2. Participation as user involvement Daniela Landert Abstract: The aim of this chapter is to explore the potential of social media for user participation from different perspectives. This includes technical factors that determine how much participation is possible on a given platform as well as the degree and forms of participation that can actually be observed. The chapter also addresses the relation between user involvement that results from interaction and other involvement strategies that can be found on social media, such as the presence of personal content and language of immediacy. It concludes with a case study that illustrates how different involvement strategies are combined in political communication on a social networking site. 1. Introduction: Interaction, participation and involvement The potential for users to participate in interaction and to contribute content is perhaps the main defining characteristic of social media (see also Hoffmann, Ch. 1, this volume). It presents a contrast to traditional mass media communication, in which communication is typically unidirectional from a professional text producer to a large anonymous mass audience. On social media, anyone is able to participate, to exchange messages and to interact with any number of other users. This chapter explores this potential from different perspectives, focussing on how it relates to user involvement. Three concepts are of central relevance when looking at participation as user involvement, namely interaction, (social) participation and involvement. There is some degree of overlap between these terms and all of them are used in various ways in different research traditions. Interaction, as understood in this chapter, refers to the exchange of messages between participants. In its most basic form, it consists of A sending a message to B and B being able to react to this message in a way that can be perceived by A. In other words, communication needs to be (at least potentially) bidirectional in order to qualify as interaction. Following this definition, a unidirectional communicative framework such as prototypical mass media communication without any opportunities for the addressees to respond to the message would not qualify as interaction. When looking at online settings it makes sense to distinguish further between the potential of interaction - i.e. whether the addressee has the option of responding to a message - and interactions in which this potential is actually realised - i.e. the addressee responds (see also Section 2 on forms and degrees of interaction and Section 3.2 on types of interactivity). From the point of DOI 10.1515/9783110431070-002 In: C. R. Hoffmann and W. Bublitz (eds.). (2017). Pragmatics of Social Media, 31-59. Berlin/Boston: De Gruyter Mouton. EBSC0 Publishing : eBook Collection (EBSCOhost) - printed on 8/21/2019 6:17 AM via MASARYKOVA UNIVERZITA AN: 1595340 ; Bublitz, Wolfram, Hoffmann, Christian.; Pragmatics of Social Media Account: S8431878 32 Daniela Landert view of pragmatics, interaction is the most accessible of the three concepts, since it can be studied directly on the basis of observable communicative exchanges. In some instances, the term participation is used more or less synonymously to interaction in order to refer to the communicative activities of participants in a given communicative situation (see also Dynel, Ch. 3, this volume). Social participation goes beyond mere interaction, though. It involves a certain degree of power, which means that participants not only have the opportunity to exchange messages, but that their messages also have an effect or, more precisely, that participants have influence on social organisation and social processes (Carah and Louw 2015: 231; Stein 2013). This is the reason why increased access to the Internet in general and social media in particular cannot be equated with democratisation. As Tredinnick (2008: 124) points out, new forms of communication technology tend to become subject to the same power relations that used to control older forms. On a very basic level, online access is not available to everyone. In addition, access to social media is sufficient for sharing messages with a large public, but whether these messages are read and whether authors are able to exert influence through them is far from guaranteed. For pragmaticists, social participation is more difficult to investigate than interaction. The social effects of individual communicative exchanges are in most cases not visible immediately. As a consequence, it is hard to know to what extent specific messages allow their authors to participate in a given social context in a meaningful way. Involvement, finally, means that individuals engage with content, typically in a way that affects them emotionally. Again, uses of the term vary and sometimes involvement is used more or less synonymously with participation and interaction, referring to situations in which participants interact actively. However, involvement can also be used more specifically to refer to internal states and emotional engagement of participants, as well as to characteristics of texts that are associated with emotional engagement (for a critical discussion of these different uses of the term, see Caffi and Janney 1994). From a linguistic point of view, the problem with involvement is that we do not have any direct access to the emotional state of interactants. If and how readers are affected by the messages they read is a question that cannot be answered based on the message alone. As a consequence, linguistic studies of involvement tend to focus on characteristics of texts that are associated with involvement, such as involving content and involving and/or involved language. To give an example, involving content may deal with personal stories and emotions with which the addressee is expected to empathise. Involved language includes, for instance, emotionally charged vocabulary, expressions that explicitly refer to the sender and addressee (e.g. first and second person pronouns), expressions of emphasis such as exclamation marks and capitalisations, the use of emoticons, and so on (see also Section 4 below). Interaction, (social) participation and involvement can and do co-occur. For example, we can picture a situation in which users interact with each other on EBSCO Publishing : eBook Collection (EBSCOhost) - printed on 8/21/2019 6:17 AM via MASARYKOVA UNIVERZITA AN: 1595340 ; Bublitz, Wolfram, Hoffmann, Christian.; Pragmatics of Social Media Account: S8431878 Participation as user involvement 33 social media in a personally and emotionally involved way on a political topic that affects them directly. If such interactions gain sufficient momentum they can contribute to the formation of a larger social movement with the power to affect real social change. The Arab Spring is often presented in such a way, although views differ as to how large a role social media actually played for the political movement (see, for instance, Papacharissi and Blasiola 2016; Thurlow 2013: 237). However, this co-occurrence of involvement strategies is by no means a given in social media. Interactions on social media are not always tied to emotional involvement and cases in which they form part of the type of participation that leads to more substantial social change are the exception rather than the rule. Moreover, even interaction in the narrow sense, i.e. with communication taking place in more than one direction, cannot be taken for granted on social media. While this is certainly the prototypical case, there are also instances in which the potential of interaction is not realised. Far from 'going viral', most messages posted by private individuals reach only a small audience and not all of them even receive a response. The relation between social media, new forms of interaction and participation needs careful consideration at various levels. These levels include whether there is technical potential of interaction on a given platform, to what extent users are free to shape their own interaction with others, whether or not users realise the potential of interaction, what content users publish and what effect this has, i.e. whether the content reaches an audience and whether it is able to lead to change. Not all these aspects are equally central to a pragmatic study of social media. Especially the last question concerning the wider social effects of social media participation is one that is more suitably dealt with in the context of social and political sciences. The remaining aspects will be discussed in Sections 2 and 3 of this chapter. Section 4 will deal with strategies of creating audience involvement through content and language, and with how these strategies are related to involvement strategies that are based on interaction. The connection between these different strategies is further illustrated with a case study on the use of social media in political communication in Section 5. 