Mediated Interaction in the Digital Age

The Rise of Mediated Interaction

The intention of The Media and Modernity was to develop a way of thinking about the media that was deeply social, or sociological, in character, and that treated the development of communication media as a fundamental and constitutive part of the formation of modern societies. In developing this way of thinking I was not advocating a media-centric approach: on the contrary, I was arguing that communication media are but one of several important factors shaping the formation of modern societies, and that communication media themselves can only be properly understood in sociological terms. I put forward a particular theory of communication media – what I called an ‘interactional theory’ of the media. The basic idea of this theory is that if you want to understand communication media and their impact, you have to analyse them in relation to the kinds of action and interaction which they make possible and create. Communication media should not be analysed on their own, in terms of their intrinsic properties, in the manner of, say, Innis and McLuhan, but should be analysed in relation to the forms of action and interaction which the use of communication media brings into being.

So the first question we need to ask about communication media is this: what is involved in the use of media to communicate and interact with others? To answer this question, we must begin by rejecting the intuitively plausible idea that communication media merely transmit information and symbolic content to individuals whose relations with one another remain essentially unchanged. We must see instead that the use of communication media involves the creation of new forms of action and interaction, new kinds of social relationships and new ways of relating to others and to oneself. So what kinds of action or interaction are we talking about here?

In The Media and Modernity, I distinguished between three basic types of interaction. I still think that was a good theoretical move, but today I would add a fourth type for reasons I’ll explain below.

The first type of interaction that I distinguished in The Media and Modernity is what we could call face-to-face interaction. It has three defining characteristics: it takes place in a context of co-presence, a common spatial-temporal setting; it is dialogical in character, in the sense that it involves, at least potentially, a two-way flow of information and communication; and it involves a multiplicity of symbolic cues – gestures and facial expressions as well as words, smells and touch (at least potentially) and other sounds and visual cues.

The second type of interaction is what I call ‘mediated interaction’ – the perfect example here is the telephone conversation, though I would also include letter-writing and a good deal of what we now do with email. It involves the use of a technical medium of communication which enables information or symbolic content to be transmitted to individuals who are remote in space or time or both. So the spatial and temporal characteristics of mediated interaction are quite different from those of face-to-face interaction. Mediated interaction is ‘stretched’ across space and time, in such a way that individuals can interact with one another even though they don’t share a common spatial-temporal setting. Mediated interaction is dialogical in character, but it generally involves a certain narrowing of the range of symbolic cues. You can see that mediated interaction is structurally different from face-to-face interaction if you compare a telephone conversation with a conversation face-to-face. In a face-to-face conversation you use a wide range of symbolic cues, facial expressions and gestures as well as the spoken word, whereas in a telephone conversation you have only the spoken word and your interlocutor must either speak or issue a constant stream of fillers to reassure you that he or she is still on the line and paying attention – ‘yeah’, ‘uh-huh’, ‘uh-huh’, etc. Without the fillers, the interaction is at risk of breaking down.

The third type of interaction is what I call ‘mediated quasi-interaction’ – this is the type of interaction created by the kind of media that were once called ‘mass communication’, like books, newspapers, radio, TV and so on, though the term ‘mass communication’ is misleading and best put aside. Like mediated interaction, mediated quasi-interaction also involves the stretching of social relations across space and time, and it involves a certain narrowing of the range of symbolic clues. But mediated quasi-interaction differs from mediated interaction in two key respects: first, it is monological in character, in the sense that the flow of communication is largely one-way (hence ‘quasi’-interaction); and second, it is oriented towards an indefinite range of potential recipients – that is, it is open-ended, unlike a telephone conversation which is point-to-point, e.g. oriented to a specific person who is at the other end of the phone line. But the important part of the argument here is to insist that our engagement with media like newspapers and TV is a form of interaction: when you watch television or read a newspaper or a book, you are not just ‘receiving’ or ‘consuming’ a media product: you are entering into a distinctive kind of social interaction with others who are remote in space and perhaps also in time.

These are the three forms of interaction that I distinguished in The Media and Modernity, but now, in the light of the digital revolution and the massive growth of the internet and other forms of networked communication, I would add a fourth type of interaction – what I’ll simply call ‘mediated online interaction’. What I want to capture with this concept is the new forms of action and interaction that have been brought into being by the computer-mediated communication that takes place in online environments. When I say it is ‘computer-mediated communication’ I don’t mean that it is restricted to desktop or laptop computers – it is not the device that matters here, it is the form of interaction that is created by computer-mediated communication. It could take place on a smartphone or tablet or other mobile device – the smartphone is a computer too, and in some ways even more important for understanding the new forms of interaction that are brought into being by computer-mediated communication and their increasingly pervasive presence in everyday life.

So what are the properties of this new form of interaction? Like other forms of mediated interaction, it involves the stretching of social relations across space and time and it involves a certain narrowing of the range of symbolic clues. But it differs from the other two kinds of mediated interaction in two key respects: unlike mediated quasi-interaction, it is dialogical in character; and unlike mediated interaction (e.g. telephone conversations), it is oriented towards a multiplicity of other recipients – it is many-to-many rather than one-to-one.

