 2002: 119). Similarly, it is reflected in how interviewees subsequently orient to questions and provide elaborated answers.

Web interviews

Example 1

A few days before Election Day, the commercial TV channel TV4 arranges its big party leader debate. This is one out of two high-profile broadcasting events (the other being the Swedish Public Service Television equivalent the day after) where the party leaders battle it out on primetime close to the election. It is one of the crucial broadcast events that can possibly make or break a party’s chances of gaining governmental power, and it is thus detrimental that the party leaders manage their visibility in such a way as to appear both ‘votable’ and statesmanlike to the public. To this end, they are primed extensively beforehand by experts and press or media people. The broadcast inquiry is still ‘the main event’ for the politicians.

After the broadcast in question, the media gather in the TV4 lobby to interview the party leaders post-debate. The tabloid Expressen airs live on the web as the politicians walk from the broadcast studio into the lobby area to the awaiting cameras and interviewers competing for the party leaders’ attention. Expressen’s camera captures the hustle and bustle of politicians and press people moving around the floor looking for an appropriate spot to stop at as well as the security people’s monitoring of the area and their guarding of the entrances. The sound accompanying the images is rather loud and consists of a multitude of intermingling voices attempting to be heard through the noise.

It is not only the lobby context that frames this situation into a backstage one (in comparison to the broadcast studio’s live interactions); it is also apparent from the somewhat relaxed behavior and loud talk of people in the lobby area that this is understood as a completely different situation than during the broadcast. At the same time, the politicians are clearly aware and accepting of the fact that they cannot just walk off this ‘backstage’ situation as if mission accomplished. They stay and move around the room showing that they know that there are media organizations, notably the tabloids, wanting interviews for their live reporting.

Expressen now cuts from showing general interactions in the lobby to a medium shot of The Green Party’s Åsa Romson standing a bit to the side of the most intensive goings-on in the room (Extract 1). The male interviewer opens with a clarification that they are now ‘live’ in ‘Expressen TV’ and then asks an experientially oriented question as opener to which Romson provides a swift answer. The reporter’s phrasing ‘how do you think the debate went’ (lines 1 and 2) invites the interviewee to assess and provide her personal opinion about the quality of the broadcast debate, treating her as an expert who can and will successfully deliver an authoritative reply (cf. Ekström and Kroon Lundell, 2011). Romson also fulfills the question’s implied expert projection by clearly stating her views with an ‘I think it went well’; ‘I think’ being an expression that may be used to, in this case, upgrade knowledge claims as it positions the speaker as authoritative and knowledgeable (Ekström and Kroon Lundell, 2011: 674; cf. Simon-Vandenbergen (1996, 2000).

Extract 1

1	Reporter:	Well Åsa Romson then we are live in Expressen TV how do you think
2		the debate went
3	Romson:	I think it went well it was a fast-paced debate with lots of topics but there
4		were good topics because school, jobs, the environment and gender
5		equality are the kinds of issues that the Alliance government has failed
6		and where Sweden needs a fresh start. We need a new roll in the film
7		about Sweden and it will start on Monday I hope.

The reply is delivered with clarity and confidence while looking steadily at the interviewer who is outside of the viewer’s vision. During the posing of the second question (lines 8–10, Extract 2), a small group of people is seen entering the lobby from an entrance to the right. Romson whose back is to these events cannot see what is happening. The general sound level in the lobby most likely contributes to the fact that she does not notice that her major political rival in the broadcast debate, The Center Party’s Annie Lööf (then part of the government), has positioned herself close-by to the right of Romson to be questioned by another media organization. Expressen’s interviewer ends his question-turn at the very same time as Lööf’s interviewer does the same whereby Lööf quickly starts responding sounding and looking very cheerful just before Romson begins her reply. Now, it becomes apparent to Romson that Lööf is indeed standing right behind her. She starts replying, but after several re-starts and repetitions, she chooses to manage the situation by making a complaint (‘it’s very hard to do this’, line 14, Extract 2). She clearly shows with her facial expression (grimaces and looks grim) and body (takes a few steps away from where Lööf is standing) that she is not happy with the conditions of the interview. The interviewer (and camera) smoothly follows her movements, re-positions, and goes on asking a third question after an excuse of sorts from the interviewer (line 15) and a short laughter.