2. Forms of interaction and participation 2.1. The technical potential of interaction Online interaction can take many different forms. One way of classifying these forms is to order them according to the degree to which user contributions are pre-shaped; or, put differently, the degree to which they allow users to contribute their own content. This classification is technical in the sense that it depends on the infrastructure of the social media platform. At one end of the scale, we find forms of interaction in which users do not contribute any content at all. An example of EBSCO Publishing : eBook Collection (EBSCOhost) - printed on 8/21/2019 6:17 AM via MASARYKOVA UNIVERZITA AN: 1595340 ; Bublitz, Wolfram, Hoffmann, Christian.; Pragmatics of Social Media Account: S8431878 34 Daniela Landert this is the tracking of user behaviour on websites, which can be used for arranging content according to what is most popular or liked. The contribution by users is very small in this case; in the case of most read articles on online news sites, for instance, users might not even be aware that their behaviour has an impact on how the content of a website is presented. Still, this immediate technical feedback is more interactive compared to offline media like books or print newspapers, where technical feedback about reader behaviour is restricted to overall circulation or sales figures. At the next level we find interaction of the type of (dis-)liking content, where users consciously submit their feedback by clicking pre-defined buttons. In this case, there is a clearer intention to contribute to the presentation of content (e.g. by increasing the visibility of certain items), but the actual content is still not changed. User polls are slightly more interactive. They present a way of collecting new content, but this content is aggregated over all participating users and participation is restricted to selecting among pre-defined options. User comments are perhaps the most basic form of individual user-generated content. Depending on the platform, there may be restrictions with respect to length and type of content, but the text can be formulated freely and users are free to choose the topic of their contribution, at least as long as they do not violate the guidelines of the platform. Even freer forms of participation can still be integrated on institutional platforms, for instance in the form of blog sections where users can freely contribute their own content. Social media platforms like Facebook, Twitter or YouTube, offer a range of different options for users to publish their own content, for instance in the form of profiles, status updates, messages and media uploads. Finally, at the opposite end of the scale, users are free to set up their own online spaces and publish content independent of any restrictions of existing platforms. Closely related to the purely technical dimension of participation is the question of user policy and regulations concerning the control over content. All major social media platforms have user policy documents that specify the degree and form of acceptable user participation. Violations of such policies can result in deletion of user content, blocking of user access and, in extreme cases, can even lead to legal action. While some common standards have developed over the last two decades, there are still marked differences across platforms in how acceptable content is defined and monitored (Buni and Chemaly 2016). In addition, Stein (2013) shows how policy documents grant users different degrees of control over their content depending on platform. Comparing YouTube, Facebook and Wiki-pedia, Stein finds that Wikipedia is the only platform of the three in which users have dominant control over the platform's content and over user data.1 In addition, However, this does not mean that every user succeeds in adding the content they want. Wikipedia has a user-based editorial system that monitors content quite closely, at least in some cases. Johnstone (2011) compares the representation of different types of EBSCO Publishing : eBook Collection (EBSCOhost) - printed on 8/21/2019 6:17 AM via MASARYKOVA UNIVERZITA AN: 1595340 ; Bublitz, Wolfram, Hoffmann, Christian.; Pragmatics of Social Media Account: S8431878 Participation as user involvement 35 policies are decided "through a process of discussion-based consensus" among the users of Wikipedia (Stein 2013: 367). In contrast, YouTube and Facebook grant their users very little influence over content and policies. Their user policy documents mainly serve to inform users about the company's practices. Thus, social media platforms differ with respect to the degree of participation they grant to users on two levels: the purely technical infrastructure and the policies that regulate how this infrastructure can be used. 2.2. Realising the technical potential: Degrees of interaction The interactive potential of social media platforms is not always fully realised. In highly interactive settings, we would expect replies to occur very regularly and we would also expect to be able to observe long interactions between small groups of participants who develop ideas and arguments over the course of their interaction. However, most studies that investigate the degree of interaction in social media come to the conclusion that such interactions are the exception rather than the rule. The likelihood for messages to receive replies and to lead to interactions varies across different communication forms and platforms. Jucker and Dürscheid (2012: 43) argue that communicative acts are characterised by different uptake expectations depending on the context in which they occur. For instance, a status update on Facebook by a user who posts holiday pictures typically has a very high degree of uptake expectation; the user posting the message expects friends to react to the message by liking and perhaps even commenting on it. The uptake expectations can be further increased when specific friends are addressed in the post, perhaps even with questions. Likewise, on Twitter, uptake expectations can be increased through the use of hashtags and the address of other users with @-phrases (see Jucker and Dürscheid 2012: 48). There are various ways of measuring the degree of interaction. The overall number of users and contributions can serve as a first indication. This can be used to compare different platforms (e.g. number of users on Facebook and on Twitter) as well as sections within platforms (e.g. number of posts on different topics on a given message board; the number of followers on Twitter). Calculating the number of contributions per user can further help identify platforms that have a stable community of regular participants. Taking this approach one step further, grouping users by the number of messages they post is a way to identify user groups with expertise in articles on Pittsburghese across different media, including print newspaper articles, a website and an email discussion forum. She comes to the conclusion that Wikipedia has the tightest editorial constraints on what types of sources are considered acceptable, and it is the only platform in her sample that does not regard personal experience as a valid source of expertise (2011: 12). EBSCO Publishing : eBook Collection (EBSCOhost) - printed on 8/21/2019 6:17 AM via MASARYKOVA UNIVERZITA AN: 1595340 ; Bublitz, Wolfram, Hoffmann, Christian.; Pragmatics of Social Media Account: S8431878 36 Daniela Landert different levels of participation on a single platform. For instance, McDonald and Woodward-Kron (2016) analyse data from an online message board on bipolar disorder by dividing the posts into sub-corpora according to "user's post count at the time of posting" (2016: 162). Posts in the sub-corpus of first posts are contributed by new members, while posts in the sub-corpora of 220-559th and 560+ posts are contributed by veteran members. This allows McDonald and Woodward-Kron to compare the language use of users with different degrees of active participation in the online community. Their findings show that the different roles in the community are directly reflected in different