Speaking into the system: Social media and many-to-one communication

Introduction

Social media complicate the still popular and seemingly simple question of who says what to whom, in which channel and with what effect (Lasswell, 1948). In addition to introducing many-to-many practices of communication alongside one-to-one or interpersonal and one-to-many or mass communication, social and other digital media have made many-to-one communication a constitutive feature of the current media environment. In and of their engagement with digital media, users leave behind bit trails that are accumulated and serve to structure future communications, with or without the users’ knowledge and consent, within and beyond the medium in question. If humans have always been ‘speaking into the air’ (Peters, 1999), we are increasingly also communicating ‘into the system’ (Jensen, 2013: n.p.). This article addresses the resulting metadata as communication, not just technical features of digital communication systems but also records of a specific aspect of communication – metacommunication – that circumscribes and conditions communication as most commonly understood.

The first section summarizes a model of communication that the authors have employed in several recent studies of changing patterns of media use and of the interrelations between one-to-one, one-to-many and many-to-many practices of communication. The model is redeveloped here in a discussion of many-to-one communication and its place in the communicative practices of contemporary publics. The second section introduces the concept of metacommunication to specify different aspects of the process of many-to-one communication. Metacommunication has most often been associated with face-to-face or interpersonal communication but helps to identify distinctive features of computer-mediated communication as well. The third section of the article, further, relates many-to-one communication to the standard transmission and ritual models that have both guided and divided the field for close to half a century. We argue that both models are needed to account for the current configuration of many-to-many and many-to-one communication in social media. In the final section, we return to some of the political and ethical implications of metadata and metacommunication in digital media. By analogy to a key legal principle of the last 800 years – habeas corpus – a complementary principle of habeas data is still in the making.

Prototypes of communication

While proliferating across diverse technologies and services, human communication can still be understood in terms of a relatively small number of prototypes. Figure 1 lays out such a typology (Jensen and Helles, 2011: 519). Along the vertical axis, a distinction is made between interpersonal communication, mass communication and networked communication; along the horizontal axis, the distinction concerns whether or not an immediate contact and transfer of information takes place. Each cell exemplifies different media that currently support each prototype of communication.

OPEN IN VIEWER

Figure 1.

Six communicative practices.

The prototypes are just that – broad categories of communicative interaction that take a variety of concrete and contextualized forms. One illustration is mobile text messaging (Ling and Donner, 2009), which might be thought of as a typically asynchronous form of interaction but which, in many instances, facilitates close-to-synchronous exchanges, for instance, among teenagers about things to do and places to meet. Another illustration is Wikipedia (Lovink and Tkacz, 2011): This online encyclopaedia has been premised on crowdsourcing, but given its few authors and many readers, it rather amounts to an instance of mass communication. As an analytical framework, however, the model captures the various flows of information and communication that increasingly intersect in the same ‘media’ or platforms. Compared to previous analogue technologies, in which one medium would typically afford one communicative practice – telephones for interpersonal communication, radio and television sets for mass communication – digital technologies represent metamedia (Kay and Goldberg, 1999 [1977]) that incorporate these ‘old’ media along with ‘new’ media, such as wikis, chat and online gaming. In view of the growing interpenetration of different media and genres, the model further suggests that it will be increasingly relevant to treat communicative practices rather than media as the central units of analysis in future media and communication studies.

One of the most important research questions of the past decade has been the extent to which the public is coming to depend on networked communication rather than on mass communication: Are the legacy media on which modern democracies have relied as sources of information and deliberation about political and cultural issues being replaced by social media, for better or worse? To begin to assess such a shift, a comparative study of nine European countries departed from the model in Figure 1. On one hand, the findings indicated that despite variations across age groups and other demographic categories, mass media, specifically broadcast television and radio, remained predominant sources of information, both in terms of their general diffusion among European populations and in terms of the time spent by these publics on various media (Helles et al., 2015). On the other hand, the same study identified a range of complex diets of old and new media in different social segments. As recognized also by other current audience and user studies (for one overview, see Webster, 2014), a central task is to tap the many steps and nuances of the ongoing transition to a thoroughly digitalized media environment. In addition to more longitudinal studies, it will be essential to further examine the significant cross-cultural variations (Jensen, 2016) of the flows of information and communication at local, national, regional and global levels.

