Share a little of that human touch: The marketable ordinariness of security and emergency agencies’ social media efforts

Today, communication specialists working for public security and rescue services increasingly use superficially personalized content, or apply ‘a human touch’, to promote their organizations in social media. To theoretically capture and understand such processes, the concept of marketable ordinariness is proposed. This refers to how the communication relates to everyday conceptions – through feelings, humor, cool vehicles or pet animals – and is made marketable, suggesting there is a promotional logic at work. Drawing on appraisal analysis of interviews with communication specialists, the article examines this strategy’s discursive elements, including the semiosis of simplicity, emotion, promotion, storytelling and quantitative success, pointing critically to the ways they aid marketization – the process whereby promotional culture encompasses increasingly more sectors and areas of life. It then discusses a number of implications. First, the public sector employees’ alignment with both informational and promotional values and communication may give rise to an authenticity paradox, leaving everyone else wondering when each standard applies. Second, a stronger promotional identity implies compromised professionalism, favoring certain abilities and choices and underutilizing communication efforts that (a) do not pursue big publicity and (b) involve any issue suspected to be challenging for the organization and mainstream culture.


Introduction
The past decades have seen public sector organizations undergo multifaceted changes (Arellano-Gault et al., 2013;Ezzamel and Reed, 2008). For instance, adoption of more commercially-oriented logics and practices is a broad reaction to allegations of lack of transparency and efficiency. Included are new forms of performance measurement and management, but also greater attentiveness to image and reputation (Fairclough, 1993;Mautner, 2010). These changes in the public sector involve the articulation of new forms of communication. Taking responsibility for information sharing and transparency in line with principles of bureaucracy and legality is not enough (Du Gay, 2000). Rather, it has become a widespread practice to extend the use of communication beyond core organizational tasks to fashion identity and ultimately create value (Mumby, 2016;Va´squez et al., 2013). The goal is to combine distinctive, positive characteristics with emotional and personal appeal to make the organization a liked and trusted partner (Bengtsson and € Ostberg, 2006;Sung and Kim, 2018;Temporal, 2015;Zhao and Zhan, 2019). Features of Web 2.0 offer some of the paramount tools for this (Johnston, 2015) including popular social media platforms. Exemplifying some of these developments, law enforcement internationally is attempting to contribute more relatable information in social media, looking to take so-called community policing to online spaces. This is not just a consequence of developments in participative digital media, but also of criticism that the police lack transparency, accountability and responsiveness (Ramirez, 2018), and need instead to 'partner with the community' (Bossler and Holt, 2013: 354), and make citizens, who have been too unengaged and fragmented, 'jointly responsible for crime reduction' (Crump, 2011: 1).
Venturing into this sector, this article draws on interviews with communication specialists working in the security and emergency services sector in Norway, including the police. While organizational research has explained how practices of branding play an increasingly important role in fashioning identity and creating value for business organizations (Mumby, 2015(Mumby, , 2016Willmott, 2010), this article demonstrates how these communication specialists, despite working for non-profit public bureaucracies, take on responsibility for creating similar value-creating communication processes. Through attention to the functions of language use for meaning-making, stance and subjectivities, the article critically conceptualizes a strategy in use that makes available more market-oriented and 'ordinary' discourse types and identities than are traditionally associated with bureaucracies. Through the practices involved in this strategy, the professionals convey a sense of control over social media and audience reactions, but also give expression to tensions and dilemmas. The study thus relates to previous work on how market and enterprise logics entail some dilemmas for rule-governed state bureaucracies (Arellano-Gault et al., 2013;Du Gay, 2000Hoggett, 2006). It also contributes to discourse analytical research that has mainly studied the promotional materials from public sector organizations Harvey, 2015, 2016;Fairclough, 1993;Han, 2013;Machin, 2015, 2016;Mulderrig, 2018) rather than the meaning-making of the professionals creating such content. From the perspective that communication contributes to constituting organizations (Fairhurst and Putnam, 2006;Kuhn et al., 2017;Putnam and Nicotera, 2008) it is critical to deliberately examine and openly discuss the workings and logics of public sector communication that may have a profound impact on these institutions and citizens, as the present study undertakes to do.

Marketization and the public sector
Marketization refers to the expansion of a competitive economic logic to other spheres than business -to the public sector and even to culture and citizens themselves. Public sector organizations that adopt marketization thus 'become more market-like in their actions, structures, and philosophies' (Eikenberry and Kluver, 2004: 133). This development entails several difficulties. When prevention and market solutions expand at the expense of state support, the transfer of responsibility from institutions to individuals makes citizens prone to increased vulnerability (Rasmussen, 2011;Rose, 1999). Moreover, if organizational identity consists of strong and jointly accepted characteristics that are slowly renewed together with contemporary identity aspirations (Golant et al., 2014), then marketization involves intensified negotiation of purposes and identity (Du Gay, 2004). A marketing-based focus on being competitive through uniqueness can undermine an entire sector's more basic, national mission to provide equal service to citizens everywhere (Brookes and Harvey, 2016;Sataøen and Waeraas, 2015). An attention-seeking, 'responsive' public sector may exceed the boundaries of organizational identity and go 'out of character' (see Czarniawska, 1997). Several identity-related questions arise in such a situation: '"Is this who we really are as an organization?" or, more provocatively, "Is this who we are becoming as an organization?" or even "Is this who we want to be?"' (Gioia et al., 2000: 76). Such negotiations both take place within communication and are prompted by communication efforts.