communicative patterns of giving advice, in metadiscourse, and in the construction of identities in relation to bipolar disorder (McDonald and Woodward-Kron 2016). Another way to study the degree of interaction on a given platform is to analyse the length and branching of message threads. In highly interactive exchanges, users engage in interactions that span several turns, with later messages replying to earlier messages. This leads to deep structures with long message threads. In contrast, flat structures are indicative of exchanges with a lower degree of interaction, in which users tend to post messages which start new topics or which only reply to one starting point message (e.g. a newspaper article, a blog post). Marcoccia (2004) is an example of a study that investigates the length of message threads. He studied French-speaking newsgroups in 1997 and found an average length of 5 messages per conversational sequence. Some approaches to interaction further study how closely messages relate to each other. Rafaeli and Sudweeks (1997) propose a three-way distinction. In oneway communication no interaction takes place at all. In reactive two-way communication, the addressee replies to a previous message, taking into account that previous message, but no messages that preceded it. In contrast, in fully interactive communication, the reply to a message takes into account the entire sequence of previous messages and the way in which they were reactive or interactive. In their early study of three online discussion groups in 1993, Rafaeli and Sudweeks (1997) found different levels of interaction in the three groups and they relate the degree of interaction to the degree of user involvement, arguing that high degrees of interaction create engagement. Arendholz (2013) develops her own measure of the degree of interaction on message boards, which is calculated on the basis of the number of messages in a given thread and the number of users who created the messages. Her results show that only a small proportion of threads in her sample qualify as highly interactive according to her measure. Two out of fifty threads contain on average roughly four messages per user, which is the result with the highest interaction quotient of her sample. In contrast, in several of the threads the number of messages is only marginally higher than the number of users, which means that interaction between users is hardly possible. Repeated replies between users would have resulted in much higher numbers of messages per user. EBSC0 Publishing : eBook Collection (EBSCOhost) - printed on 8/21/2019 6:17 AM via MASARYKOVA UNIVERZITA AN: 1595340 ; Bublitz, Wolfram, Hoffmann, Christian.; Pragmatics of Social Media Account: S8431878 Participation as user involvement 37 Similar results were found in studies of user comment sections on online news sites by Kleinke (2010) and Neurauter-Kessels (2013). Kleinke (2010: 202) observes in her data from BBC Talk and the Spiegel Online forum that messages that refer to previous messages usually do not receive a reply by the user who posted the earlier message. This results in largely unidirectional interaction, in which messages sometimes build chains without ambidirectional interaction: User A may post a reply to user B's message, which was posted in reply to user C's message, but C does not reply to B and B does not reply to A (Kleinke 2010: 202-205). A similar observation is made by Neurauter-Kessels (2013), who studies conflic-tive exchanges in user comments of five British online newspapers. All conflictive exchanges start with an initial message that contains a personal attack against other users, journalists or moderators. For four of the five newspapers, the percentage of such attacks that did not receive any response was very high, between 72 % and 88%. Only for one newspaper, the Express Online, a clear majority of attacks received a response, namely 66 %. This was also the only newspaper for which conflictive exchanges regularly reached a length of four messages and more (21 %). Neurauter-Kessels (2013: 228-229) explains this difference with the smaller number of users that are active on the Express Online compared to the other news sites, and the fact that the users who were involved in the conflicting exchanges were all quite active on the site. In other words, the higher degree of interaction is likely to be related to a more close-knit community of users (2013: 263). Likewise, for blogs the general finding is that the interactional potential that is offered by the technical infrastructure is often not fully exploited by users (e.g. Bolander 2013; Hoffmann 2012). Studying data collected from ten different blogs, Hoffmann (2012: 202) observes that "interaction in comments is [...] extremely limited". In his data, most comments are directed at blog authors and if blog authors respond to these comments at all, their comments do not receive any further reply (2012: 202, 211-212). Bolander (2013) made very similar observations in her study of eight diary blogs. Of the 717 reader comments she analysed, 74% were written in response to the blog post (2013: 106), which leaves only about a quarter of reader comments that can potentially be used for longer interactions. She also found that 75 % of readers in her data produced only one comment, and that cases in which readers submitted more than one comment in the same section are rare (2013: 107). The typical patterns of interaction that can be observed consist of two to three turns: a post by a blogger, responded to by a reader comment which, in some cases, receives a reply by the blogger (Bolander 2013: 113). In contrast, longer sequences and interactions between readers are far less common in her data. However, Bolander also observes differences across her data. The type of discourse move has an impact on the degree of interaction; more precisely, disagreements and criticisms are particularly common in reader comments directed towards other readers and they seem to be more likely to trigger further reader comments (Bolander 2013: 128-129, 204; see also Bolander 2012: 1614). In addition, she EBSC0 Publishing : eBook Collection (EBSCOhost) - printed on 8/21/2019 6:17 AM via MASARYKOVA UNIVERZITA AN: 1595340 ; Bublitz, Wolfram, Hoffmann, Christian.; Pragmatics of Social Media Account: S8431878 38 Daniela Landert finds differences between the eight blogs in her sample. In one of the blogs, the largest group of comments are comments by readers in response to a previous reader comment. On this blog, this group accounts for 41 % of all comments -compared to less than 10% for all the other blogs (2013: 114). Bolander relates this difference to group size and participant relationships, especially with respect to how well users know each other. Similar to Neurauter-Kessels' (2013) findings mentioned above, the higher degree of user interaction in Bolander's study occurs in the setting in which the number of active users is smallest. Bolander (2013: 115) is careful to point out that we cannot assume a simple causal relationship between these factors; not all small online communities of users who know each other well show a very high degree of interaction. Still, the available evidence so far indicates that social factors like group size, familiarity of users, the type of discourse move and the interactional norms within a given group play a very important role for the degree of interaction. In sum, the results so far show that it is certainly not enough to look at the technical infrastructure of a given platform in order to judge the degree of interaction between users. While a certain technical infrastructure is required for interaction, such technical potential is not sufficient to ensure that users actually engage in interaction. 