Social network sites (SNSs) are the perfect setting for this kind of mediated online interaction: on Facebook, Twitter, YouTube and other social media platforms, individuals create or continue social relationships with distant others, some of whom they know from contexts of face-to-face interaction but many of whom they know only through the social media site. A distinctive feature of these sites is that they enable users to make visible not only their profiles but also their social networks, thus enabling individuals to make connections with a multiplicity of distant others that would not otherwise be made (boyd and Ellison, 2008). The character of the relationships they form and the interactions they have with the different ‘friends’ or ‘followers’ they have on Facebook and similar sites are shaped both by the properties of the platform they are using – it’s ‘affordances’, to use Gibson’s felicitous term (Gibson, 1979) – and by the extent to which these relationships and interactions are tied to this medium (your relationship to someone known only through Facebook is quite different from your relationship to someone with whom you also interact face-to-face in the shared locales of everyday life). In other words, Facebook and other SNSs facilitate a distinctive form of social interaction online, creating a constantly expanding network of social relationships characterized by varying degrees of familiarity and tenuousness and by the exchange of symbolic content in multiple formats and modalities – messages, comments, photos, videos, news feeds, etc. – that is made available to others with varying degrees of openness and restrictiveness.

We can therefore specify the four types of interaction and distinguish between them by using four key characteristics: (1) space-time constitution, (2) range of symbolic cues, (3) degree of interactivity and (4) action orientation. Figure 1 summarizes the interactional properties of these four types of interaction using these characteristics. Face-to-face interaction is situated in a context of co-presence and oriented towards co-present others who share the same spatial-temporal locale; it is dialogical in character and can use the full range of symbolic cues. Unlike face-to-face interaction, all three forms of mediated interaction are stretched out in space and possibly also in time, and all involve a certain narrowing of the range of symbolic cues; but in other respects they differ from one another. Both mediated interaction and mediated online interaction are dialogical in character (at least potentially), whereas mediated quasi-interaction is monological; and the action orientation of each is different: mediated interaction is one-to-one, mediated quasi-interaction is one-to-many and mediated online interaction is many-to-many. OPEN IN VIEWER

Having distinguished between these four types of interaction, let me immediately add four important qualifications and clarifications. First, I don’t want to suggest that this typology is fixed for all time, or that the lines of demarcation between the different types of interaction are always clear-cut – they aren’t. Technologies are constantly changing, and as they change so too do our ways of using technologies to interact and communicate with others. Moreover, specific technologies or apps may blur the boundaries between these types of interaction, enabling individuals to interact in ways that blend together the characteristics of different types. Nevertheless, it is helpful in my view to distinguish these different types because it enables us to see that the use of communication technologies involves the creation of new forms of action and interaction, and it enables us to focus our attention on the crucial ways that these forms of action and interaction differ from one another. The aim here is not to present a definitive typology but rather to outline a way of thinking about communication media and their connection to forms of action and interaction that can be developed and revised as communication technologies evolve.

The second point is that, in distinguishing these four types of interaction, I don’t want to suggest that every actual instance of interaction in everyday life will reflect one of these types. In the actual contexts of everyday life, these different types of interaction are often woven together in complex ways and individuals are constantly moving between them, or even interacting in several different ways simultaneously – for example, you may be watching TV and engaging in face-to-face conversation with a friend or family member at the same time, and while you’re doing that, your mobile phone might ring and you might start another conversation that is carried on concurrently. This is, in fact, pretty common: many individuals now live in media-rich environments – what Madianou and Miller call ‘polymedia’ (Madianou and Miller, 2012) – where different communication media exist alongside one another and in relation to one another, and where individuals switch between these media continuously, choosing which medium to use depending in part on the kind of interaction and interpersonal relationship they want to initiate and sustain with distant others. A typology of forms of interaction is not intended to be a description of the actual flow of social life. But one of the merits of a typology is that it enables you to separate out the different forms of interaction which are often woven together in the complex flow of day-to-day life, to analyse their characteristics and to make explicit certain similarities and differences that might otherwise be difficult to see.

The third point is that, while SNSs provide a perfect setting for mediated online interaction, this is not the only form of interaction or communication that takes place on these sites. There is not a one-to-one correlation between a site or platform, on the one hand, and a form of interaction, on the other. Most SNSs are multifaceted platforms that combine different features and apps to enable users to communicate and interact with others in various ways (Baym, 2015: 18). For example, SNSs like Facebook commonly have, in addition to the news feed, a private messaging facility similar to email that allows individuals to send messages to particular individuals or groups – hence mediated interaction and mediated online interaction are facilitated by the same platform and users can move easily between them. Moreover, the news feed itself, together with the online status indicator and the ‘always on’ character of online connectivity, can give users a peripheral awareness of distant others even when they are not interacting with them (Madianou, 2016). Many SNSs are also commercial organizations that generate revenue by harvesting personal data about users and selling this data to advertisers who promote products or services by targeting messages at particular users or sets of users. These messages can appear as banner ads on a user’s home page or as posts in the news feed and may, in terms of appearance, be very similar to posts from other members of a network. Third parties can also create bots that can be plugged into a messaging app like Facebook Messenger, so that when a user clicks on an ad in their Facebook news feed or Instagram feed, a ‘conversation’ with the bot will open in the Messenger app. But in analytical terms, we need to distinguish between promotional communication of this kind, even communication that takes the form of a ‘conversation’ with a bot, and mediated online interaction with the other participants in a network. Promotional communication of this kind is ‘interaction’ in a very specific sense: the user is receiving messages that are targeted at him or her for the purposes of selling products or achieving some other end. It differs from traditional forms of advertising because it takes place online and because it is much more narrowly targeted, since the recipients are selected on the basis of the personal data harvested by the SNS, but in interactional terms it is very similar to a traditional ad: this is promotional communication that is governed by a commercial or other logic and shaped by the aims of third-party organizations who are using networks to pursue their ends. As a form of interaction, it is entirely different from the dialogical interaction between multiple members of a network, even if it appears on the same platform and is embedded in the same feed where mediated dialogical interaction is taking place.