Still missing in many research designs, including the model in Figure 1, has been an operationalization of the many-to-one communications that users enter into and which affect the one-to-one, one-to-many and many-to-many communications that they later participate in. Many-to-one communication provides a conceptual and empirical interface for studies of the interplay between media systems and individual users: Through minimal acts of logging on, clicking or liking, communicative agents can be seen to contribute to the ongoing structuration of media, always conditioned by the pre-existing structural frameworks of communication and other social interactions (Giddens, 1984). One example of the role of synchronous many-to-one communication is real-time bidding for advertising space, in which the information that is available about users and their trajectories serves to focus and personalize the push of commercial messages. The current automation of the process of matching products and potential consumers has long-term implications for social structuration, ‘something akin to the automation of the financial markets in the 1980s’ (The Economist, 2014: n.p.). A corresponding example of asynchronous many-to-one communication is the accumulation of user preferences over time, which drives the recommendation of genres and titles in systems such as Amazon, Netflix and Spotify. In each case, what occurs is not so much an intentional input by users but an instrumental registration of users – in a minimal sense, what they do with media (Katz, 1959). What users do with media in sequences and networks later comes to inform what the same media, or third parties, do with or to users.

Users necessarily speak into digital systems; whether and how these become systems of surveillance or exploitation is a matter of design and policy. To clarify the nature of the choices involved, it is important to specify conceptually what it means to speak into the system. While familiar, not least, from interpersonal communication research (e.g. Pearce, 2007), the concept of metacommunication also captures important aspects of users’ interaction with digital media (see further Jensen, 2013).

Metacommunication as process and products

It was Gregory Bateson (1972) in particular who developed the concept of metacommunication to capture some of the implicit, yet essential conditions that make communication possible in the first place. Focusing on face-to-face interaction, he suggested that ‘human verbal communication can operate and always does operate at many contrasting levels of abstraction’ (p. 150). Not only does any individual statement carry several potential meanings but also the interpretation of these meanings will influence the further course of a conversation and its outcome. In face-to-face settings, communication evidently includes not just verbal language but also bodily expressions and various aspects of the context of interaction. And even though much of our face-to-face interaction is outside our conscious control, it still qualifies as communication, at least as far as other people are concerned. Humans constantly ascribe meaning to each other as well as to their shared surroundings, nature and culture. As students of Bateson summarized the insight, humans ‘cannot not communicate’ (Watzlawick et al., 1967: 49), productively and receptively.

For analytical purposes, Bateson (1972) identified two aspects of metacommunication. On one hand, all communication depends on codification: conventions of expressing and interpreting facts and ideas. In a pinch, we may ask, ‘what do you mean’, or ‘what do you mean by “facts”’? In most cases, however, communication will proceed smoothly without the need for such metacommunicative interruptions. People are remarkably good at making sense of what they say to each other by relying on repetitions, rewordings, examples, and so on; this also applies to more challenging instances such as humour or irony. On the other hand, metacommunication serves to establish and maintain social relationships between people exchanging both serious and humorous messages. Most communicative interaction is part of ingrained social practices that assign more or less formal private and public roles to the participants, and which manifest in the participants’ turn-taking, openings, responses, elaborations, and so on. If the communicative contract breaks down for a moment, it can mostly be repaired without any need to ask: ‘Who are you?’, ‘Do you know who I am?’ and, by implication, ‘Who are we?’