A debate has transpired over the extent to which marketization and enterprise logic reconfigure the public sector; that is, organizations that primarily get their funding from state resources and operate under politically agreed policies (Hoggett, 2006). These institutions were established to maintain the wellbeing and productivity of the population and thus the security and advancement of the state (Dean, 1999;Foucault, 2008). Many commentators describe marketization as complex and multilayered, with diverse tendencies across countries and parts of the world (Ezzamel and Reed, 2008). Norway, where the present research was conducted, displays some of this complexity. It is considered to have a strong public sector, but has nevertheless carried out some privatizations. Its institutions are implementing new public management reforms modeled after private sector quality management (Christensen and Laegreid, 2009) and prior studies have found that some principles of branding are being incorporated in public agencies' communication (Sataøen and Waeraas, 2015). Fairclough (1993) examines a variety of communication practices that exemplify marketization. Many of them fit into what he calls promotional discourse, including claims of achievements at organizational, group or individual level, contributing to the building up of a genre of prestige. Other public sector studies have shown that the promotional discourse of universities constitutes a form of values-based advertising. Glossy images featuring large open spaces and welldressed people who typically find themselves in these open areas rather than in real working situations, connote freedom, dynamism, movement and success (Ledin and Machin, 2016;Mautner, 2010). The language consists of positive but generic markers of prestige such as 'excellence', 'performance', 'quality', 'flexibility', and 'innovation' (Mautner, 2010;Morrish and Sauntson, 2013). Two aspects of the language of marketization, however, are the overlapping features of conversationalization and personalization (Fairclough, 1993). These feature the use of personal appeal, emotive lexis, slang and dialect, conveying directness, spontaneity and liveliness, and thus creating a seemingly personal relationship with the target audience. Studies of healthcare campaigns have demonstrated the use of similar forms of address, and of images and texts conveying feelings of fear and discomfort concerning health problems, so that the target audience would want to avoid the danger by taking preventive steps for themselves and their family (Brookes and Harvey, 2015;Mulderrig, 2018). Related uses of emotive discourse and a personto-person address also appear when one takes into account developments in branding and uses of social media.

Branding and social media
Branding, Willmott (2010: 519) explains, functions through three productive phases: (1) it utilizes consistent semiotic choices, conveying 'reassurance of standardized quality indicated by a brand name and/or logo to prospective purchasers of impersonally traded goods'; (2) it associates the organization or product with particular attitudes, values, beliefs and lifestyles, conveying another level of quality and desirability beyond mere consistency, increasing the so-called brand equity; and (3) value is increasingly placed on this manufactured entity that is communicated and circulated through consumer/user involvement, and it 'assumes a monetized value as an intangible asset'. This value may, in the case of successful brands, well exceed the organization's physical assets.
For state bureaucracies, the purpose of branding is obviously not the creation of value for shareholders, but rather the creation of other intangible assets such as trust and involvement from citizens that improve efficiency and cost-effectiveness. Another justification of the public sector's adoption of branding is that its work is organized (or at least viewed) as if it were taking place in competitive markets. A recent best-practice book states that: 'Increased competition in every sector makes branding an imperative. From nations to non-profit organizations, from civil services to cities, from sectors to services, there is no escape from the tough requirement of the need to create a point of differentiation and a positive image' (Temporal, 2015: 7). Furthermore, branding makes communication, as Mumby (2015: 25) describes it, 'a key element in the production and reproduction of value'. Mumby illustrates this with two examples. First, communication practices constitute professional and organizational identities in the workplace. Second, communication is the linchpin for involving consumers/users so that the organization can adjust products/services to their tastes, but also for influencing the value of products/services by putting together meaningful language and visuals to convey attractive ideas and stimulate meaning-making processes (Mumby, 2015). In these circumstances, social networking sites become strategic devices as part of planned professional work, but also because 'any free, autonomous act of communication has the potential to become free labor that is brandable' (Mumby, 2016: 886). Social media are defined by this dual function. They are used to capture, channel and manage social interactions potentially among several users simultaneously, but they also continue 'capital's attempt to valorize social labor' in that the voluntarily involved users/consumers are made productive (Beverungen et al., 2015: 474).
The use of personality and emotions in organizations' communication, which Fairclough (1993) critically analyzed early on, has become increasingly widespread with the rise of social media. Most empirical research, however, focuses on understanding and developing methods for how organizations should exploit these communicative affordances for organizational efficiency and success (Barcelos et al., 2018;Gretry et al., 2017;Liu et al., 2017;Sung and Kim, 2018;Zhao and Zhan, 2019). Several studies find it effective for organizations to create vivid posts containing photos and videos to attract attention, and to convey emotional appeals such as happiness or sympathy to achieve more audience interaction (Liu et al., 2017;Sung and Kim, 2018;Zhao and Zhan, 2019). Other research results point to specific conditions under which this informal style of communication is beneficial to the organization, for example whether an organization is well known (Gretry et al., 2017) or if its offerings to the general public involve low involvement and risk, and thus do not concern major financial decisions or health (Barcelos et al., 2018). It has also been shown that social media accounts are not followed by the public unless they provide content designed to convey a meaningful message (Ramirez, 2018). Furthermore, the use of personalization and emotional appeals in organizations' communication via social media is not a free-for-all; managerial measures are taken to establish control over the discourse and keep it within certain boundaries (Banghart et al., 2018). Various uses and control measures have also been demonstrated in research on police authorities' use of social media (Brainard and Edlins, 2015;Meijer and Thaens, 2013;Rasmussen, 2017).
Some police departments will allow informal communication, with police officers occasionally sharing information about their personal lives on police Twitter accounts (Meijer and Torenvlied, 2016).
Regarding such practices, Thumim (2012) explains that the digital era has seen a proliferation of self-presentation and more ordinary identities. This theoretical understanding of the term ordinary includes, however, something other than its everyday meaning, as it refers to an easily recognizable cultural package of signs and cues that you would associate with the broad populace rather than experts or bureaucrats (Thumim, 2012). As a communicative resource, ordinariness does not just present itself, but requires work. According to Sacks (1984: 415), the 'work of being ordinary' is a resource for creating mutual relatedness and community. It is an interactional accomplishment through which any event or person is made normal and intelligible. Doing 'the work of being ordinary' requires knowledge about what is perceived as recognizable and how these features may be represented. Sacks describes some of the work as searching for storyable characteristics.
Currently, this appears to be an underutilized concept in the literature on organizational social media use.