Extract 2

8	Reporter:	You in the red-green opposition got criticized for not being unified when
9		it comes to nuclear power. What do you think about the criticism you
10		received from the Alliance parties?

		[Annie Lööf walks up behind Romson and starts answering her interview question just before Romson starts to speak]
11	Romson:	Well it was not- it was not a new-, it was not a new issue when it comes
12		to which direction we should have in Sweden I spoke very clearly about
13		what we need to see in order to- /breaks away/

14

Romson:

it’s very hard to do this

15	Reporter:	Difficult to hear we continue Åsa Romson Björklund thought that you
16		can take the piece of coal home with you because you will need it in the
17		future if you are to phase out nuclear power what do you think about that
18		suggestion

There are several conditions surrounding the Expressen interview that do not apply for the common type of live broadcast news interview, and some which overlap. It is a post-interview in that it takes place in a backstage setting after a main broadcast event. Backstage settings commonly allow for a more informal type of talk and interaction, but here talk is still organized as a structured form of Q-A with the same kind of divided responsibilities between interviewer and interviewer as in broadcast equivalents. The encounter is filmed live for the immediate consumption of absent web audiences, and the interviewer represents a well-known media organization, if a tabloid. The affiliation of the interviewer can, however, complicate matters when it comes to what is expected of the interviewee, as can the platform used for the interview’s mediation. A tabloid is about conveying intensity, vivacity, drama, and feeling, but here, they are conducting live interviews with politicians a few days before the election. With what degree of seriousness should one orient to the questions and design one’s answers, also given that this is aired on the web with potentially different demands on how to talk and act? Although the interview starts off with a purely experiential question which could be perceived of as more lightweight and tabloid-like, the second and third questions ask the politician to elaborate on criticisms she received during the prior broadcast and can, therefore, be seen as a kind of extension of what went on in the studio. At the same time, the questions’ variations of ‘how do you think’ and ‘what do you think’ invite the politician to reflect on prior interactions rather than holding her accountable – in that sense, they are not really challenging her personally or her politics.

The situation where Lööf comes up from behind of Romson and starts talking is able to happen because Expressen’s camera is nonexclusive when it comes to who is in the shot and who is not. Typical broadcast interviews are exclusive, that is, one politician is interviewed at a time in a focused interaction with a clear beginning and a clear end, often also edited before being broadcast. Here, not only does the live camera sweep all over the room to capture various interactions while waiting for an interview to get going, once the interview is underway, the camera person makes no attempt to solely focus on the interviewee but allows the shot to include also the entrance of Lööf and the start of her interview. What would normally be seen as a disturbance to an exclusive interview becomes part of the interview performance as ‘staged’ by Expressen. As mentioned, it also impacts Romson’s behavior, causing a mini-drama as she disengages herself from the interview and states ‘it’s hard to do this’ while making a sour face. In combination with her actions of stepping aside and looking away while the camera is still rolling, what she is seemingly saying is that she cannot or will not do this unless the interactional conditions change, and Lööf is removed from the shot. This implicit message seems to get through quickly enough as Expressen’s reporter (and the camera) adjusts to the new positioning and allows the interviewee to regain her composure and again become the exclusive object of the tabloid’s attention.

Romson can be said to exercise her power as interviewee by demanding an interactional change and getting her way. At the same time, she takes a risk when it comes to the management of her visibility. If the interviewer had not followed suit and glossed over the disturbance with a comment that it was ‘difficult to hear’, but instead asked a challenging question about Romson’s relations with Lööf (which at the time were frosty after Lööf’s criticism of Romson’s politics during the broadcast), Romson would have found herself in a much more fragile position. At worst, such a sequence could have gone viral and caused the tabloid the following day to talk about ‘The War between the Women’ racing for governmental power or something equally dramatic.

Even the interactional move that she does make is so unusual had it been a regular broadcast interview that it could potentially have been recontextualized both by Expressen and other media and made into something juicy. With the appropriate headline, it could have served as ‘proof’ that Romson was a stuffy and inflexible politician unable to cope with minor disturbances with a sense of humor: Instead, here she is pulling a face like a child who has just lost her lollipop to a rival kid in the sandbox. Apparently, the communicative project of the tabloid was not to dramatize the actions of Romson at this particular point in time as no such clips were exploited.