The fourth and final point is that the different types of mediated interaction, and the communication media with which they are interwoven, are themselves always embedded in social organizations of various kinds and, like all social organizations, these organizations are structured in certain ways. Just as face-to-face interaction always takes place in structured contexts or ‘fields’, so too mediated interaction is always embedded in organizations that develop, control and make available under certain conditions the communication media that enable the mediated interaction to take place. Whether these are the telecommunications companies that provide the facilities and networks for the use of telephones or the media organizations that produce and transmit radio and TV programmes or the technology companies that develop and control the platforms that host social media, these organizations are the social infrastructure that makes possible and supports mediated interaction in its various forms – without these infrastructures, or something similar to them, these forms of mediated interaction would not exist. These organizations, and the individuals who own, manage and work for them, have their own interests, priorities and concerns. They provide the institutional bases for the intensive accumulation of resources – economic, symbolic and informational – and for the exercise of power, including the power to control access to the communication channels and networks, whether this is in the form of journalists and editors who decide which stories will appear in the news or in the form of the terms and conditions to which individuals must adhere in joining a network or posting a message online.

The typology I’ve outlined here is similar in some respects to the three forms of communication distinguished by Castells in Communication Power, but it also differs in certain ways (Castells, 2009: 54–5). Castells distinguishes between what he calls ‘interpersonal communication’, ‘mass communication’ and ‘mass self-communication’, where by the latter he means ‘a new form of interactive communication … characterized by the capacity of sending messages from many to many, in real time or chosen time, and with the possibility of using point-to-point communication, narrowcasting or broadcasting, depending on the purpose and characteristics of the intended communication practice’ (Castells, 2009: 55). While Castells’ notion of interpersonal communication corresponds to what I’m calling face-to-face interaction and his notion of mass communication corresponds to what I’m calling mediated quasi-interaction, his notion of mass self-communication merges together two forms of interaction that I would wish to distinguish – what I’m calling mediated interaction and mediated online interaction. The difference between these two forms of interaction lies not in their dialogical character (they are both dialogical or interactive, at least potentially) or in the medium used (they might both involve the use of digital devices like laptops or mobile phones, though mediated interaction can involve the use of other media too, including old-fashioned pen-on-paper) but rather in the action orientation: mediated interaction is oriented towards a specific other, i.e. it is point-to-point, whereas mediated online interaction is oriented towards a plurality of distant others, i.e. it is open-ended. On my account, using email to communicate with a specific other is a form of mediated interaction – no different, in terms of its basic interactional properties, from making a phone call or writing a letter and sending it by post: the latter is much slower, of course, and hence it is stretched out in time as well as in space, but its interactional properties are the same. But this is very different in interactional terms from posting a message on Facebook or a tweet on Twitter or a video on YouTube, where the message or tweet or video is available for a plurality of others to see, respond to, comment on, retweet, share, etc.

The Social Organization of Mediated Interaction

Having distinguished these four types of interaction, I now want to develop this further by analysing the social organization of these different types. In order to do this, I am going to borrow Goffman’s helpful distinction between the ‘front region’ and the ‘back region’ of action (Goffman, 1969: 109ff). Any action or performance takes place within a particular interactive framework which involves certain assumptions and conventions as well as physical and other features – furniture, décor, ‘props’, layout, etc. – that are part of the setting. An individual acting within this framework will adapt his or her behaviour to it, seeking to project a self-image which is more-or-less compatible with the framework and with the impression that the individual wishes to convey. This action framework, and the features that are accentuated by the individuals acting within it, comprise what Goffman calls the ‘front region’. Actions and aspects of the self which are felt to be inappropriate, or which might discredit the image that the individual is seeking to project, are suppressed or reserved for ‘back regions’. In back regions individuals often act in ways that knowingly contradict the images they seek to project in front regions. They also relax and lower their guard – that is, they no longer require themselves to monitor their actions and utterances with the same high level of reflexivity generally deployed while acting in front regions. The back region is typically separated from the front region in some way, so that an individual moving into a back region can assume that the back region will be hidden from the front-region audience. In a restaurant, for example, the front region is the area where the customers sit and eat, and the back region is the kitchen area; the regions are often separated by a swinging door so that customers can’t see what goes on in the back region – here, waiters and waitresses can laugh and joke among themselves, complain about customers and make fun of them, knowing that what they say will not be seen or heard by the customers.

We can use Goffman’s helpful distinction to elaborate the four types of interaction. In the case of face-to-face interaction, the interaction situation is quite simple – we can represent it as in Figure 2. There is one common front region in which individuals interact, and there are various back regions to which these individuals can retreat if they wish. The social organization of mediated interaction looks rather different – Figure 3. Here, the interactive framework is fragmented into two or more front regions that are separated in space and perhaps also in time. Each of these front regions has its own back regions, and each participant in the mediated interaction has to manage the boundary between the front region in which he or she is situated and the relevant back regions. In a telephone conversation, for example, an individual may try to suppress noises arising from the back region, like the comments or laughter of a friend – they could do this by closing a door, or moving into another room, or calling them back when they’ve isolated themselves from potential interference from back-region behaviour. OPEN IN VIEWER

OPEN IN VIEWER

Figure 3.

The social organization of mediated interaction.