Rationale and research questions
The present study adds to previous research in three different ways. First, in relation to discourse analytical research, it provides insight into an understudied aspect since the field has examined the promotional materials of public hospitals and higher education (see Harvey, 2015, 2016;Fairclough, 1993;Han, 2013;Ledin and Machin, 2015;Mulderrig, 2018) but not the discourse of the professionals involved or other sectors. In particular, the analysis of interviews gives a more in-depth and nuanced understanding of the workings and dilemmas of communication efforts at a time when public organizations to varying degrees are adapting to 'brand society'. Second, the study seeks to contribute to the theoretical understanding of value-creating communication and branding, which has been developed through organizational research that has focused on the mechanisms of major societal and organizational change (e.g., Mumby, 2015Mumby, , 2016Willmott, 2010), and that therefore might benefit from the present study's emphasis on conceptually capturing and explaining everyday communication efforts. Third, it is clear that the majority of research examining the incorporation of interpersonal strategies into social media among business brands (Barcelos et al., 2018;Gretry et al., 2017;Liu et al., 2017;Sung and Kim, 2018;Zhao and Zhan, 2019) as well as in the public sector (Brainard and Edlins, 2015;Crump, 2011;Meijer and Thaens, 2013;Meijer and Torenvlied, 2016) focuses on understanding the practice in order to make it more effective (e.g., reaching consumers, involving citizens), which also means that more could be done to underpin critical reflection on these communicative developments. The present study thus aims to analyze whether and how the social media efforts of state bureaucracies entail professional adoption and negotiation of marketized discourses of communication and self-management. To fulfill this aim, three research questions are addressed: (1) What kind of strategy emerges that can enable the professionals to manage social media? (2) What are some of the key discursive elements and subject positions of such a strategy? (3) How do the professionals evaluate and negotiate it, or convey interdiscursive features that involve dilemmas or tensions?

Methodology
In order to investigate the above-mentioned aim and research questions, a qualitative research design was adopted including interviews with communication specialists working for security and emergency services in Norway. The motivation for this choice was that previous studies have observed gradual changes in Norway's public sector brought about by the incorporation of corporate management programs and branding (Christensen and Laegreid, 2009;Sataøen and Waeraas, 2015). This trend can also be seen internationally (Ezzamel and Reed, 2008), which indicates that further observations and insights from our region are likely to be relevant for similar research and practice in other parts of the world. As in other countries (see e.g., Du Gay, 2008), the sector faces market-oriented requirements to be 'responsive' to citizens. In addition, it has a substantial social media presence in Norway (Politidirektoratet, 2011). Moreover, the interviewees represent a professional segment of 'communication labor', which refers to 'the myriad ways in which the creation of mutuality, intersubjectivity, and authenticity has become a key element of value production under communicative capitalism' (Mumby, 2016: 896). Notably, it is a category of employees that has grown significantly in numbers (Christiansen and Henriksen, 2014; US Bureau of Labor Statistics, 2017).
The study uses semi-structured interviews, which suit research that aims to examine the sense-making of employees, particularly when the researcher is not looking to dig deeply for the truth beyond language (Alvesson and Deetz, 2000). Actually, on the positive side, culturally scripted phrases and positive storytelling make up some of the rules, norms and presentational techniques that, in a discourse analytical strand of thinking, are keys to achieving an understanding of the phenomenon under study (Potter and Wetherell, 1987). Moreover, with the ethnographic interview as a model, the semi-structured interviews involved making an effort to achieve a good rapport and a reasonably long conversation with the interviewees, so that they could articulate some of their more personal considerations as well (Spradley, 1979).
Since communication managers had the overall responsibility for communication strategies including on social media, it seemed reasonable to interview them along with at least one of their employees who spend most of their time working hands-on with social media. Six different security and emergency organizations responded positively to the invitation to participate, including police and fire departments at the local level, and security, police and health organizations at the national level. With 14 interviews lasting an average of 48 minutes, the material covered enough detail and variation (see Creswell, 1998) to explore communication specialists' accounts of social media use and content. The respondents consisted of nine women and five men, with an average of 15 years' experience in communication. All interviews were conducted in person at the respondent's place of work. The interviews were recorded and transcribed verbatim with some adaptation to written language. Excerpts for detailed analysis were translated by the author from Norwegian into English.
In exploring the first research question about a communication strategy that appears to be important and makes it possible to manage social media, according to the professionals' narratives, the analysis identified recurring characteristics in stories as well as between interviews. The emerging pattern led to a reflexive 'back and forth' interplay between interview data, theoretical literature and a creative search for how to conceptually capture the phenomena observed. The interviewees' narratives of attempts to communicate in such a way that people would take interest and relate to the content meant, analytically, that they were clearly involved in creating strategy; that is, in finding 'winning solutions' and 'modes of action' to affect the possible actions of others (Foucault, 2000: 346). More specifically, they seemed to be involved in creating an 'intersubjective strategy' (Martin and White, 2005: 6), in that their communication was supposed to stimulate the emergence of shared meaning. Related to this, interviewees repeatedly represented communication efforts linguistically and thematically in a way that could correspond to the private life of anyone and anybody. In the analysis, this came to be understood as a fashioning of 'ordinariness', since the term had previously been used to distinguish the communicative construction of community, as opposed to elite, expert and strictly professional discourse that establishes a distance to both the general public and private life (Sacks, 1984). Furthermore, in addition to the narratives testifying to strategic, promotional intentions, they systematically included certain elements and not others, so as to create, as Willmott (2010) described regarding branding, another level of quality and desirability. In this regard, the analysis could identify an order or rule of discourse and exclude the possibility that the communication specialists would construct any 'ordinary' images of working life. Just as materials and products are refined to become marketable in any business, the process of content development appeared to involve a form of discursive purification. Therefore, it seemed fitting to continue analyzing this particular strategic business of communication specialists using the concept of marketable ordinariness.