Whereas Bateson had focused on face-to-face communication, his categories lend themselves to the study of digital media as well. Metacommunication is a constitutive element of the interactions not only between the users of digital media but also between user and medium. Codifications and social relationships are documented and accumulated through the architectures of digital technologies. Metacommunication is a continuous process; it also generates products which, in the case of digital media, are spoken into the system. Whereas these products are most commonly referred to as metadata – data about data – at a technical level of description, they also yield rich information that can be recommunicated to and by system administrators, advertisers, regulators and other stakeholders. The question is who codifies what – and whom – resulting in which social relationships, not just here and now but in the subsequent and frequently unforeseen uses of all these communications and metacommunications.

Because of the flexibility of digital technologies, they give rise to at least two and rather different kinds of metadata or metacommunicative products. First, in the interface between system and user, communicative events are registered at varying levels of granularity, for example, as individual time stamps and as user trajectories across the wider system. Depending on the specific configuration of the system in question, they document important aspects of the conditions under which communications took place in the past and which may be recommunicated with a view to future communications. Second, beyond the exchange of ‘messages’, the interactions between users include statements on both codification and social relationships. Emojis, tags and other commentary contribute, at once, to the meaning of a message and to the experience of self and other among the participants. Together, the programmed outputs of systems and the inputs of users deliver detailed records about the texts as well as the contexts of communication.

The multiple types and levels of communication and metacommunication intersect in digital systems in complex ways. Digital media, further, highlight one key insight of 20th-century philosophy that was summed up in speech-act theory (Austin, 1962) and fed into communication theory: To communicate is to act. In online as in face-to-face exchanges, whenever one says something, one does something, to and with others. Different media, of course, afford different resources for action. The field of communication research was founded, in part, on the concern that mass media and one-to-many communication would uproot traditional societies; the coming of social media and many-to-many communication has reintroduced a widespread notion that new media may create new forms of community. Also many-to-one communication constitutes a form of action. The question is who does what, in and through many-to-one communication, and what forms of community and society may be produced and maintained by many-to-one communication.

Being social

Social media remain under-defined in the literature: ‘No medium is more social than any other medium. But each medium is social in distinctive ways’ (Jensen, 2015). As a working definition, ‘so-called social media are distinguished by their potential for many-to-many communication, drawing on and feeding into networks of one-to-one and one-to-many communication, as well’ (p. 1). Social media participate, alongside other media, in the ongoing structuration of society (Giddens, 1984).

The concrete effects and broader social consequences of new media forms have been studied and evaluated for decades with reference to two models that have typically been thought of as opposites: the transmission and ritual models of communication. James Carey’s (1989 [1975]) benchmark article questioned a dominant understanding in American social science of communication as the transmission of information from senders to receivers in a centralized system of mass communication. Instead, Carey sought to promote an understanding of communication as a distributed ritual – a sharing of meaning and a condition of community. In this spirit, he quoted the pragmatist philosopher, John Dewey: ‘Society exists not only by transmission, by communication, but it may fairly be said to exist in transmission, in communication’ (in Carey, 1989 [1975]: 13f.). Societies exist in the process of communication, not only in its products: attitudes, purchases, votes and other behaviours. Although Carey was careful not to paint the two definitions as a dichotomy, critical media studies have often treated the ritual model as the preferable alternative. In a media environment with many-to-many communications in the ascendance, the ritual model might come closer to its realization. Social media have been said to empower distributed users, perhaps in the longer term shifting the balance between structure and agency. Enough communicators might agree to change the social structures embedding them.

To assess the place of communication in the current balance between agency and structure, it is essential to include many-to-one communication in the examination of communication as both transmission and ritual. ‘Societies exist by transmission as well as in ritual. Whether pushed or pulled, information is transmitted; rituals motivate transmissions’ (Jensen, 2010: 51). Users transmit tags as well as time stamps into the system, thus participating in a wide variety of rituals. Carey (1989 [1975]) had specified the nature of that participation in his frequently cited definition of communication as ‘a symbolic process whereby reality is produced, maintained, repaired, and transformed’ (p. 23). The definition signalled a broadly constructivist notion that realities – in the plural – are articulated, shared and contested in explicit expressions and exchanges. Social media add to the number and nature of voices that may participate in the construction of reality (Berger and Luckmann, 1966). Social media are social to the extent that they include more people as participants in important rituals of communication and as agents in the social construction of reality.