Now let’s compare this with mediated quasi-interaction – Figure 4. Here, there is no common front region; rather, there is a production context in which there is a front region and back regions, and there is a plurality of spatially dispersed reception contexts, each of which has its own front and back region. In the case of television, the front region of the production context might be the area of a studio which is covered by the cameras and microphones – for example, the panel of public figures in a current affairs programme like Question Time, as well as the members of the audience who are in the studio and may be given the opportunity to ask questions or make comments from time to time. This front region will be separated off from back regions like the production control room, the make-up and changing rooms, the reception area, etc. Other features of the studio, such as camera crew and teleprompters, are also aspects of the back region as they are visible only to the individuals situated in the production context. The members of the panel and of the studio audience who appear in these front regions of the production context will be visible to individuals who are situated in dispersed contexts of reception – that is, in the many homes, dispersed in space and time, where the programme is being watched. The individuals in the studio will try to manage the way they appear and the way they speak, conscious of the fact that they are being seen and heard not only by the other individuals in the studio but also by many thousands or even millions of distant others, and they will try to prevent their back-region behaviour (or aspects of their back-region behaviour) from appearing in the front region. But the leakage of back-region behaviour into the front region can easily occur – this is a constant hazard that stems from the social organization of mediated quasi-interaction. Often the consequences of such leakage are trivial but sometimes they are not – indeed, sometimes the consequences can be hugely damaging and disruptive. OPEN IN VIEWER

Consider an example. In the run-up to the 2010 General Election in the UK, the then-Labour-leader Gordon Brown was campaigning in Rochdale, near Manchester in the north of England. During a television interview he was challenged by a pensioner, Gillian Duffy, about Labour’s plans to cut the deficit; Brown ignored her on that occasion, but his aides thought it would be a good idea for him to meet her face-to-face, so they arranged a televised walkabout in the neighbourhood of Rochdale where Gillian Duffy lived. Brown met her in the street and nodded courteously as she expressed her concerns about various issues, including her anxieties about immigrants from Eastern Europe – it was classic front-region behaviour intended to show that Brown, a politician not known for his common touch, was listening to ordinary voters. Then Gordon Brown returned to the seclusion of his chauffeur-driven car and closed the door: assuming that he was now in a safe back region, he let down his guard and described Gillian Duffy in unflattering terms. ‘That was a disaster – they should never have put me with that woman’, said Brown. ‘Whose idea was that? It was just ridiculous.’ ‘What did she say?’ asked an aide. ‘Oh everything’, replied Brown, ‘she was just a sort of bigoted woman.’ But what Brown didn’t realize is that he was still wearing the microphone on his lapel and it was still on, so his words were picked up and recorded and then replayed later on radio and television to millions of listeners and viewers. And it got worse. Shortly after the incident, Brown went into a BBC studio for a live radio interview and, without warning, his unflattering remarks about Gillian Duffy were played back to him in the interview itself. As the scale of the PR disaster began to dawn on Brown, he buried his head in his hands, unaware that this interview was also being filmed, and issued a humiliating apology:

Of course I apologize if I have said anything that has been offensive … You’ve got to remember that this was me being helpful to the broadcasters with my microphone on, rushing into the car because I had to get to another appointment. They have chosen to play my private conversation with the person who was in the car with me. I know these things can happen. I apologize profusely to the woman concerned.

Leakage of this kind is very common today – indeed, more common than ever, thanks to the digital revolution. I’ll come back to this in a minute. But before I do this, let’s look at the social organization of mediated online interaction. The social organization of mediated online interaction could be represented as in Figure 5. Here we have a set of individuals interacting with one another in their front regions, which are themselves separated in spatial and perhaps also temporal terms, creating a form of interaction that is stretched out in space and time. Each front region has its own back region, and each participant to the interaction has to manage the boundary between them. The interaction is dialogical (hence the arrows are pointing in both directions) and each individual is linked to others in a network involving multiple participants, so each individual is orienting himself or herself towards multiple others dispersed in space and time. Every individual participating in this interactive situation knows that his or her utterances, expressions and communicative outputs are available to a plurality of distant others, and each knows that these distant others can also contribute to the interaction by posting comments or joining the interaction in some other way. OPEN IN VIEWER

Let’s now return to the phenomenon of leakage. We’ve seen that one common form of leakage is the seepage of content from the back regions into the front regions of mediated quasi-interaction, as in the case of Gordon Brown’s ‘bigoted woman’ comment, but the phenomenon is much more widespread than this and is increasingly linked today to new sets of players who emerge in the networked space of the internet. One set of players here are bloggers like Matt Drudge and celebrity gossip sites like TMZ that are actively involved in leaking rumours and stories about celebrities, public figures and others – and we shouldn’t forget that Matt Drudge played a crucial role in leaking rumours and stories about the Clinton-Lewinsky affair well before the traditional news media were willing to report it. But we’ve also seen the emergence of a new set of players whom we could aptly describe as ‘digital whistleblowers’, like Julian Assange, Chelsea Manning and Edward Snowden. Digital whistleblowers are actors who see the leakage of sensitive information into the public domain as part of their ‘mission’ or self-understanding – that is, it is part of a political project or set of political beliefs. Here leakage is not an accidental or unintended slippage that occurs when the boundaries between back regions and front regions become porous or blurred; rather, it is the intended outcome of an action seeking to expose activities that took place secretly, in back regions that were hidden from view, and the action is animated by a belief that others have a right to know about these hidden activities. So what we see happening in cases like Wikileaks’ disclosure of thousands of confidential US diplomatic cables in 2010 and Snowden’s disclosure of thousands of top-secret NSA documents in 2013 could be represented as in Figure 6. Online sites like Wikileaks leak information into a public domain where it is picked up by mainstream news media and disseminated to a much wider range of recipients through the traditional channels of mediated quasi-interaction. In the case of digital whistleblowers like Manning and Snowden, they can leak either to an online site like Wikileaks, as Manning did, or to mainstream news media, as happened in the case of Snowden, who reached out to Glenn Greenwald, a journalist working for The Guardian, and Laura Poitras, a documentary filmmaker who had a track record of critical work on state surveillance and strong connections with the mainstream media. The first article based on the documents disclosed by Snowden was published in The Guardian on 5 June 2013 and quickly picked up by mainstream media across the world. When the mainstream media pick up and report material leaked in this way, they also endow it with a kind of legitimacy that it wouldn’t have if it were just a story on the Drudge Report or a post on a site like Reddit – in other words, they endow it with a form of symbolic value, to use Bourdieu’s term, that derives from the accumulated symbolic capital of the mainstream media organizations, and this greatly increases the potential symbolic power and disruptive potential of the leak. It gives the leak a kind of credibility and visibility that it would not have had otherwise. OPEN IN VIEWER