Pursuing the second research question, the analysis explored how discursive elements were composed and put to use; that is, how the 'building blocks of strategies' that 'have no identity (as strategies or parts of them) . . . begin to form part of a tactics that will surface in a strategy' (Strozier, 2002: 123-124). The analysis also took into account how language use affects identity, how communities are positioned by specific configurations of discourse (Martin and White, 2005), and how actors align themselves and call on others to take up certain social positions; that is, it took into account the differentiated and interconnected roles and statuses of a specific discourse (Schatzki, 2002). It then took note of instances of interdiscursivity, in this context taking the form of communication work constituted through diverse discourses that oftentimes have developed in different time periods and spheres (Fairclough, 1993). Finally, when exploring the third research question regarding how the professionals evaluate and negotiate their work, the analysis drew on appraisal analysis (Martin and White, 2005) to detail how speakers invested in value positions, how they aligned and dis-aligned, or how they brought up the practices and opinions of presumed others in the conversation. Meaningful discursive choices consisted of valuing lexis, bare assertions assuming there is consensus, dialogic contraction and expansion, proclamations and disclaimers, and constructions of force and necessity (Martin and White, 2005). Several seemingly small discursive moves carry important meaning potential.

An analysis of the strategy of marketable ordinariness
The following sections analyze communication specialists' accounts of social media use, demonstrating how a set of discursive elements come together in a strategy that I conceptualize as marketable ordinariness. Collectively, these elements form building blocks with specific, strategic meaning potential (see Strozier, 2002). As will be shown, fundamental elements such as the semiosis of simplicity, the everyday and emotion are coupled with an element of promotion. The latter is indicated not only by semantic choices (marketing lingo) but also by representations of successful organizational and communicative tasks. Furthermore, storytelling performs the function of tying together and articulating the aforementioned elements, potentially attracting and retaining social media user attention through continuitycreating meaning-making. Beyond these features of social media content, the professional discourse also includes assessment of performance through tokens of quantitative success. This particular discursive element performs the function of justifying and thereby contributing to the strategy's continuous reproduction. Finally, the analysis demonstrates interdiscursive features that are noticeable in interviewees' discursive realizations of, on the one hand, the impartial and impersonal attitude of the bureaucracy and, on the other hand, the promotional, emotive and personalized qualities of branding. In this regard, social media use appears as a vehicle for ongoing negotiation of old and new practices and organizational identities. The strategy of marketable ordinariness thus involves contradiction and it hinges on diversification. It hinges on communication specialists taking on the task of managing diverse content channeled towards customized social media accounts, which enables more 'ordinary' and promotional identity work but also risks disintegrating the ethos and identity of bureaucracy.
The discursive elements of simplicity, the everyday, emotion and promotion The following example, taken from an interview with a police communication officer, deals with the department's content on Facebook and Twitter and highlights discursive elements involved in the articulation of marketable ordinariness: First, she represents the social media content as easy to understand. The interviewee begins by asserting that the content can consist of a picture, implying that text or sound are unnecessary for the message to come across. She mentions that 'dogs and horses are currently very popular', a bare assertion treating the subject as not requiring any explanation. As if to suggest that some other material, perhaps more serious and advanced, would have been expected, the interviewee initiates these first propositions with distancing formulations ('Other things are simply, it can almost only be a picture'). The formulations are dialogically expansive, signaling that other ways of communicating have been common and could be possible still.
If a first element of constructed, marketable ordinariness is simplicity, then a second element is the everyday, which is realized by the noun itself and adjectives; that is, 'those little everyday things' (lines 7-8) and the 'broad scope' and 'breadth' of police work (lines 4 and 6). It is also realized by her invoking the context of news and newsworthiness, in relation to which she aligns with their own social media content through positive assessment ('things that wouldn't get any column space anywhere, but are fun to read about'). Hence, the type of content that commonsensically is understood to get column space -the news -is also commonsensically understood to be about serious events and criticized for always dealing with the misery in society. These comparisons indicate that different attitudes (positive/ negative) to media content are at stake, and they invite alignment with the social media practices of the police. Finally, some promotional identity work is going on. She conveys an element of care and compassion by drawing on preventive work (line 5) instead of any actions with the meaning-making potential of casting the police as a coercive force. Signifiers such as animals, children and adolescents perform a similar function through associations with the care and compassion provided by the adult world. Social positions are further produced through promotional discourse, in that the police are represented as aware of and catering to audience tastes (lines 3-4 and 8-9), implying that citizens make up an entertainment-seeking crowd. Moving on to an example from an interview with another communication specialist working for the Norwegian police, this short excerpt presents more evidence of the elements of the everyday and emotions: Ordinariness is invoked through several collectivizing verb clauses, the first one asserting a direct relationship between social media communication and cognition ('Write what we experience'), and the other one utilizing an idiom which also asserts that their social media content emanates from an unfiltered direct source ('we speak straightforwardly'). The original idiom in Norwegian ('snakke rett fra levra') is used in particular to signify someone expressing anger or dissatisfaction. The interviewee thus pronounces the viability of letting feelings show in the social media content. It underlines a pull towards ordinary behavior as opposed to bureaucratic impartiality or rule following. The next example captures a stretch of an interview with a communication officer at one of the municipal fire and rescue services: Excerpt 3 1 IR: Yes. Which social media were you using then? 2 IE: [. . .] And Twitter we mostly really just use as a mere push channel where we 3 give people some insight into the callouts we attend. We post some of the 4 events we are called out to. Also, we use it in a somewhat out of the ordinary 5 way, because we have quite many followers, we have about thirty-five thou-6 sand now or something, and so we can use it for important messages we're 7 announcing, but we are taking care not to dilute the channel with too many 8 informational and moralistic messages, because that's not why people follow 9 us. They would like to get some insight into the somewhat hidden life that 10 firefighters lead and what callouts involve. Therefore it is important to cul-11 tivate the channel a bit in that direction, we think. (Communication officer, local fire and rescue services) The ordinary is again inscribed through discursive choices that convey that their social media content features ongoing realities as seen from within the occupational community, as opposed to extraordinary, superficial or edited one-off events. It is suggested by collectivizing verb clauses featuring the fire and rescue service (e.g., 'we give people some insight into the callouts we attend') and the audience ('They would like to get some insight into the somewhat hidden life that firefighters lead'). Even more discursive choices (e.g., 'we post some of the events we are called out to') convey a direct relationship between social media content and work realities, and exclude the representational act of choosing and creating the content. These articulations of the everyday as seen from within have the ideological effect of conveying authenticity. Moreover, promotional discourse resurfaces in that the respondent commits himself strongly to knowing and meeting audience needs (e.g., 'that's not why people follow us') and by aligning himself with the lifestyle marketing strategy of inviting people to identify with people and environments rather than instructing them and moralizing (lines 7-11). Accordingly, he evaluates 'important' messages dealing with 'out of the ordinary' situations less positively and states that they are used less frequently. The corresponding position of the 'people' is a passive one, at the receiving end when Twitter is characterized as a 'push-channel' (line 2), which refers to the well-known dissemination strategy of marketing, and is somewhat limited to the consumer's binary choices to follow or not follow accounts, and to like or not like contents (lines 5 and 9).