But communication also consists of implicit expressions and exchanges, as noted here with reference to the concept of metacommunication. Many-to-one communication in digital media serves to structure – condition and circumscribe – the rituals that produce, maintain, repair and transform current social realities. Social media are a particularly rich source of data and metadata about the everyday practices in and through which people become and remain social. Like other data, metadata are transmitted; they are accumulated and they carry over to later instances of communication, as exemplified by Facebook’s EdgeRank algorithm (Bucher, 2012). What I see in and perhaps recommunicate from my news feed is the outcome of many previous communications; my current communications and metacommunications prefigure many later communications. They are, at once, structured transmissions and social rituals.

To sum up, social media must be considered social in at least two different, if interrelated, respects. On one hand, social media recognize far more voices than either mass media or interpersonal media; they extend communicative agency to the many. On the other hand, social media enable the exercise of not only communicative but also other forms of agency by the few. Social media are part of a duality of structure (Giddens, 1984), involving all four prototypes of communication: many-to-one, many-to-many, one-to-many and one-to-one. Societies exist in communication, in rituals with many agents. Societies exist by communication, as well as, by transmissions including metacommunications whose products are accumulated in digital infrastructures. Far more than technical units of analysis, bit trails are thoroughly human and social phenomena that are produced, maintained and repaired in individual acts of communication with structural consequences. A communicative conception of what users do with social media and what social media do with users is important for any assessment of how digital systems might be transformed.

Habeas data

Political and ethical deliberations about communication have traditionally centred on the rights and responsibilities of speaking into the air and, increasingly, through technologies. The classic freedoms of expression, assembly, access to information and political organization have been seen to afford self-definition and self-determination for individuals and collectives alike. In and through communication, humans would be in a position to define the ends and means of their existence and coexistence. These ideals have been associated with both ancient and modern technologies and institutions – the forum and the library, the printing press and broadcasting – and they remain a central legacy for social media. In this last part of this article, we briefly address some additional normative implications of many-to-one communication.

Current political and ethical debates on digital media typically centre on issues of privacy and surveillance, as illustrated with reference to themes such as ‘the right to be forgotten’, at least as far as some transmissions and rituals are concerned. As noted by Agre (1994) more than 20 years ago, digital systems have entailed a shift in the understanding of issues of privacy. What used to require dedicated activities of surveillance can now be accomplished through ‘automatic real-time data collection and analysis’ (p. 103) of user actions and other events within digital systems. The final word of the quotation in the last sentence – analysis – is important because it suggests the extent to which the configuration of digital systems prefigures certain kinds of information about who did what. The relationship between data collection and data analysis is not what it used to be: A vision of ‘collect now, analyze later’ is emerging; this approach has been advocated in the growing literature on so-called big data. The suggestion is that instead of sampling data in order to infer to a wider population, research should gather ‘as much [data] as possible, and if feasible, getting everything: N = all’ (Mayer-Schönberger and Cukier, 2013: 26).

A detailed response to such visions has been offered by Helen Nissenbaum (2011) in a theory of contextual integrity, which further encapsulates a communicative conception of the principles and procedures at stake. In sum, Nissenbaum argues that the normative implications particularly of personal information cannot be decided once and for all, according to general terms of service or at the moment of capture. Information travels through unpredictable contexts; it may be accessed and acted upon by third or fourth parties in an uncertain future in equally unpredictable ways. Users cannot – should not – waive their rights to be associated or not with information that they originated or co-created as part of human–machine interactions. It is in both the private and the public interest to regulate the social uses of information according to certain ‘context-specific substantive norms’ (p. 32). To specify these norms, Nissenbaum refers to three parameters – actors, attributes and transmission principles: Who are the subjects, senders and recipients in question; which types of information are part of their transactions; and ‘what are the constraints under which information flows’ (p. 33) – who else might become end users of the information in question? The multi-step flows of information and communication in the digital media environment make public debate about principles and procedures in this domain more important than ever. For now, ‘the dominant approach to […] achieving privacy online is a combination of transparency and choice’. The unrealistic presumption is that complex terms of service will be transparent to users before they go on to accept them. But like information in other markets, such information in digital media remains imperfect. And with a view to future unknown and unknowable contexts, evidently, so-called ‘notice-and-consent, or informed consent’ (p. 34) cannot be considered informed.