However, to do justice to this complex information environment as it is evolving today, we need to elaborate the model. A more accurate representation of information flows in the kind of mixed online-offline media environment – or ‘hybrid’ media system, to use Chadwick’s (2013) term – that we live in today would look like Figure 7. This includes mainstream media organizations like The Guardian, The New York Times, the BBC, CNN, etc., online sites like Wikileaks, the Drudge Report, Reddit, the Huffington Post, etc., and SNSs like Facebook, Twitter and YouTube, and the flow of information between these organizations is two-way. The picture is complicated by the fact that most mainstream media organizations have a substantial online presence today, so they can’t be treated exclusively as offline organizations – they are themselves mixed or hybrid media organizations; but they are mainstream in the sense that they have an established presence and they use traditional media as well as online media. In the online environment, the recipients of information are also often producers of it in numerous ways – they are constantly interacting with others on social media sites, posting comments, responding to the comments of others, and so on. So the flow of information in the online environment is really two-way between online sites and platforms and networked individuals, and indeed in many cases the platforms are the very medium in which much of the mediated online interaction between networked individuals takes place, so the interaction is not so much between networked individuals and platforms but rather between networked individuals via platforms. In the case of mainstream media organizations, the flow of information between producers and receivers is largely one-way, as the notion of mediated quasi-interaction suggests, though the opportunities for recipients to input into mainstream media organizations is now greater than it was in the past thanks to the fact that many of these organizations now have a presence online. OPEN IN VIEWER

The Transformation of Visibility

By distinguishing the new types of action and interaction brought into being by the use of communication media and analysing their distinctive characteristics, we are able to gain a fresh perspective on some of the historical transformations associated with the rise of modern societies. For most of human history, most forms of social interaction have been face-to-face. Prior to the early modern period in Europe, and until quite recently in some other parts of the world, communication and the exchange of information were, for most people, processes that took place exclusively within contexts of face-to-face interaction. Traditions were primarily oral in character and depended for their survival on a continuous process of renewal, though story-telling and related activities, in face-to-face interaction. Forms of mediated communication did exist, of course, but they were largely restricted to political and religious elites who possessed the necessary skills, such as reading and writing, to use these media. But with the rise of the printing industry from the 15th century on, with the expansion of literacy from the 19th century on and with the development of electronic media in the 19th and 20th centuries, face-to-face interaction was increasingly supplemented by various forms of mediated interaction, mediated quasi-interaction and mediated online interaction. Face-to-face interaction remained very important, of course, but the ‘interaction mix’ of social life changed in fundamental ways as these three other forms of interaction became increasingly important and widespread.

As the interaction mix of social life changed, so too did the ways in which individuals appear to and before others. In face-to-face interaction, the individuals who participate in the interaction are immediately visible to one another as they share the same spatial-temporal context and they are all situated in the same front region of interaction. In face-to-face interaction, visibility is tied to the spatial and temporal properties of the interaction situation and is reciprocal in character: each participant is visible to everyone else and all are visible to each. But with the development of communication media, visibility is freed from the spatial and temporal properties of the here and now. In the new forms of interaction created by the use of communication media, the visibility of individuals, actions and events is severed from the sharing of a common locale: one no longer has to be present in the same spatial-temporal context in order to see the other individual or individuals with whom one is interacting or to witness an action or event. Just as the interaction is stretched out in space and time, so too is the field of vision.

But the properties of communication media shape this field of vision in certain ways. In the case of mediated quasi-interaction, visibility is no longer reciprocal in character. The medium changes the ‘directionality’ of vision: the viewers can see the distant others who appear on their TV screens or in other media but the distant others cannot, in most circumstances, see them. Visibility in mediated quasi-interaction is one-way: some individuals can be seen by many others without themselves being able to see these others, while these others can see distant individuals without being seen by them. In the case of mediated online interaction, the directionality of vision is altered in other ways because many participants in the interaction may have means at their disposal to make individuals, actions and events visible to distant others. The simplest and most effective of these means is the camera in their smartphones: this gives every individual who has a smartphone the ability to photograph or video an individual, action or event and make it available, more or less instantly, to a plurality of distant others. This is not the same as the reciprocity of vision that is characteristic of face-to-face interaction because it is not a matter of each participant in the interaction being visible to all others. But it is not the same as the visibility characteristic of mediated quasi-interaction either because the directionality is more complex: many actors in the network can now use the means at their disposal – e.g. their smartphones – to make individuals, actions and events visible to a plurality of distant others. Uni-directionality has been replaced by multi-directionality.