Example 2

The second example involves the same politician, now interviewed by the other main tabloid Aftonbladet, again after a TV4 broadcast 4 days before the election. Various media have been allowed access to the studio-floor at TV4 to conduct interviews and take pictures. As in the prior analyzed sequence, journalists, camera people, politicians, press secretaries, and others are moving around the room, and several interviews take place simultaneously in different parts of the studio. The interview is conducted in what seems to be a corner in the back region of the studio in front of some colorful screens. Romson is first seen finalizing an interview with Expressen before being quickly directed by an assistant to the Aftonbladet team standing by. Meanwhile, people from the former interview move away from the shot. The clip is filmed with a slightly shaky hand-held camera.

The interviewer, judging from her voice a rather young woman, opens the encounter by stating the politician’s name and mentioning her affiliation (line 1, Extract 3). The opening fills a double function as it both introduces the interviewee to the viewers and works as a self-presentation aimed at the politician. This is an opening that could be expected in this context, but the politician seems to treat it as a greeting and responds rather informally ‘hi there’ (Sweden ‘hejsan’, line 2), a phrase that has colloquial even slightly childish connotations. The reporter returns this greeting with the same phrasing a bit hesitantly, possibly indicating that she is a little surprised by the interviewee’s choice of wording. She then pursues her introductory question (lines 3 and 4).

Extract 3

1	Reporter:	Åsa Romson we are live now here on Aftonbladet dot se
2	Romson:	hi there
3	Reporter:	hi there we heard you say here that I’m bad at planning financial matters
4		what is it that you find
5	Romson:	heh
6	Reporter:	difficult

In the question is embedded an alleged quote by Romson regarding her (dis)ability to plan her finances implied to have been uttered during the broadcast (‘we heard you say here …’). The interviewer adds a question to the quote which beckons the politician to elaborate what it is that she meant by the quote. In the midst of the question, Romson smiles and elicits a quick laughter (line 5) to which the interviewer responds by audibly smiling as she finishes her question. Notably, it is not a question that asks of Romson to generally assess her performance or the quality of the debate, treating her as an undisputable expert as in Example 1 (Extract 1); it is an accountability-like question that, although posed in a seemingly innocent sing-song tone of voice, could potentially threaten the authority and legitimacy of the politician depending on how satisfied (or not) the interviewer is with the politician’s subsequent response.

The opening of this interview clearly deviates from the formality of a regular broadcast interview. Viewers both see the ending of the former interview and the slight delay before the interviewee is directed and positioned in front of Aftonbladet’s camera. This is normally not the case on TV where bridges between communicative activities are generally edited away or not shown to the audience. In the opening phrases, one can detect an uncertainty on part of the interviewee as to the formality of the interaction. As in Example 1, it is a live interview, but at the same time, it is a post-interview after a broadcast; the interviewer’s affiliation is a tabloid, and moreover – which could have bearing on how questions are treated – the interviewer is a young-ish woman with a sing-song pitch and not a renowned male reporter booming out questions in an assertive manner. These conditions likely influence Romson as she formulates her ‘hi there’ instead of using a more somber greeting, projecting her presuppositions about a more leisurely encounter. When replying to the interviewer’s accountability-like question about her bad financial planning, she chooses to keep an informal approach (e.g. choosing words as ‘hung around’, line 9, and ‘super stable’, line 11, Extract 4).