2.3. Horizontal and vertical communication When approaching questions of participation, there is another aspect that becomes important, namely who interacts with whom. Chung (2012: 43) introduces the distinction between horizontal and vertical communication. Vertical communication describes interactions between journalists, public figures, and other professional communicators on the one hand, and users on the other. Typical examples are online newspapers, where readers can post comments in reaction to news articles, and social media profiles by politicians and political parties, which allow supporters to follow, like, share and comment on content. In contrast, horizontal communication refers to interaction between (non-professional) users. Typical examples are social media interactions between friends who also interact with each other offline and message boards devoted to a topic where users can meet other users who share their own interests. These two types of social media interaction create different forms of user participation. Vertical communication leads to a form of invited participation in which the previously largely invisible mass audience is given an opportunity to talk back. Motivations for inviting such user participation are manifold. Apart from gaining insights into the opinions and preferences of their audience, social media interaction with and among users can also be used to increase involvement and to create a sense of community, which can lead to a higher degree of identification with (and thus loyalty to) products, companies and political parties. In political campaign- EBSC0 Publishing : eBook Collection (EBSCOhost) - printed on 8/21/2019 6:17 AM via MASARYKOVA UNIVERZITA AN: 1595340 ; Bublitz, Wolfram, Hoffmann, Christian.; Pragmatics of Social Media Account: S8431878 Participation as user involvement 39 ing, this use of social media has been gaining importance over the last decade to an extent that social media competence is now viewed as an important factor in winning elections (see also Section 5). In the case of public service providers, vertical communication can also be a way to fulfil their public service mandate (Enli 2008; Allan and Thorsen 2011: 29). As far as participation is concerned, vertical communication typically takes place on platforms that are maintained by institutions or professional communicators who consequently can control the degree and form of participation that is available to users. Horizontal communication takes place between peers who may or may not know each other outside of social media contexts. In some cases, social media simply provide an additional channel of communication alongside face-to-face interaction, phone conversations, and other forms of mediated communication. In other cases, social media enable users to interact with likeminded others with whom interaction would be unlikely, difficult or even impossible offline. Sometimes such interaction results in the formation of relatively stable groups or communities online, and sometimes interaction takes place in more ad-hoc groups, which are, for instance, based on specific hashtags. This latter option has been characterised as "ambient affiliation" by Zappavigna (2012, 2014; see also Zappavigna, Ch. 16, this volume). Social media can also provide spaces for horizontal interaction in contexts in which free speech is restricted through political control and censorship (see, for instance, Han 2016). While only accounting for a small share of all interactions, this is the dimension of social media that is most clearly tied to social participation and democratic processes and which, therefore, is of high symbolic value. To what extent social media interactions of this kind are actually able to influence political processes remains controversial, though (see Han 2016; Khia-bany 2016; Papacharissi and Blasiola 2016; Thurlow 2013: 237). The distinction between horizontal and vertical communication is analytically useful, but it is important to note that most social media platforms do not fall neatly into one or the other category. Platforms which, at first sight, appear to be designed for vertical communication are sometimes used mainly for horizontal communication among users. User comment sections on online newspapers are an example. While some users explicitly address their comments to journalists, it is quite rare that journalists actually reply to such messages and it is often unclear if the comment is actually read by the intended addressee. In an early study on user comment sections on the New York Times website, Schultz (2000) found that comments were only contributed by readers, and most journalists he interviewed said that they did not even read the user comments. In contrast, interactions between users are more common, although they also tend to be limited to occasional short sequences of messages on most websites. At the same time, platforms which are often associated with interaction among friends, such as Facebook, also have frequent institutional uses by companies, associations and public figures who address their customers, supporters and voters. Such institutional uses of social media often show similar EBSC0 Publishing : eBook Collection (EBSCOhost) - printed on 8/21/2019 6:17 AM via MASARYKOVA UNIVERZITA AN: 1595340 ; Bublitz, Wolfram, Hoffmann, Christian.; Pragmatics of Social Media Account: S8431878 40 Daniela Landert characteristics as user comment sections on news sites. Politicians may address their voters in the first post and users may respond to this by liking, sharing and commenting (vertical communication), but subsequent turns in the interaction in comment sections often take place between users (horizontal communication). In sum, horizontal and vertical communication lead to different forms of participation. Vertical communication is a form of invited participation between professional and non-professional individuals whereas horizontal communication takes place among equals. This distinction plays an important role for the analysis of wider social and political effects of social media interaction. 3. New models for new forms of participation Social media interaction and participation have a number of consequences in different areas. One that has already been mentioned and which will not be further discussed here is their potential to contribute to democratic processes. Other consequences relate to shifts of roles in the widest sense and are often characterised as blurring of boundaries of one type or other. A typical characterisation of this can be found in Chung: "Interactivity fundamentally challenges existing models of communication and blurs the lines between mass and interpersonal, sender and receiver, and traditional and new media" (Chung 2012: 37). It is certainly true that new media have led to various innovations and that they have had a profound impact on some areas of communication and social innovation. However, it is also true that many of these changes are not quite as categorical as might seem at first sight. Moreover, in some cases the speaking of a blurring of lines can block the view for the complexities of specific communicative constellations. Many of the apparently blurred distinctions can be disentangled, sometimes with the help of new models and concepts that are better suited to representing the new communicative forms. Recent years have seen a number of suggestions for such new frameworks, and some of these will be discussed briefly in this section. 3.1. Interpersonal and mass media communication First, there have been suggestions for new models of mass media communication and interpersonal communication. A traditional view of the two types of communication presumes that mass communication is publicly accessible, unidirectional, and addressed at an anonymous mass audience (Luhmann 1995; McQuail 1987: 31-32), while interpersonal communication is interactional and takes place between a closed group of participants. In social media interactions, this binary opposition is clearly not tenable. However, it is important to note that the clear dichotomy of mass and interpersonal communication has been challenged long before social media came into being. As early as 1972, Chaffee talks about the fact that mass media and inter- EBSCO Publishing : eBook Collection (EBSCOhost) - printed on 8/21/2019 6:17 AM via MASARYKOVA UNIVERZITA AN: 1595340 ; Bublitz, Wolfram, Hoffmann, Christian.; Pragmatics of Social Media Account: S8431878 Participation as user involvement 41 personal interaction are interrelated. Mass media try to integrate characteristics of interpersonal communication, for instance through letters to the editor, and people rely on knowledge acquired through mass media when engaging in conversations with others (Chaffee 1972: 95, 114). Referring to media change, he further comments on the fact that the audience also shapes the way in which media develop: "It seems that the influence of the audience on the medium will be at least as important in this case as will the more conventional question of the medium's influence on the audience" (Chaffee 1972: 115). Still, social media present new challenges and make it necessary to reconsider mass and interpersonal communication. Liiders (2008) proposes a model in which she characterises the relationship along two independent axes. The fist axis