In the next excerpt, another communication staff member at the fire service talks about the communication department's use of social media, beginning with Twitter and then moving on to Facebook: Ordinariness is 'done' from the outset, beginning with the interviewee's self-repair where she categorizes their social media use as sharing 'events', like keeping an 'event log', rather than as an informational activity (lines 2-3). What characterizes any event log is that it does not differentiate very much between events, and if we take into account its meaning in the world of computers, it records all user activity. The event log is therefore a powerful metaphor suggesting that all everyday operations are communicated. What is omitted, then, is the act of representation in content creation. Beyond the construction of the everyday, her account also draws on emotive content that also conveys authenticity. The fire services' posts are described as 'a little bit about the inner life of the fire service, trying to present a bit of a human picture of it' (lines 5-6). The metaphorical term 'inner life' signifies the actions and processes within an organization that are not usually accessible or visible to outsiders. It also conveys processes of consciousness -especially feelings and moods -that take place within a person, and that are also inaccessible to others. Another metaphorical term, the 'human picture', signifies that messages convey an authentic and emotional reality, as opposed to an edited and superficial one. However, in contrast to the construction of marketable ordinariness, this stretch of talk also features accounts of informational social media use (line 4) that appears to be important to the interviewee. And when she describes the fire service's use of Facebook, there is a shift towards talking about more traditional emergency communication (i.e., 'Of course it's a channel we are going to use during emergencies'), to fend off a putative audience's view that social media would be used only for trivial matters. This exemplifies how producing a variety of communication styles involves negotiating professional practices and identities. As the analysis of the above excerpts has shown, marketable ordinariness is marked by distinct discursive elements including simplicity, the everyday, emotion and promotion. Combined they produce values that are fundamentally different from a bureaucracy's traditional demands on, for example, impartiality. Another discursive element, storytelling, provides logic and structure to how these social media features are constructed.

The continuity-creating discourse element -storytelling
Storytelling performs the function of (a) tying together the discursive elements previously analyzed, such as simplicity, emotion, everyday work and promotion, and (b) providing context and continuity with the potential of grabbing and retaining the social media user's attention. The following stretch of talk focuses in particular on the events and contents featured in posts by the police on Facebook. The excerpt presents premises for stories about people who have been caught up in some innocent entanglement that soon ends happily. Later, it is also asserted that several posts can be presented sequentially to create a storyline: Excerpt 5 1 IR: You mentioned something about good news? 2 IE: That's when our Facebook comes in really handy, for example. That's where 3 you can post stories about people who have regained a stolen cell phone or 4 you can post, well, good things. Children who were separated from their 5 parents, but have been reunited. Yeah, things like that. Much of the work 6 done by the canine squad, with horses, that's the type of content that many 7 enjoy reading. And then various units in the building or in the police districts 8 also get to profile themselves through positive stories on Facebook. And then 9 we get many likes, of course. We got a police boat last summer, for example. 10 Then we posted lots of material about the boat and what it does for the Oslo 11 Fjord and for everyone who uses the fjord and islands and all that, and there 12 are certainly many who thought it was fun to read about. And it's also about 13 showing the scope of police work. The police are not only working with drugs 14 or only with home burglaries, the police are working with a lot of things. And 15 we have the opportunity to profile that on our Facebook pages, for example. 16 So we have the aim to get a couple of positive things out every week about 17 work that's been done. And sometimes it becomes a serial, because someone 18 contributes something, questions, and we might add something more, well, 19 yes. (Communication officer, national police organization) Storytelling emerges as yet another discursive element of the strategy of marketable ordinariness. Notice how the colloquial item 'good things' (line 4) as well as 'positive things' (line 16) establishes some boundaries on content, maintaining the objective of promoting a positive image of the police. Likewise, all the examples in the excerpt (lines 3-5, 8-11) describe successful policing. The premises for the stories also construct a sense of ordinariness through easily recognizable lostand-found stories (lines 3-5); stories about animals; and, finally, attractive images, like cruising the fjords in a new boat. These are descriptions of ordinary troubles and amusements that even young children can understand. Note also all the discursive choices that represent promotional discourse. There is the imageconsciousness inherent in the verb 'profiling' (lines 8 and 15); truth claims regarding audiences' content preferences (lines 6-7, 9 and 11-12), which invoke the logic of 'knowing your customer'; and the pronouncement on the viability of strategizing social media efforts by ensuring continual delivery of positive content (lines 16-17). Along with other similar practices, these discursive choices have the ideological effect of refashioning positions, so that the bureaucracy is expected to act more like a brand, with its awareness of image, while citizens are cast as those who 'use or choose' among social media options. This next excerpt comprises further accounts of storytelling. The interviewee, who has worked for many years at an operations center, provides assessments of how the police should communicate in social media:

Excerpt 6