It is difficult to over-estimate the significance of these transformations in the nature of visibility and the ways that they are reshaping the fields of social and political life today. Individuals, actions and events are now visible in ways that they simply were not visible in the past, and anyone equipped with a smartphone has the capacity to make things visible to hundreds or even millions of others in ways that were not possible before. Of course, not everyone avails themselves of this capacity, and not everyone – or every organization – has the same power to make an image or video visible to others: like all social processes, the capacity to make visible, and the extent to which something can be made visible or kept hidden, made available or removed from view, depends on the power and resources that individuals and organizations have at their disposal. But these differentials, important though they are, should not blind us to the fact that the fields of social and political life have been, and are continuing to be, reconstituted by these new forms of interaction and visibility brought into being by the use of communication media. Let me examine this further by focusing on the implications of these changes for the ways in which political power is exercised.

Reconstituting the Political Field

Prior to the development of print and other media, political rulers interacted primarily with other members of the political elite in the relatively closed circles of the assembly or the court. Their visibility was restricted to those with whom they interacted face-to-face in these shared locales of political power. There were occasions when rulers appeared before wider audiences comprising, among others, some of the subjects over whom they ruled. These occasions included major public events such as coronations, royal funerals and victory marches. The pomp and ceremony of such occasions enabled the ruler to maintain some distance from his or her subjects while enabling them temporarily to see and celebrate the ruler’s existence in a context of co-presence. But for most individuals in ancient or medieval societies, the most powerful rulers were rarely if ever seen.

With the development of print and subsequently electronic media, however, political rulers were able to interact with the subjects over whom they ruled and become visible to them in ways that were simply not possible before. The relation between rulers and subjects was increasingly shaped by the characteristics of mediated quasi-interaction, and the political field itself was partly reconstituted by these new forms of action and interaction. The distinctive properties of communication media and the institutions of which they were part shaped the nature of the relation between rulers and subjects and both enabled and constrained the ways in which they were able to interact with one another. On the one hand, rulers were able to use these new media to reach out to the subjects over whom they ruled and address them in particular ways, while on the other hand subjects were now able to see and engage with rulers as never before, though in ways that were limited by the non-reciprocal character of mediated quasi-interaction and the narrowing of the range of symbolic cues. The development of electronic media, of radio and then television, greatly increased the range of symbolic cues and made it possible to compress the temporal delay to close to zero, giving rise to a distinctive form of despatialized simultaneity: distant others could be rendered visible in virtually the same time frame, could be heard at the very moment they spoke and seen at the very moment they acted, even though they did not share the same spatial locale as the individuals to whom they were now visible. Radio enabled the oral quality of the human voice to be encoded and transmitted to a plurality of distant others, while television enabled both oral and visual cues to be recorded and disseminated. With the advent of television, therefore, individuals were able to see persons, actions and events, as well as to hear the spoken word and other sounds, in a way that was similar in some respects to face-to-face interaction but crucially different in other respects: it was stretched out in space, non-reciprocal in character and dependent on a range of technical and institutional considerations.

Among the institutional considerations was the fact that any communication medium of this kind was always part of an organization that involved large concentrations of resources and personnel who had their own interests and aims. Journalists and other media professionals were never neutral conveyors but always interested parties who, in both facilitating and producing mediated communication, also framed it in certain ways, drawing on norms and ethical codes specific to the media field and on priorities specific to their organizations in order to guide their decisions and practices. In conceiving of themselves as media professionals with their own codes of conduct and their own aims, journalists and other practitioners played, and continue to play, a significant role in shaping the ways in which political leaders, actual and aspiring, are able to appear before others and, more generally, shaping what is communicated to whom and what is made visible and what is not. These organizational players become gatekeepers who shape the flows of communication, the ways that messages and symbolic content are presented and the visibility or invisibility of actors in the field.

Part of the importance of the growing role of mediated online interaction in the political field is that it disrupts the settled roles of these institutional gatekeepers. The power of the established media organizations to shape the agenda is disrupted by the emergence of a plethora of new players who are able to use communication media to interact with others while bypassing the established channels of mediated quasi-interaction. Moreover, by enabling any player in the network to communicate with others, the power of media professionals to shape the agenda is attenuated. They must now compete with a growing number of platforms, information sources and actors who are able to communicate to distant others without going through the channels controlled by established media organizations. This is not to say that the new platforms that enable individuals to bypass the traditional media channels are not themselves gatekeepers – they are. They set the rules that determine who can participate and under what conditions, they decide what kinds of communication are permissible and what kinds are not, they moderate – using both human operators and automated processes – to remove content deemed to be offensive or violent or inappropriate in some way, they develop algorithms to anticipate users’ presumed aims and interests based on their previous practices and preferences and they use this form of knowledge to shape the news and other content that appears in users’ news feeds, to make recommendations and to generate advertising revenue, among other things. These processes work in ways that are quite different from the gatekeeping activities of established media organizations – they are based on what Tarleton Gillespie calls an ‘algorithmic logic’ that uses proceduralized choices and accumulated data to automate a proxy of human judgement or unearth patterns across collected social traces, as distinct from the kind of editorial logic that involves the choices of experts who belong to organizations that claim a certain cultural authority (Gillespie, 2014). But they are forms of gatekeeping nonetheless, and they are not peripheral to the activities of platforms but central to them (Gillespie, 2018). And in some ways they are even more consequential than the gatekeeping activities of traditional media organizations because the dominant platforms are so few in number and have such an enormous scale and reach.