distinguishes between institutional or professional content on the one hand and de-institutional or de-professional content on the other. The second axis describes whether communication is symmetrical interaction or asymmetrical quasi-interaction. Mass media are characterised as formal/professional content and asymmetrical quasi-interaction. In contrast, social media - called personal media by Liiders - contain de-institutional/de-professional content and can be placed at various points on the interaction axis. While Liiders' model helps describe similarities and differences between mass media and social media, one of its limitations is the fact that it cannot accommodate social media use by institutional or professional users. Another limitation is that it cannot describe settings in which more traditional mass media communication is combined with interpersonal communication. Janoschka's (2004) model of interactive mass communication deals with this aspect. She addresses the fact that many online settings provide at the same time one channel for unidirectional mass communication and other channels for interactive communication between sender and addressee. For instance, a corporate website may include many parts that are unidirectional, providing information about the company and its products. In addition to this, some interactional channels might be provided, too, such as the option of submitting comments online, or of engaging in an online chat with a support agent. In contrast to Liiders (2008), Janoschka's model can account for such cases in which mass communication and interpersonal communication are combined on the same platform without giving up the analytical distinction between the two. Janoschka's model has been further developed for the case of online news sites, which provide a particularly rich example of combining unidirectional mass media communication (e.g. publication of news stories) with various forms of user interaction and user-generated content (Landert 2014a). This revised model takes into account that not all sections of online news sites provide the same options for user interaction. Some online news sites have sections that are devoted to user comments and other forms of user-generated content. In these sections, users can actively participate and may even have a relatively high degree of control over their content, even though it is usually moderated by the news site. At the same EBSC0 Publishing : eBook Collection (EBSCOhost) - printed on 8/21/2019 6:17 AM via MASARYKOVA UNIVERZITA AN: 1595340 ; Bublitz, Wolfram, Hoffmann, Christian.; Pragmatics of Social Media Account: S8431878 42 Daniela Landert time, online news sites always include sections devoted to editorial content, such as news articles, where the control over content lies entirely with journalists and other professional news workers. The relationship between editorial content and user content - in terms of relative space of each and degree of differentiation between the two - is one of the aspects in which online news sites differ from each other quite considerably (Landert 2014a). 3.2. Users and collaborative content Bublitz (2012a, 2012b) addresses another boundary that is blurred in social media, the boundary between readers and writers. He argues that the duality principle of communication is undergoing erosion in computer-mediated communication. Older forms of communication, such as handwritten letters or manuscripts, clearly distinguish between the person who writes and the recipient who reads. In social media communication, these roles are no longer clearly distinguished. Instead, the new concept of the user replaces both writers and readers; users receive and read messages, and they also participate in the interaction more actively. Bublitz (2012b: 155) adopts Eisenlauer and Hoffmann's (2010: 103) distinction of three types of interactivity, cognitive, structural and productive. Cognitive interactivity refers to the cognitive processes of meaning making on the part of the reader which take place in all forms of communication. Structural interactivity occurs whenever readers take an active role in deciding in which sequence they read a given text, for instance by following hyperlinks to read texts in a non-linear fashion. This type of interactivity is typical of computer-mediated texts but not restricted to them. Examples of printed texts in which structural interactivity occurs regularly are, for instance, encyclopedias and gamebooks (Bublitz 2012b: 155). Finally, productive interactivity refers to the option of modifying and supplementing texts. In many online settings, the degree of productive interactivity is much higher than in offline mediated communication. This is the main factor that distinguishes interactivity in online settings from earlier forms of communication, and it is the reason why Bublitz (2012b) argues that participants in online interaction are more appropriately characterised as users than as readers and writers. A related tendency is noted on a slightly different level by Bruns (2008). He looks at collaborative content creation online and speaks of produsers, i.e. users who not only consume media products, but who actively contribute to them by producing their own content. Wikipedia is a prominent example of online content that is produced by its own users. Other examples of "participatory culture" that have been analysed include fan fiction (e.g. Barton and Lampley 2014; Jenkins 1992, 2006) and citizen journalism (e.g. Bruns 2005; Papacharissi 2009). Both forms have had precursors that predate the widespread use of the Internet, but social media have helped them gain momentum to a very considerable extent. From the perspective of pragmatics, such new forms of collaborative content EBSC0 Publishing : eBook Collection (EBSCOhost) - printed on 8/21/2019 6:17 AM via MASARYKOVA UNIVERZITA AN: 1595340 ; Bublitz, Wolfram, Hoffmann, Christian.; Pragmatics of Social Media Account: S8431878 Participation as user involvement 43 creation have not been fully explored yet. The studies that have been carried out so far point to the potential of collaborative content to provide new insights on a number of issues. For instance, Bartlett (2012) presents a case study on the negotiation of the wording of a sentence that caused disagreements on Wikipedia. He draws attention to different ways in which the editors' decisions are influenced by their role as "prosumers" of Wikipedia and points to the need to analyse the negotiation of objectivity and "neutral point of view" in the context of the norms of their community of practice. Page (2013) focuses on the open-ended and collaborative narration on Wikipedia. She analyses article versions from two different language versions of Wikipedia (English and Italian) and two points in time on one topic ("Murder of Meredith Kercher") and shows how the relative prominence of different narratives varies across cultural contexts and how it changes over time. Like Bartlett (2012), she emphasises the importance of context for the analysis of the negotiations of dominant versions of the narrative and she argues that the analysis should not be restricted to the final product (one version of the article) but take into account the production process that is characterised by open-endedness and multiple tellership of the narrative (see also Page, Ch. 19, this volume). 