In a longer historical perspective, bit trails recall classic legal questions concerning measures ‘for the protection of the person, and for securing to the individual what Judge Cooley calls the right “to be let alone”’ (Warren and Brandeis, 1890: n.p.). In the late 19th century, Samuel D. Warren and Louis D. Brandeis had worried about the impact of ‘instantaneous photographs and newspaper enterprise’ on ‘the sacred precincts of private and domestic life’. Private information and individual identity have remained two sides of the same coin. The phenomenological philosopher, Maurice Merleau-Ponty (1962 [1945]), made the point that whereas people typically say that they have a body, they also are a body (p. 174). Humans have information about themselves and others; they also are information that they themselves and others may access and act upon, depending on the practices and institutions regulating the constitutive representations and interactions.

Habeas data has been advanced, by analogy to habeas corpus, as a legal principle addressing such concerns. Habeas corpus, deriving from the English Magna Carta of 1215 and integral to modern jurisprudence around the world, literally means ‘you may have the body’. The principle has been intended to protect individuals from unlawful imprisonment through the possibility of appeal, and courts are thus required to establish the whereabouts of a prisoner – his/her corporeal existence. Habeas data – ‘you may have the data’ – has been designed, in turn, to give individuals access to registers in which information is held about them and, if relevant, to have the information corrected or removed. The principle has been given several different interpretations and was first implemented in the constitution of Brazil in 1988 (Doneda and Mendes, 2014).

The general idea of habeas data refocuses attention on the fact that people exist and coexist in and by information and communication. Communities are imagined (Anderson, 1991), in part, through metadata and metacommunication. Both online and offline, individuals and societies will be social in distinctive ways, depending on how data about them are documented, processed and applied. Like habeas corpus, habeas data is no absolute principle; societies will only assign certain rights for individuals to have, hold and use their information as well as their bodies. However, compared to the attention that has been given to rights and responsibilities of information access and communicative expression in recent decades, it is high time for media and communication research to conceptualize and examine rights and responsibilities pertaining to metadata and metacommunication. Media users are increasingly speaking into systems that structure their communications and condition their agency also beyond systems of communication.

Conclusion

Social media have been associated above all with the coming of many-to-many forms of communication, but they also depend for their operation on many-to-one communication, the input that the users of digital media leave behind and which serve to structure future communications. In examining the long-term shift from mass communication towards networked communication, research has tended to neglect how the users of social media speak into the system, above and beyond what they say to each other. To qualify and focus on both empirical studies and normative assessments of social media, we suggest the importance of conceptualizing in communicative terms both what users do with digital media and what digital media do with users.

Departing from a model of prototypical forms of communication, this article has reemphasized the role of many-to-one communication in shaping other flows of communication in the digital media environment. The concept of metacommunication helps to cast metadata as one constituent of wider processes of social interaction. Like other media, social media are social in distinctive ways. To recognize the structuring features of digital media, we note, it is necessary to consider communication not only as a ritual but also as a set of transmissions that are fed back into the system. Minimal acts of communication maintain the structures in which other communicative rituals unfold.

We finally argue that the rights and responsibilities of users as citizens should be addressed and elaborated beyond the most common categories of copyright and its reverse: the right not to be associated with particular items of information and instances of communication. The stakes are on the scale of habeas corpus: Habeas data begins to draw attention to information and communication as fundamental conditions of individual existence and social coexistence. Social media provide the latest opportunity for communication research to reflect on the meaning of being social, and for societies to deliberate on the organization and regulation of the means of public communication.