President Donald Trump, by choosing to communicate via Twitter where he has more than 50 million followers, made a calculated decision to prioritize mediated online interaction over mediated quasi-interaction as his preferred mode of interacting with citizens and with his political base. This enables him to bypass the established media channels, which he accuses of bias and of peddling fake news, and to say what he wants to say without the framing and commentary of the traditional media gatekeepers – albeit within the constraint of 140 (now 280) characters. It also enables him to insert his messages into Twitter feeds that include tweets and updates from family and friends, a feature that gives him and his messages a kind of mediated intimacy that they would not have if they were being reported by mainstream media channels (Turner, 2018: 148). The fact that Trump’s language is often crude and overstated, that his opinions are strongly expressed and his assertions are blunt, is not necessarily a weakness in the medium of Twitter but could, in the eyes of his supporters, be a strength, as it enables him to come across as an ordinary human being with real feelings and strong beliefs, rather than as a calculating politician who weighs every word carefully and studiously avoids any display of emotion. And by choosing to communicate via Twitter, Trump is able to speak not only to those who actively decide to follow him on Twitter but to a much wider audience too, as the mainstream media scrutinize his tweets with fine-tooth combs and comment on them frequently, thereby giving Trump a great deal of visibility in the mainstream media but on terms of his own choosing.

While the emergence of these new forms of interaction and visibility created by the use of communication media has reconstituted the political field and created new opportunities for both political leaders and ordinary citizens to participate in political life and interact with others, they have also created new obligations and new risks. New obligations because political leaders, actual and aspiring, cannot not adapt themselves to the new forms of mediated action and interaction that now partly constitute the political field. They have to adapt their ways of acting and speaking to fit with these new conditions. They have to give a great deal of attention to their appearance, their dress, their way of speaking and relating to others through mediated forms of interaction – in other words, they have to pay a great deal of attention to what we could call ‘the management of visibility’. In fact, they really have no alternative now: this is not just an opportunity, it is an obligation, something they have to do; it is part of the structure of a field that is partly constituted by mediated visibility. This also partly explains why, in our heavily mediated democracies, individuals who have forged careers in the media – from Reagan to Trump – can move so easily into the political field: it is not just because they are media celebrities, known for their well-knownness, as Daniel Boorstin once put it (Boorstin, 1961: 57), but also because they are well-versed in the arts of mediated visibility. They know how to present themselves to others through the media and communicate effectively with them – for them, the management of visibility through the media has become second nature. As Fred Turner aptly observes, Trump mastered the idiom of mediated authenticity as the host of reality TV’s The Apprentice and then transferred this seamlessly to Twitter, where his petulance, self-congratulatory pronouncements and bombastic attacks on others are seen by his supporters as signs of his authenticity as a person (Turner, 2018: 147). Some commentators lament the extent to which the entertainment industry has infiltrated politics and see this as another expression of a culture that has become obsessed with celebrities, but this overlapping of entertainment and politics is a symptom of a more fundamental structural transformation in which mediated forms of interaction and visibility have reconstituted the political field.

The New Fragility

These new forms of interaction and visibility create new risks too. Today leaders are much more visible than political leaders were in the past, and however much they may try to manage their visibility, they cannot completely control it. So the visibility created by the media can become the source of a new and distinctive kind of fragility in the political sphere.

There are just too many sources of information, and too many ways in which actions or utterances made in the past, or made in back regions that were seemingly closed off from the front regions of mediated interaction and quasi-interaction, can be captured, preserved, transmitted and re-transmitted into front regions where they can disrupt, compromise or undermine the self-images that leaders wish to project.

These forms of disruption, which have existed since the emergence of print and electronic media, have only been exacerbated by the digital revolution – and for two interconnected reasons. First, the digital revolution has made it much easier to capture and preserve actions and utterances and to copy and reproduce images and information. Now everyone with a smartphone in their pocket or their bag is equipped with the means to photograph, video and record the actions, events or utterances that occur in their immediate locale, thereby creating an indelible digital record of phenomena that would otherwise be fleeting and transitory. Everyone using email and participating in social media is creating durable digital content that, as soon as it is sent or posted, escapes from their control. This banalization of recording creates a vast and rapidly expanding reservoir of digital content that can be recovered and transmitted much more quickly and easily than was the case with content recorded in non-digital forms.

Second, the proliferation of online media means that anyone can now transmit (or re-transmit) images, messages and other symbolic content with relative ease: there is no need to try to persuade the institutional gatekeepers of the established media channels to pay attention to and transmit your content. Uploading a video on Facebook or YouTube, or a photo on Twitter or Instagram, couldn’t be easier. This democratization of transmission turns everyone into a potential source of viewable and shareable content (though potential is not the same as actual, of course).

The banalization of recording coupled with the democratization of transmission means that social and political life is now awash with digitized symbolic content that outstrips, at an ever-increasing pace, the ability of any individual or organization to control it. It renders much more complex the power dynamics embedded in the acts of recording and transmitting symbolic content, turning every bystander into a potential witness who has the means at their disposal to provide audio-visual evidence for what they are witnessing – exactly in the way that Ramsey Orta recorded the manhandling of Eric Garner by the NYPD on Staten Island one afternoon in July 2014, capturing the events on his cell phone and enabling millions of others, widely dispersed in space and time, to see and hear Garner being forced to the ground and utter ‘I can’t breathe’ 11 times before he passed out and subsequently died – a tragedy that, thanks in part to the video clip that was uploaded onto YouTube, sparked off protests against police brutality in many cities across the US and helped to fuel the sense of injustice that has underpinned the Black Lives Matter movement.