4. Strategies for creating involvement In the previous sections of this chapter we looked at involvement that relied on actual user interaction through a social media platform. However, involvement can also be created in other ways. On the level of content, the topic of a message and the way in which the topic is presented play an important role. A focus on private topics and personal stories tends to create more involvement than abstract and generalised topics. And linguistic immediacy - i.e. language that is typically associated with informal private conversations - can be used to create involvement through language (see, for instance, Chafe 1982; Koch and Oesterreicher 1985; Tannen 1982, 1986). Both these strategies can be used to personalise mass media communication (see Landert 2014b). They foreground the individuals that participate in the interaction and they present them not as professional communicators, but as private individuals. Despite the fact that the author and the addressee are not co-present and that they may not even know each other, the communication shows characteristics that are typical of personal face-to-face interaction between friends, either in terms of topic or language.2 These alternative strategies for creating user involvement are not restricted to social media. For instance, tabloid newspapers are well known for using stories For a detailed discussion of how these characteristics relate to orality and literacy, see Bos and Kleinke, Ch. 4, this volume. EBSCO Publishing : eBook Collection (EBSCOhost) - printed on 8/21/2019 6:17 AM via MASARYKOVA UNIVERZITA AN: 1595340 ; Bublitz, Wolfram, Hoffmann, Christian.; Pragmatics of Social Media Account: S8431878 44 Daniela Landert about the private lives of individuals as well as language that violates the journalistic norms of objectivity and detachment in order to involve their readers (see, for instance, Conboy 2003, 2006; Sparks and Tulloch 2000). Advertisement is an area in which linguistic strategies of involving the addressee are very prominent. Hermeren (1999) argues that print advertising uses questions and forms of direct address as personalisation strategies to "create the impression that the product or service promoted is tailor-made" for the recipient of the mass media message. Similar means of "[creating] the impression of personal communication" were found by Janoschka (2004: 132) in non-interactive forms of web advertising. Likewise, Barron (2012) demonstrates how public information messages make use of various linguistic strategies to create the impression of interaction, even though no interaction was taking place. The linguistic means that she identifies include, for instance, first person pronouns, deictic reference, questions, directives, and lexical references to the target group. Fairclough ([1989] 2001) coined the term "synthetic personalization" to refer to such linguistic means of simulating interaction. He describes synthetic personalisation as a linguistic strategy of mass media: Synthetic personalization simulates solidarity: it seems that the more 'mass' the media become, and therefore the less in touch with individuals or particular groupings in their audiences, the more media workers and 'personalities' (including politicians) purport to relate to members of their audience as individuals who share large areas of common ground. (Fairclough [1989] 2001: 160) Thus, mass media use such linguistic strategies to address their anonymous mass audience as individuals and to simulate interaction linguistically. The three main strategies for creating involvement - interaction, linguistic immediacy, and private content - can be used independently. Business emails, while being interactive, may be formulated in very formal language and often deal with content that has little potential to involve the addressee emotionally. In contrast, a death notice published in a newspaper may present involving content in formal language and with no immediate options for interaction. Linguistic immediacy is often used in print advertising, where the content is not emotionally involving and where interaction is not immediately available. These differences can be visualised with a three-dimensional model, in which each axis stands for one of the three involvement strategies (see Figure 1, based on Landert 2014b). Each communicative act can be placed in this model with respect to how much it makes use of each involvement strategy. A communicative act that uses hardly any involvement strategies - such as a prototypical article in a print newspaper on a hard news topic - is placed at position A in the model. A business e-mail with some degree of interaction, non-private content and no linguistic immediacy can be placed at position B. A death notice would correspond to position C and an interactive print advertisement might correspond to position D. EBSC0 Publishing : eBook Collection (EBSCOhost) - printed on 8/21/2019 6:17 AM via MASARYKOVA UNIVERZITA AN: 1595340 ; Bublitz, Wolfram, Hoffmann, Christian.; Pragmatics of Social Media Account: S8431878 Participation as user involvement 45 Non-private content Language of distance J ✓ High degree of interaction 1 Low degree of _> interaction O ŕ Private content Language of immediacy Figure 1. Visualisation of involvement strategies in a three-dimensional model (based on Landert 2014b: 30) Despite this independence of the three strategies, they are often used in combination on social media platforms. Thus, a prototypical Facebook post deals with a personal story, written in informal and involving language, and posted with the intention of eliciting a reaction quickly in the form of likes, shares or comments. This corresponds to position E in Figure 1. The co-occurrence of strategies is not entirely coincidental. Social media tend to be very supportive of interaction, the quasi-synchronous nature of communication encourages the use of linguistic immediacy (Dürscheid 2007: 38), and the fact that social media are often used for informal interaction between friends and acquaintances means that private content is quite common. In this sense, the three strategies for creating involvement are closely related to social media. However, this does not mean that the technological development of social media is the sole driving force behind increased levels of linguistic immediacy and private content. Instead, we can look at the use of social media for involved interaction as a technically enabled continuation of the much older aim to simulate solidarity and to relate to members of the audience as individuals described by Fair-clough ([1989] 2001). A deterministic approach to media change would assume that technological developments lead to changes in language and communication in a unidirectional way. Contrary to this, there is ample evidence that the relation between technological change, on the one hand, and social practices - including EBSC0 Publishing : eBook Collection (EBSCOhost) - printed on 8/21/2019 6:17 AM via MASARYKOVA UNIVERZITA AN: 1595340 ; Bublitz, Wolfram, Hoffmann, Christian.; Pragmatics of Social Media Account: S8431878 46 Daniela Landert communication -, on the other, is more complex. Herring (2003) sees technology as facilitator of change. New forms of technologically-mediated communication can be "a site for the emergence and evolution of linguistic norms" (Herring 2003: 1). Other supporters of a weak social constructionist perspective argue that developments in technology and in social practices are interdependent, and that social factors often influence technological developments (e.g. Bijker 1997; Bolter and Grusin 1999: 73-77; Bublitz 2012a). Such a perspective sheds new light on the relation between social media and involvement. Rather than assuming that the new technological setting of social media has caused users to communicate in more involved ways, one can argue that the technological innovations enabling social media platforms were, at least in part, motivated by the aim to relate to the mass audience in more personalised and involved ways. Finally, the frequent co-occurrence of involvement strategies is a consequence of genre conventions that have developed. Many social media platforms are strongly associated with the private use by individuals who exchange personal content in informal language. This prototypical use has played an important role in establishing the norms of interaction on these platforms. Today, many social media platforms are also used by professional communicators, such as commercial businesses, organisations, politicians and celebrities (see also Thurlow 2013 for a critical discussion). For instance, Facebook states that there were more than 50 million active business pages on their platform by the end of 2015 (Facebook for Business 2015). There is not a great deal of research so far about the differences between the communicative practices of private individuals and professional communicators on social media platforms. However, it seems that professional communicators adjust to the involved style that is the norm on many platforms. To give an example, Puschmann (2010) analyses the use of interpersonal pronouns on corporate weblogs and identifies possible explanations for their strong presence. Similar to their use in advertising, they fulfil functions that help further the communicative goals of corporate weblogs by personalising the communication between companies and their customers. At the same time, Puschmann (2010: 188) emphasises the role of genre in shaping readers' expectations about the texts. In other words, readers of corporate weblogs who are familiar with the norms of personal weblogs expect to find a certain degree of similarity between the two genres. 