1 IE: Also, we try to have a storyline in the message when there are bigger things. I 2 think it's such a good example, I heard someone saying that Volvo had a 3 terrific car engine and they wanted to go out on Twitter and like speak 4 about their fantastic car. And they came out with the Volvo logo and that 5 the car is so good and there have been no problems and stuff, and no one had 6 any interest in it. The police are here cast as if they have to carefully shape their communication for maximum impact. How their messages are designed plays a much more prominent role than what they say, suggesting that a good product does not generate attention without the right communication backing it up (lines 4-6). A technique he attests to concerns creating a storyline, exemplified by the section about Volvo, which he assesses with tokens of appreciation and enthusiasm ('I think it's such a good example'). Plenty of advice-giving then invites alignment with a particular story. While he casts product-oriented marketing as failing ('no one had any interest in it'), he attests to storytelling and a plain style of language, fashioned via the blue-collar worker's expressive value statement ('it's a damn good Volvo engine') and humorous, counter-expectational message ('nothing to do today either'). Such emotionalized, bottom-up discourse is typical of ordinariness (Higgins, 2008). So too are the final assessments of obligation to avoid 'finely crafted, polished messages' and a 'too rule-based' approach, and instead to show spontaneity ('It should . . . be a bit more direct'). Several discursive choices have the effect of dialogically contracting what is appropriate, for instance the assessments of intention (e.g., 'we try to have a storyline'), of obligation (e.g., 'we shouldn't', 'it shouldn't', 'it should') and generalizations ('That's kind of what life is like on Twitter'). The authorial voice of the interviewee rejects possible alternatives that seem at odds with prevailing wisdom. Hence, somewhat contradictorily, there is a great deal of emphasis on the viability of communicating in a very specific way, customizing messages to audiences' tastes for maximum reach and effect, but also on communicating spontaneously. Thus, he does not support any 'ordinary' work discourse, but a promotional variety. In several cases, quantitative success is drawn upon as a factual token justifying these practices and reinforcing the promotional tendency.

The justificatory discourse element -quantitative success
Justifications for constructing social media content characterized by what this analysis refers to as marketable ordinariness consist of a wealth of assertions about public taste, popularity and quantitative success. Audience sentiments are thus assigned crucial importance. With the analysis of the next excerpts, I hope to shed some more light on this phenomenon, beginning with an account given by a police employee: Excerpt 7 1 IE: Actually, Oslo has engaged in relationship building with its Twitter . . . which is 2 somewhat known for its humor. So they have used it for reputation manage-3 ment really. Somewhat unintentionally, but that's how it turned out. So, well, 4 I don't know if it's the fourth largest Twitter account in Norway or whatever it 5 is, the operations center in Oslo. And that's pretty amazing. And therefore 6 they have been invited to conferences to give lectures in both Sweden and 7 Denmark. Because no one has nearly as many followers, not the 8 Copenhagen police, no one. So they have been quite curious about it.
(Communication adviser, national police organization) The justification of the police's social media activity consists solely of assertions about and positive attitudinal assessments of quantitative measures, and not of quality of content. The interviewee, like any influencer or celebrity, endorses the significance of becoming big in social media. Size implies excellence. There is an ideological effect here of the finality of customer choice. This strong value position is realized by means of comparatives. The interviewee contrasts the police's largest Twitter account with other accounts nationally (lines 4-5) and with other police accounts in Scandinavia, at this point via triadic repetition and superlative formulations (e.g., 'no one has nearly as many followers, not the Copenhagen police, no one.'). Further positive evaluation is offered via the description of invitations to lecture at conferences in neighboring countries, which connotes positive social esteem, a specialist role, and a context of peers offering a like-minded assessment of their social media performance. Attributions of curiosity and lexical choices at the top end of the scale of positive attitudinal value ('pretty amazing') invite sympathy as well. Altogether, these discursive moves serve the semantic function of assessing success. Next, as is recognizable from previous analyses, a respondent aligns himself with the viability of (a) defining successful communication by how happy social media crowds are with it and (b) using emotionally charged messages about the everyday: Excerpt 8 1 IR: Okay. But then, on Facebook, you said that you went out and tried to correct 2 misunderstandings and provide the right information and stuff. Did you get 3 any reactions? 4 IE: We received quite a few comments and lots of shares and lots of likes. So it had 5 great reach. And of course, it was written with the idea that we wanted it to get 6 as good circulation as possible. Therefore we wrote it to, you know, in a way, 7 we tried to appeal to people's feelings so they would get an idea of what it was 8 like for us out on a job, and how this could happen, and why we had to act 9 quickly in an emergency. And people obviously had great understanding.
(Communication officer, local fire and rescue services) To the interviewer's question about the reception of a Facebook post, the respondent shows maximum commitment to reporting success, using factual tokens and upscaling of positive evaluation including modal quantification ('quite a few comments and lots of shares and lots of likes') and intensification ('So it had great reach', 'And people obviously had great understanding'). These attitudinal evaluations provide evidence of just how successful their communication was and, simultaneously, justification for their social media efforts. Moreover, the respondent demonstrates alignment with two very different communicative undertakings that are not easily reconciled. One seeks confirmation from social media audiences ('to get as good circulation as possible'); the other seeks to be true to the business of firefighters ('what it was like for us out on a job'). It is not just any ordinary work experience or explanation that can be shared on social media but, again, specific representations of work that appeal to large social media crowds and their feelings. Thus, quantitative success can be understood as a discursive element that justifies these choices and at the same time reinforces the continued orientation. However, the strategy of marketable ordinariness and its elements that so far appear rather solid, show signs of inconsistencies and dilemmas when certain interdiscursive features are taken into account.