Extract 4

7	Romson:	well it is my own finances that I’ve never really been especially
8		interested in
9		I’ve .hh hung around- studied a lot at university and of course that’s not
10		good to do for too too many years /main studio lights go out/

11		because then you won’t end up with a super stable a super stable eh
12		economic situation

13		now they seem to be out of lamps in this studio maybe they don’t have
14		such a good financial situation at TV4 either
15	Reporter:	heh-heh-heh-heh
16	Romson:	I don’t know whether they’re in control but-
17	Reporter:	but we have a camera lamp anyway so you are still visible
18	Romson:	OK OK but you know I haven’t lived such a life that I have maximized
19		my choices when it comes to what is optimal for my own private
20		finances

As Romson starts elaborating as to what she meant by her statement regarding her financial situation, the lamps in the studio go out. While still talking, she twists her head and glances backward as if trying to see what is going on. Without pausing, she cracks a joke about the incident at her own expense by connecting the dark lights to the broadcaster possibly having similar financial problems (‘maybe they don’t have such a good financial situation at TV4 either’, lines 13 and 14, Extract 4). This elicits a laughter from the interviewer (line 15).

Romson thus uses the lamps going out as a resource in the ongoing interview, and she does so in a way that exceeds the traditional institutional roles and conventions of the news interview. Her joke and the interviewer’s laughter break the institutionalized discursive roles of questioning and answering; these are, if only momentarily, dissolved. Joking and laughter are actions that can exhibit solidarity and shared values. Joking in interaction is generally seen as an action generating a feeling of friendship and belonging (Zajdman, 1995: 327–331). Laughter is (in many cases) associated with appreciation (Hay, 2001: 76). Here, the interviewer’s laugh expresses an understanding of the politician’s orientation to winning the viewers’ appreciation, something which is an obvious breach of the relationships set up in formal news interviews.

After the interviewer has clarified that the politician is still visible on camera, Romson keeps explaining the context to her quote embedded in the first question whereupon she receives a third question pursuing the matter further (Extract 5).

Extract 5

21	Reporter:	but if you’re bad at planning your finances is that something a minister
22		that you maybe will become here soon should be saying really
23	Romson:	no I took it as a pretty personal question if I was worried about my
24		pension and the best thing in such a situation for a politician is to answer
25		honestly and eh since I’m not someone who plans for my pension then
26		it’s better to own up to that fact /…/

The interviewer now orients toward a more adversarial stance pushing for a particular answer (lines 21 and 22, Extract 5). She uses a journalistic technique common when journalists perform more critical interrogations involving a preface followed by a yes–no question (Clayman and Heritage, 2002). This preface puts the interviewee in a position in which she is most likely to start her answer with a no (which she does, line 23). The question functions as an accountability question (Clayman and Heritage, 2002), and Romson provides an answer which contains a justification of her answer in the TV4 interview. From this point onward, the interview continues with more formal questions and answers. The journalist orients toward a neutral stance and poses questions concerning energy supply and the potentialities and probabilities of different government coalitions while the interviewee acts as a representative of her party. Except for the shaky camera and the somewhat disorderly backdrop, this part of the interview shares more of the characteristics of the traditional news interview.

The interview sequence illustrates that this kind of web interview is perceived of as a considerably fuzzier communicative activity type than the regular broadcast ditto. It seems that the interviewee goes into this interview expecting it to be an experiential interview (Montgomery, 2007). Initially, the interviewer’s actions do not challenge the interviewee’s understanding of the situation which in a sense facilitates the joke about the lamp; a more personal, informal atmosphere provides a good ground for joking. The politician’s orientation to a lighter approach is both acknowledged and resisted by the interviewer who shifts between echoing Romson’s colloquial greeting phrase and appreciating her joke while posing accountability questions, pushing for answers regarding Romson’s legitimacy as a potential part of government. The interviewer uses the power that comes with the discursive role as a questioner with the rights to visualize and distribute whatever the camera registers in order to turn this into an accountability interview characterized by more formal and impersonal roles and relationships.

Romson exploits the fuzziness and the messiness of the interview surroundings as a resource in order to appear personable and witty which is generally believed to be beneficial for the politician’s public persona. However, she needs to rely on the interviewee to give her appropriate responses in order to pull off being this casual. In that sense, she subjects herself to a risk when it comes to her visibility. If she made herself fragile in Example 1 by possibly appearing too stuffy when disengaging herself from an uncomfortable interview situation, Romson here dodges the underlying seriousness of the questions by virtually brushing them off with a joking attitude. This could potentially be damaging in the opposite way; will she appear too nonchalant or, perhaps even worse, will it look as if she is not recognizing the young female interviewer as having the legitimacy and authority to hold her accountable in this context? As in the former example, the tabloid’s interviewer, despite pursuing her questions as described above, makes no explicit attempt to expose Romson for choosing a relaxed mode of address during the encounter.