5. Political communication on social media Politicians are a group of professional actors for whom social media have gained importance over the last decade. A crucial point in this development was Obama's first presidential campaign in 2008 and his re-election campaign in 2012, in which social media are said to have played a decisive role - to the extent that Obama has been referred to as "the first social media president" (Rutledge 2013). Indeed, his EBSC0 Publishing : eBook Collection (EBSCOhost) - printed on 8/21/2019 6:17 AM via MASARYKOVA UNIVERZITA AN: 1595340 ; Bublitz, Wolfram, Hoffmann, Christian.; Pragmatics of Social Media Account: S8431878 Participation as user involvement 47 election was in part attributed to the fact that he was able to mobilise new segments of voters through the successful use of social media (e.g. Bimber 2014; Enli and Naper 2016; Kaye 2011; Nagourney 2008). Since then proficiency in social media campaigning has become a key skill of any election team, even though only few politicians succeed in drawing attention to their social media presence on a large scale (Nielsen and Vaccari 2013). Larsson (2015: 151) points out that most research on social media use by politicians so far has focused on election campaigns, despite the fact that social media use has long become part of the everyday practices of many politicians. Everyday use of social media can serve different purposes. For instance, politicians regularly use social media to release information directly, without relying on news media as gatekeepers (Riboni 2015: 260), and sometimes strategic communication on social media is even used for agenda building and to influence the press (Kreiss 2016: 1475). How widespread the practice of releasing information on social media is could be seen at the conclusion of the Iran nuclear talks in April 2015. After a series of meetings over several days, the agreement on Iran's future nuclear programme was announced by Iran's Foreign Minister Javad Zarif via Twitter directly from the meeting. His tweet "Found solutions. Ready to start drafting immediately." was quoted verbatim in a breaking news story on BBC News before the press conference could take place (see Figure 2). This example is particularly compelling if one takes into account that Twitter was banned in Iran at the time of this tweet (see Murgia 2016). By releasing the information on his Twitter account, Zarif was able to draw attention to his own role in the successful negotiations. Another function of social media use is to keep in touch with their electorate between election periods, which can make it easier to secure their support during election periods. Larsson (2015: 150) refers to this aspect as "permanent campaigning". A successful election campaign can lead to a very strong emotional response of the supporters (see, for instance, Zappavigna's (2012) analysis of the Twitter response to Obama's first election). The hope is that if voters follow politicians on their social media profiles, then part of this emotion may be kept alive and could even lead to some sort of community building among supporters, which in turn would increase affiliation with the politicians and the likelihood of continued support in the future. 5.1. Case study: The White House on Facebook Obama was the first US president to communicate through Facebook. Less than three months after he assumed office, the page "The White House" was launched on Facebook. This Facebook page is only one of the channels of communication, together with the White House website, which also includes a blog, a Twitter feed, and presence in traditional mass media. It is also important to note that the Face-book page of The White House is not maintained by Obama personally, but by his EBSC0 Publishing : eBook Collection (EBSCOhost) - printed on 8/21/2019 6:17 AM via MASARYKOVA UNIVERZITA AN: 1595340 ; Bublitz, Wolfram, Hoffmann, Christian.; Pragmatics of Social Media Account: S8431878 48 Daniela Landert O +± Foliow Javad Zarif OJZarif Foreign Minister of Islamic Republic of Iran Tehran, Iran 15 FOLLOW IN G 229.7K FOLLOWERS Javad Zarif ©JZarif 29m Found solutions. Ready to start drafting immediately. 4% tl3'9io •ky&s = Sections □□Hnews aA kí Tweets Media i Favorites TOP STORIES ■ ■ 'Solutions1 in Iran nuclear talks Last updated 8 minutes ago Iran's Foreign Minister Javad Zarif says "solutions" have been found to disputes over its nuclear programme with six major powers during marathon talks in Switzerland that could lead to a deal. "Found solutions, ready to start drafting immediately," he said on Twitter after eight days of talks in the Swiss city of Lausanne, A comprehensive nuclear accord is meant to be drafted by the deadline of 30 June. Figure 2. Tweet by Iran's Foreign Minister Javad Zarif (left) and quotation of Tweet in breaking news report on BBC News (right) staff. In addition, Obama has been present on Facebook with his personal profile since November 2015. In what follows, I present a very briefcase study on the Facebook page of The White House. The case study is based on a small set of data consisting of 37 posts that were published by The White House between 8 March and 9 April 2015. The question that I will deal with is how and to what extent the posts create involvement. More specifically, I will look at how the posts make use of interaction, content and language to involve followers of The White House. Undoubtedly, interaction plays a very central role for creating involvement in the collected posts. The posts offer various opportunities for interaction through liking, sharing and commenting. All of these options were used very actively for all of the posts; every post received at least 10,000s of likes, 1,000s of shares and 100s of comments. The post that received most comments contained Easter greetings from the Obama family. It received 83,684 comments and more than 2 million likes within the year following the publication of the post. For technical and ethical reasons, comments to the posts were not collected and analysed separately. However, EBSC0 Publishing : eBook Collection (EBSCOhost) - printed on 8/21/2019 6:17 AM via MASARYKOVA UNIVERZITA AN: 1595340 ; Bublitz, Wolfram, Hoffmann, Christian.; Pragmatics of Social Media Account: S8431878 Participation as user involvement 49 a cursory view at the top-listed comments indicates that some of them are addressed to Obama while others are addressed to the community of followers.3 There is also interaction between followers, with some of the top-listed comments receiving more than 100 replies. This high number of comments and replies suggests that the Facebook page succeeds in creating personal involvement through interaction, even though there is no indication that Obama or his staff reacted to the comments directly. In addition to enabling interaction, the communicative setting of the Facebook page plays a role for creating involvement in two more ways. First, it is relevant that most non-professional users of Facebook use the platform typically to stay in touch with friends and acquaintances. As a consequence, when they follow The White House, political posts are displayed on the users' news feed in between their friends' status updates on private topics. This reception of political posts together with private content is certainly a very important factor in presenting Obama in a personal way and in creating involvement. Second, the platform supports multimodal communication and hyperlinking. All of the posts in my sample contain either an image or a video, most posts contain hyperlinks, and many contain hashtags. In contrast, there is relatively little text included in the posts. The number of words (including hyperlinks and hashtags) ranges from 0 to 185 with an average of 40. Both multimodality and hyperlinking serve to increase involvement. Hyperlinks are a form of structural interactivity (see Section 3) inviting readers to select their own reading paths, to interact with content, and to follow up on stories by visiting other sites. Visual elements are generally associated with a higher degree of involvement compared to text, which is sometimes attributed to a more immediate perception of content (Eisenlauer and Hoffmann 2008: 8) and a stronger reliance on association and higher emotional response compared to text (Kappas and Miiller 2006; Miiller and Kappas 2011). In addition, a study on Obama's Facebook posts during the 2012 election campaign found that posts with photos of Barack Obama had a higher likelihood to receive likes, comments and shares than other