Interdiscursive features
Interdiscursivity is particularly apparent as interviewees' draw on representations of personalized and promotional content as well as bureaucratic information and restraint. The following excerpt offers an example of this, as the interviewee first conveys that employees have little opportunity to express their views and personality through the state agency's social media, but then presents a story, featuring a dialogic counter viewpoint, about a project that allows both greater freedom of expression and even public sector commerce: Excerpt 9 1 IR: What can that mean in concrete terms? 2 IE: That we can never say, 'yes that's good'. That we can almost never say, for 3 example, that we believe that someone else is writing a good post, or we 4 cannot say anything, we cannot have opinions about anything leading in 5 some political direction. We can just add a link to a website we have about 6 it, neutral information about the topic. All this business about being a per-7 sonal presence when in the end you can't say if you are for or against some-8 thing. Of course we can say that we exercise, but we cannot say that we prefer 9 one gym or another that's operating and doing their thing. We cannot do that. 10 Or we cannot say anything about Høie, the politician, or if some youth party 11 or something like that does something, we do not have any opinion about it. 12 In a way it is our position then, as subject to the agency. So, it makes personal 13 presence a little more cumbersome. //laughter// 14 IR: Yes, I can imagine. 15 IE: Yes. So, but this is very fun, it has nothing to do with crisis, but the one called 16 Dine30, our Facebook page, which deals with physical activity, there we have 17 been exempted completely. We do not act in the agency's name. We act as the 18 Dine30-training gang. And they have opinions, they share and they advertise 19 their products. So they have a completely different, special position.
(Communication officer, national health organization) This excerpt features several discursive moves that indicate strong investment in a position supportive of freer and more personal communication on the agency's social media. All the verbal negations of free expression invoke the corresponding positive as a dialogic option, especially when there are so many negations (Martin and White, 2005). Intensifying negations like 'we cannot say anything' and 'we cannot have opinions about anything' (lines 3-4) construct their freedom of expression as being subject to extreme restrictions, and function hyperbolically to convey strong investment in dialogic alternatives. As Martin and White (2005: 142) explain: 'There is a proliferation of options at this maximising end of the intensity spectrum.' She further constructs this communicative restraint as at-issue by never taking responsibility for it. 'We cannot' repeatedly conveys that the cause is some external source outside her control. Not expressing opinions is articulated as a consequence of the identity position of being 'subject to the agency'. Additionally, she draws on contradictory circumstances that have the effect of creating irony. This discursive move contains a two-part concession and counter-proposition: 'All this business [evidence-based proposition] about being a personal presence when in the end [counter] you can't say if you are for or against something.' Similarly, the proposition that 'personal presence is a little more cumbersome' is an understatement that also creates an ironic effect and, in fact, causes laughter (lines 12-13).
These statements have the effect of inviting the interviewee's interlocutor to entertain the possibility of reasonable alternatives, such as the project that she goes on to talk about. This project is immediately constructed as different from the previously defined mode of work, through a discursive counter and positive evaluation: 'So, but [counter] this is very fun [positive evaluation]'. She conveys that the possibilities for communication, with the Facebook page Dine30 (Eng. Your30), expanded significantly, with products of their own, advertising and opinion sharing. Again, the cause of the change is externalized ('we have been exempted completely') and never explained, in keeping with a hierarchical order. So there is a sudden discontinuation of a discourse of bureaucratic restraint. The new opportunities for self-expression, identity and commerce are positively assessed with lexical choices ('fun') and maximizing discourse ('completely different'), conveying enthusiasm and amazement, and therefore also some awareness of the bureaucratic tradition of cautious self-expression and impartiality.
Moving on to an example from an interviewee with one of the communication officers working for the Norwegian police, we see how in this instance the respondent subtly evaluates different departments' social media activity: Excerpt 10 don't use that for here-and-now things, but for example, well, last week we 7 provided information about statistics from the year that had passed, The interviewee uses a metaphor to suggest that the operations center's Twitter activity is too unbridled, and then admits only partial affiliation with the proposition to soften the internal critique ('it can be everything from A to Z, I'd almost say'). By not providing an explanation of what 'everything from A to Z' means, she appears to assume the existence of some mutual, interdiscursive awareness between her and the interviewer. For instance, the operations center's tweets had been perceived as humorous to such an extent that a mid-tier book publisher decided to put out a collection of their funniest tweets. Moreover, this positioning of different social media practices continues with the description of a social media account and practices that she does identify with ('But our Twitter account. . .'), with the example of a recent post consisting of statistics. This stretch of talk thus demonstrates subtle negotiation of a variety of police accounts and contents, including the 'here-and-now things' that transgress discursive boundaries to which others adhere with more traditional, bureaucratic public communication.
In the following excerpt, a communication adviser for a security agency presents disparate social media practices and the matter 'at issue' concerns where to set boundaries for professional and personal content on organizational Twitter accounts for employees: Excerpt 11 communication. However, the director of communication, he shares mostly 10 security stuff. So it became a problem. He couldn't be himself. He was just a 11 security guy. So we created an agency Twitter account a month ago, which we 12 have also recently begun to use more, and will also use in a crisis situation.
(Communication adviser, national security organization) A series of discursive moves constructs a value position favoring the inclusion of some private content on professional social media pages. First, there is the bare assertion ('if you follow me then you know which football team I support') that constructs some personal information as a taken-for-granted feature, and invites others to share his position (see Martin and White, 2005). A linguistic counter then signals that his colleague used social media differently ('However, the communication manager. . .'), followed by negative evaluation of the identity affordances of his content on security, suggesting that some more personal content could have been expected and preferred ('So it became a problem. He couldn't be himself. He was just a security guy'). The solution presented was to create another Twitter account -a diversification of communication options with new opportunities for identity construction. It is also notable how the interviewer's assumption that the organization lacked a Twitter account proved somewhat face-threatening, sparking a response with recurring promotion of their social media use, detailing their number of followers on LinkedIn, positioning the agency as an early adopter of blogs (lines 4-6), and steering the interview back to something positive and content rich. Overall, it is pertinent from the past examples how interviewees' associate social media use with both personalized, promotional discourse and an impersonal, bureaucratic one. It has furthermore been demonstrated how this inconsistency is, in their view, at least temporarily resolved using channel segmentation, again realizing a function of marketing.