The new forms of visibility created by the media can be a risk for political leaders too. However much a political leader, actual or aspiring, might seek to manage the way that they appear before the distant others who know them only through mediated forms of interaction, they cannot completely control it: there’s always the risk that something they said or did in the past, or in settings that they regarded as back regions or private spaces, will be recalled and injected into the front regions of mediated interaction in ways and at times that could disrupt, compromise or undermine the self-image they wish to project, as Anthony Weiner, the US Congressman who used Twitter to send a sexually explicit image of his erect penis to a young woman, discovered to his cost, as did Trump when the video recording of his lewd and sexist remarks about women (‘Grab ’em by the pussy’) was released by The Washington Post in the midst of the 2016 US presidential election campaign. However much an individual, group of individuals or organization may think that their actions or conversations are confidential and clothed in secrecy, they may find that information documenting these actions or recordings of these conversations are leaked into the public domain.

This is exactly why phenomena like leaks and scandals have become so significant in the mediated field of modern politics today: they are occupational hazards in the age of mediated visibility. Leaks, as we’ve seen, typically involve the divulging of information about back-region behaviour into the front regions of mediated interaction. Scandals do this too, but scandals involve more than this because scandals presuppose a certain kind of response. Leaks can fall on deaf ears and still be leaks; scandals can be scandals only if the disclosures precipitate some degree of public disapproval or outrage.

Of course, scandals are not new – or, to be more precise, the word ‘scandal’ is not new: it predates the rise of the media by many centuries (Thompson, 2000). Indeed, the word can be traced all the way back to Ancient Greek and early Judaeo-Christian thought. But from the 17th century on, ‘scandal’ was increasingly interwoven with claims and counter-claims articulated in print. By the early 19th century, a new social phenomenon had come into being – what I call ‘scandal as a mediated event’. This modern phenomenon of mediated scandal had a distinctive structure and dynamic: it involved the disclosure through the media of some action or activity which was previously hidden from view, which involved the transgression of certain values and norms and which, on being disclosed, elicited public expressions of disapproval and outrage. Activities that were carried out secretly or in privacy were suddenly made visible in the public domain, and the disclosure and condemnation of these activities in the press served in part to constitute the event as a scandal. Mediated visibility was not a retrospective commentary on a scandalous event: rather, it was partly constitutive of the event as a scandal.

While the 19th century was the birthplace of mediated scandal, the 20th and early 21st centuries would become its true home. Once this distinctive type of event had been invented, it would become a recognizable genre that some would seek actively to produce while others would strive, with varying degrees of success, to avoid. The character and frequency of political scandals varied greatly from one national context to another and from one time period to another. But it is undoubtedly the case that political scandal has become a more prevalent feature of public life in the US, Britain and many modern liberal democracies since the early 1960s – from Watergate in the US and the Profumo scandal in the UK through to the great MPs’ expenses scandal of 2009–10, the sexual harassment scandals that have shaken Westminster and the numerous scandals and potential scandals that continue to swirl around Trump. Almost every country you look at today has its many political scandals, great and small – Brazil is currently in a state of crisis because of a huge corruption scandal, so too is South Africa, and the list goes on. But, at the end of the day, do these scandals matter? Maybe scandals are just the superficial froth of political life, whipped up by unscrupulous tabloid newspapers that want to sell more copies – that is the way they are seen by many. But in my view, this way of thinking about scandals is completely wrong. Scandals are much more important than this. And the reason why scandals matter so much is simple: they matter because they touch on real sources of power.

To understand this, we have to distinguish between two different forms of power, what I call political power and symbolic power, and we have to see that, in modern democratic societies, the exercise of political power actually depends on symbolic power. To exercise political power in modern democratic societies, you have to use symbolic power to win and secure the support of others – in other words, you have to persuade them to support you. And to do this, you need specific kinds of resources: you need resources like prestige, reputation and trust. You need these resources to exercise symbolic power, and since political power depends on symbolic power, you need them to exercise political power too. And this is why scandals matter so much in the world of politics: they matter because they can deplete the resources on which the exercise of symbolic power depends. Scandals threaten to drain away reputation and trust – they don’t necessarily drain away these resources, but they have the capacity to do so. They are potential reputation and trust depleters. But these are exactly the kinds of resources you need if you want to be able to exercise symbolic power, and hence to acquire, exercise and retain political power as well. This is why politicians are so worried about scandals: they know that scandals can damage their reputation and destroy trust, and hence can undermine their ability to acquire and exercise political power.

So if this account is right, it would be quite mistaken to think of scandals as superficial or frivolous distractions from the real substance of politics: scandals matter because they impinge on real sources of power. They tell us something important about the very nature of power and about the way that power is acquired, exercised and lost in the mediated field of modern politics. They help us see that the exercise of political power in modern democratic societies depends on another form of power – symbolic power – which has to be carefully nurtured and protected if an individual wishes to acquire, exercise and retain political power. And they also help us to see that the mediated field of modern politics is a battlefield in which actors are using whatever media channels they have at their disposal – newspapers, TV, Twitter, etc. – to intervene in the field, pursue their agendas, exercise symbolic power and contest the symbolic power of others, seek to tarnish the reputation of others and to protect and defend their own.

Today we live in an age of high media visibility, which means that political life now unfolds in an information environment where the capacity to reveal and conceal, to make things visible and keep other things hidden, is much more difficult to control, creating a permanently unstable arena in which leaks, revelations and disclosures are always capable of disrupting the most well-laid plans. The digital revolution and the proliferation of new networks of communication and information flow have only exacerbated these developments, creating an information environment in which the leakage of information and the disclosure of previously hidden actions and events is a constant risk and threat. In this respect, the digital revolution has not altered my way of thinking about these issues. But it has reminded us all of just how fragile this arena now is, just how fluid the boundary between public and private life has now become and just how important it is for us as social scientists to try to understand this turbulent new world of mediated visibility in the digital age.