Discussion and conclusion

Through detailed analyses, this article sets out to explore the altered conditions of political interviews primarily resulting from a change of communication platform (from broadcasting to the web), and media organization producing the interviews (from established broadcaster to tabloid). Although following the well-known institutional structure, organization, and role division of any media interview, it is clear that the live web interview is not yet a recognizable communicative activity type with an obvious set of norms, rules, and routines. What is or is not an allowable contribution is still up for negotiation not only in the local interaction but in terms of the format as such, both on part of the interviewer and on part of the interviewee. Pushed in a studio corner after a ‘real broadcast debate’, the politician may be faced with experiential questions mingled by accountable ones causing confusion about the seriousness in approach called for by the politician. Potentially harmless experiential questions where the politician is oriented to as an unchallenged expert may unexpectedly turn into a fragile moment because of the nonexclusiveness of the camera which shows more than what is known by the interviewee. Suddenly, the politician may be faced with a competitor on camera and is forced to take an immediate decision how to handle it without losing authority.

We propose that the most altering condition impacting politicians’ behavior during web interviews in contrast to broadcasting – and consequently that which is the core of the new visibility facing politicians in these instances causing fragilities – is the nonedited, constant live camera. The tabloids’ expressed ‘ideology’ when it comes to producing TV for the web is simply ‘let the camera roll’. They are not bound by strict time constraints where they need to cut down a longer sequence to a few seconds to fit in a news slot. Audiences have come to expect that the web contains plenty of complementary ‘extra’ material anyway, so imperfections, unplanned events, and dead time need not be cut. The web has also long promoted a do it yourself (DIY) aesthetics where it is perfectly fine – if not even expected – that you supply material that looks a little rough and scruffy around the edges so as to appear more authentic.

The apparent ‘messiness’ in which web interviews occur, including mess as a result of things happening within the encounter itself, is thus not a problem to be managed by the tabloids. Messiness needs only be controlled so that it does not disturb the actual access to the politicians and/or creates obstacles when it comes to getting the interview going. Whatever happens after that is part of the performance, that is, that which is offered to audiences. The roaming nonexclusive camera positions the politician as a constant visible prey. No clear boundaries exist between one mediated encounter and another. Although the politician might not be aware of it, a multitude of simultaneously live filming cameras treat you as a front stage person also in transit between interviews, and most certainly also during them, whatever happens. As long as you are in the kind of post-interview lobby area previously described, although it may appear to be a backstage setting, you are really in the midst of an area of high media visibility.

Although no politician has yet been outed in any dramatic way for doing or saying the wrong thing during these types of web interviews, it is a form of visibility that is unpredictable and in that sense obviously uncontrollable for politicians. The recontextualized clips that were posted from web interviews during the 2014 election contained ‘funny moments’ which rather boosted the politician in question than worked in the opposite direction. This lack of exploiting potentially colorful material in a negative way for an individual politician has a plausible explanation. The tabloids are relatively new on the scene and need access to politicians in order to be able to stream live material with them as elite sources. Politicians may, therefore, have experienced a particularly kind treatment by the tabloids this time around so as not to interfere with their budding relations and their roles as legitimate producers of political TV content. But politicians cannot rely on this kindness in the future as the tabloids are always hungry for videos that have a potential for viral success. Politicians’ wrongdoings in these circumstances create new opportunities for the tabloids and other media to accomplish such uncontrollable coverage.

We predict that these types of web-related interview practices will eventually lead to the altering of broadcast interview practices when it comes to structure, content, and organization of interactions. It is, of course, impossible to say exactly how they will alter at this point in time when the web phenomenon is still so new. However, it is not unlikely that the ‘let-the-camera roll’ practice will be used to a larger extent also in ‘polished’ broadcast news with the potentially damaging visibility consequences for politicians described above. As of now, further research should trace how similar developments occur in various international contexts in order to explore and understand exactly how technological developments impact both the interview format as an institutionalized type of interaction, as well as on the more long-term relations between politicians and journalists worldwide.