Conclusion
This article aimed to analyze whether and how the social media efforts of state bureaucracies entail professional adoption and negotiation of marketized discourses of communication and self-management. Drawing on interviews with communication specialists working in various security and emergency services, the analysis suggests that their accounts of social media use, foregrounding promotional, uncomplicated, seemingly mundane and personable content, can fittingly be conceptualized as marketable ordinariness. What the communication specialists present is not the authentic ordinary per se, because they also assume responsibility for the social media content being positive and crowd-pleasing. As a strategy, this makes available both more commodified and personalized discourse types and identities than are traditionally associated with state bureaucracies.
While organizational research has explained how practices of branding play an increasingly important role in fashioning identity and creating value for organizations (Mumby, 2015(Mumby, , 2016Willmott, 2010), this article has shown how communication specialists, despite working for non-profit public bureaucracies, take on responsibility for creating similar value-creating communication processes. As expected, the material does not lend itself to making claims that bureaucracies have similar financial incentives as business organizations. But their construction of what I term marketable ordinariness still functions as a strategy, as a mode of action to influence others, to achieve at least in part the same intangible values, including trust and engagement, through its logic of calling for the attention of as many people as possible and trying to align them with a positive image of the authorities.
The results indicate both similarities and discrepancies in relation to findings from research examining healthcare communication efforts Harvey, 2015, 2016;Mulderrig, 2011Mulderrig, , 2018. The present analysis demonstrates promotional discourse, but not the same degree of citizen responsibilization. It did examine an account of an initiative to increase citizens' willingness to take care of themselves through 30 minutes of physical training a day (see Excerpt 9), and it represents the type of voluntary prevention that critics of neoliberalism are afraid will weaken collective welfare security and lead to a transfer of responsibility so that a major reason for illness may become neglect of proper self-care. However, beyond this example, accounts of the police and fire services' communication indicate a focus on self-promotion rather than any significant transfer of responsibility to the public. Quite the contrary, the respondents claimed to be cautious about instructional rhetoric, and instead attempted to capture the public's attention through positive, emotive posts and storytelling about everyday events, something akin to lifestyle marketing that attempts to elicit identification with the feelings and values of certain environments, practices and individuals rather than presenting rational arguments.
The study has also demonstrated other discursive practices than are found in research on the public sector and higher education. Discourse analytical studies focusing on universities have found their promotional discourse to convey abstract and visionary values that breathe class, with spacious, immaculate areas, happy people, and assertions of excellence and innovation (Han, 2013;Ledin and Machin, 2015;Ng, 2014). Representing another part of the public sector, and conveying more of a physical-work identity, representatives of the police and fire services emphasize the viability of describing their everyday work and, in contrast with higher education, some grittiness rather than gloss. A telling example could be the police communicator's assessment that it is necessary to 'speak straightforwardly', exemplified by his idea that Volvo would have done better if their commercials featured a car mechanic's straightforward and brash value statement instead of polished informational messages. Together with previous research, the study thus shows that some level of marketization through communication is common across different branches of the public sector, but the co-created strategies and identities do indeed vary from one professional realm to another.
We might also enter into a discussion about the potential for bureaucracies to create new difficulties. To begin with, the article has analyzed interdiscursive features that were noticeable in interviewees' discursive realizations of, on the one hand, the impartial and impersonal attitude of the bureaucracy and, on the other hand, the promotional, emotive and personalized qualities of branding. Several passages that comprised such disjunctive discourses were once again reminiscent of marketization, in that respondents enunciated the viability of diversification -of having different social media accounts linked to their organizations. The most striking example of this came from the health sector, where a department was allowed to engage in a commercial fitness project, even though it was organized entirely under the auspices of a government agency. Hence, the respondents take on the task of managing more customized social media accounts, which reportedly better contribute to representing particular organizational and employee identities, but may have the effect of fragmenting organizational purposes and identity.
In addition, when bureaucracies embrace promotional discourse, it is likely to become more difficult for citizens to understand their communicative purpose (see Fairclough, 1993) and to what extent content is staged. Are dog and horse patrols portrayed in social media because the police have something to inform citizens about, or primarily to gain followers and esteem? Public organizations' attempts to create authenticity and generate goodwill by way of the marketable ordinariness strategy described here can lead to an authenticity paradox due to growing uncertainty among the public about the purpose of the communication from situation to situation. In this way, the analysis also points to discursive practices with ideological effects on social positions, which at least move towards the model of enterprises looking for customers' appreciation, and citizens whose scope of action seems limited to the binary choice of the consumer.
The last significant problem is that the communication specialists' role in making competent and experience-based decisions may be compromised by prioritization of that which generates desired results in quantitative terms (see Du Gay, 2000: 93). While they construct success in social media in terms of quantitative attention, other possible achievements, such as obtaining information or establishing dialogue, are not treated as priorities, despite the latter being qualities many have hoped would be achieved by social media (Brainard and Edlins, 2015;Brainard and McNutt, 2010;Meijer and Thaens, 2013). More importantly, from a critical point of view, in their competitive pursuit of positive feedback en masse, the professionals present some versions of the job and not others, which necessitates selective use of experience, expertise and decision-making skills. They will likely exclude any content that they anticipate could be challenging for the organization or for mainstream culture. The analysis thus suggests that the professionalism of the public sector employee, which is made up of the complex application of rules, experience, competencies and professional discretion (Hoggett, 2006), is to an extent compromised under marketization.
Finally, this study was based on intensive analysis of only 14 interviews conducted in Norway. It would thus be valuable if future studies included larger samples of communication specialists working in the public sector in both Scandinavia and other parts of the world. Furthermore, the emotive-expressive strategy that has been described by professionals in this study and shown to be part of public sector campaigns (Brookes and Harvey, 2015) can also result in a kind of backlash and be used by citizens against organizations (Bell and Leonard, 2018;Rasmussen, 2015). This points to the importance of more closely understanding audience reception of social media content. Hence, an avenue for further research would be to focus on what citizens make of the social media content that professionals such as the interviewees in this study produce.

Funding
The author received no financial support for the research, authorship, and/or publication of this article.