posts, and for photos including Michelle Obama or one of their daughters, the effect was even stronger (Gerodimos and Justinussen 2015: 125). This suggests that this type of image can further increase involvement through leading to a higher degree of actual user interaction. The images and videos that are included in the posts from The White House further create involvement through their content and style. Out of the 37 images and videos, 27 showed Barack Obama and in 3 he appeared together with members of his family. Of the 10 remaining images and videos, 2 showed his handwriting and one showed Michelle Obama. This strong visual focus on Obama serves to foreground his person rather than his office. This is even more the case since By default, two "most relevant" comments are displayed directly below each post. How exactly relevance is assessed by Facebook is not clear. EBSC0 Publishing : eBook Collection (EBSCOhost) - printed on 8/21/2019 6:17 AM via MASARYKOVA UNIVERZITA AN: 1595340 ; Bublitz, Wolfram, Hoffmann, Christian.; Pragmatics of Social Media Account: S8431878 50 Daniela Landert Like ' Comment ■ Share Like ■ Commant ■ Share Figure 3. Two playful images posted on The White House Facebook page in April 2015 many of the images are quite informal and playful. To give some examples, one of the images shows Barack Obama with boxing gloves standing next to Michelle Obama who is holding a dumb-bell. In another picture we see Obama from behind sitting next to an Easter bunny in front of the Washington monument (see Figure 3). These are pictures that are designed to show Obama from a private side, rather than in his professional role as President of the United States. Some other pictures show him in professional contexts, but in situations in which it appears that he is not posing for camera. An example of this is a picture that shows him during a visit of a solar energy plant. Neither he nor the person who is with him look into the camera and the picture gives the impression of presenting a glance behind the scenes of the visit. Overall, the images and videos contribute to the creation of involvement by focusing on the person of the private, backstage Obama. In terms of topic and content, most posts deal with current affairs (see Table 1). These posts typically provide updates about ongoing political issues, often with quotes and sometimes with videos from Obama's speeches. This group of posts is least involving, and it shows that despite the more personal and playful nature of some of the content, the main aim of the Facebook presence is still to transport political messages. The group of posts classified as ceremonial deals with anni- EBSCO Publishing : eBook Collection (EBSCOhost) - printed on 8/21/2019 6:17 AM via MASARYKOVA UNIVERZITA AN: 1595340 ; Bublitz, Wolfram, Hoffmann, Christian.; Pragmatics of Social Media Account: S8431878 Participation as user involvement 51 versaries and historically significant events, such as the 50-year anniversary of the Selma to Montgomery marches, the start of a year-long space mission, and St. Patrick's Day. The posts on these topics include quotes by Obama that acknowledge the significance of the event, sometimes taken from speeches. Posts dealing with events at the White House covered the White House student film fest and the five-year anniversary of the West Wing Week, a weekly behind-the-scenes video show of events at the White House. These posts do not focus on Obama but promote events and related web resources. The six Easter posts include Easter wishes by Obama and his family and Easter events taking place at the White House. They are among the posts that present Obama from a more personal perspective. Finally, the three posts classified as "other" include one post on the First Lady's visit to Cambodia, an update of the cover photo, and Obama's basketball tournament bracket. This last post is another example of focusing on Obama in a non-professional role as someone who is interested in basketball and who makes his own predictions about the tournament, noted down in handwriting. Table 1. Content classification of 37 Facebook posts of The White House Topic Number of posts Current affairs 15 Ceremonial 7 Events at White House 6 Easter 6 Other 3 Total 37 Concerning linguistic immediacy, 17 of the posts address readers directly by using second person pronouns, questions and imperatives (see Examples 1 and 2). While the overall language use is not especially informal, there are a few instances of creative word play, such as egg-cellent in the context of an Easter post (Example 2) and Baracketology when referring to Obama's basketball tournament bracket. Such interpersonal and playful language use again contributes to involvement by staging an interaction with readers. In addition 19 of the 37 posts consist mainly of direct quotes, 16 of them quotes from Obama (e.g. Examples 3 and 4). (1) Share the good news: President Obama just announced new steps to train more Americans and veterans for clean-energy jobs —► http://go.wh.gov/YYmtFD #ActOnClimate (4 April 2015) (2) Watch some egg-cellent highlights from today's White House Easter Egg Roll! wh.gov/ EasterEggRoll #GimmeFive (7 April 2015) (3) "Today, the United States—together with our allies and partners—has reached an historic understanding with Iran." —President Obama: wh.gov/irandeal #IranDeal (2 April 2015) EBSCO Publishing : eBook Collection (EBSCOhost) - printed on 8/21/2019 6:17 AM via MASARYKOVA UNIVERZITA AN: 1595340 ; Bublitz, Wolfram, Hoffmann, Christian.; Pragmatics of Social Media Account: S8431878 52 Daniela Landert (4) "Good luck, Captain. Make sure to Instagram it. We're proud of you." —President Obama Congrats to NASA Astronaut Scott Kelly as he takes off for the International Space Station and his #YearInSpace: go.nasa.gov/1 xFtpnu (27 March 2015) The presence of Obama's own voice contributes to involvement to some extent. However, it is important to note that his voice is presented in quoted speech. Outside of quotation Obama is referred to in the third person. This is different on Obama's own Facebook profile, which was created a few months after the end of the period from which I collected data, on 9 November 2015. Here, Obama appears as the first person author, and even comments on this fact in his first post (see Figure 4). This new Facebook presence marks a new degree of involvement. Obama appears as author and speaks about his personal attitudes, opinions and preferences in the first person. He combines his professional role ("I'm heading to Paris to meet with world leaders") with aspects of his private person ("something I try to do at the end of the day before I head in for dinner"), positioning himself as "one of us", so to speak. Readers are addressed directly and invited to share their own content. This new, even more involved channel of interaction was established almost seven years after launching the Facebook page of The White House, which shows that social media use by politicians is still gaining importance. This and similar newly developing forms of interaction, participation and involvement are only waiting to be studied. 6. Conclusion The potential of participation is a characteristic that all forms of social media have in common. However, this potential is realised to different degrees, in different forms, and with different effects across social media platforms. Such differences require careful consideration and critical examination. Perhaps the most important result of research so far is that there is often a considerable discrepancy between the degree of participation that is possible and the degree of participation that is actually realised. Truly collaborative and interactive online environments tend to be the exception rather than the rule. Nevertheless, such cases provide exciting new research opportunities for pragmatics. This chapter also addressed the fact that user involvement can be created in different ways on social media. Interaction is perhaps the most obvious factor in creating involvement, but content and language often play an important role, too. The relation between different strategies of creating user involvement presents promising avenues for future research, especially in the context of commercial and political uses of social media. EBSCO Publishing : eBook Collection (EBSCOhost) - printed on 8/21/2019 6:17 AM via MASARYKOVA UNIVERZITA AN: 1595340 ; Bublitz, Wolfram, Hoffmann, Christian.; Pragmatics of Social Media Account: S8431878 Participation as user involvement 53 ¥**M President Obama Mia November 9- 2015 ■