Affordances-in-practice: An ethnographic critique of social media logic and context collapse

Abstract

Drawing on data gathered during ethnographic fieldwork in Mardin, a medium-sized town in southeast Turkey, this article shows that social media users actively appropriate online platforms and change privacy settings in order to keep different social spheres and social groups apart. Keeping different online social contexts distinct from each other is taken for granted as a way of using social media in Mardin. By contrast, social media scholars have extensively discussed the effects of social media in terms of context collapse. The article highlights how context collapse is the result of patterns of usage within Anglo-American contexts and not the consequence of a platform’s architecture or social media logic. It then suggests a theoretical refinement of affordances, and proposes the concept of affordances-in-practice.

Introduction

“Really? Do people in Milan and London only have one Facebook account? Don’t they use pseudonyms or fake profiles? I can’t believe it. How could it be possible?”

A young Kurdish man from Mardin, a medium-sized town in southeast Turkey, in this way expressed surprise at hearing that in other parts of the world, social media users tend to have only one Facebook account that is used mostly under a real name and identity. In Mardin, people often have multiple Facebook profiles used with different groups of people, and can employ them under fake names and pseudonyms. They often create different closed groups within the same account, extensively use the private chat of Facebook for one-to-one communication, or deliberately take advantage of the visibility offered by more public accounts. The public-facing Facebook inhabited by a few hundred friends, relatives, and neighbors is indeed only one kind of place within a configuration of different online environments and scalable socialities (Miller et al., 2016), and it is purposely used to broadcast updates, images, and news to a large audience. The public Facebook wall is ruled by clear social norms. Here, people have produced a clear normativity that fits the expectations of one specific audience, the general public formed by family members, friends, and neighbors. In Mardin, social media users carefully mold their online space in order to keep different social spheres apart. Keeping different online social contexts separate from one another is the normal way of using social media. Interlocutors of my research mastered Facebook privacy settings excellently and often changed them in order to intentionally create online spaces that are divided from each other. In each online context, they follow social norms, values, and codes of behaviors that reproduce and remediate (Bolter and Grusin, 1999) those existing in the social contexts of the offline world.

Media scholars have extensively discussed the effects of social media in terms of context collapse (among others, see Marvin, 2013; Marwick and boyd, 2011; Marwick and Ellison, 2012; Vitak, 2012; Wesch, 2008, 2009), seen as the collapsing of several contexts upon one another (Wesch, 2009). The concept draws from Erving Goffman’s (1959) argument according to which people portray different images of themselves to different audiences in different social contexts. boyd (2002), inspired by the work of Joshua Meyrowitz (1985), introduced the expression collapsed context in her first study on social media and repeatedly used it in other research (among others, boyd, 2008a, 2008b, 2010, 2014). Meyrowitz’s (1985) main goal was to study social interactions in the context of media environments. He argued that media professionals who speak in front of a camera transcend physical boundaries and talk to different publics; as a result, places lose the social significance they had before. Likewise, social media users would be “forced to grapple simultaneously with otherwise unrelated social contexts that are rooted in different norms and seemingly demand different social responses” (boyd, 2014: 31). The concept of context collapse has remained largely unchallenged in the literature on social media. Yet, more recently, Davis and Jurgenson (2014) highlighted the limits and ambiguities of this theory, and proposed a theoretical refinement that could balance the emphasis given to the “normative structure of social media platforms” (p. 479), by bringing into attention the active role of social actors and their context.

In this article, I argue that the uses of social media in Mardin do not give rise to a collapse of context. My ethnographic evidence demonstrates that Facebook users tend to communicate to imagined audiences that already exist in the offline world. My research participants hadn’t come to terms with the emergence of new online spaces that mixed up unrelated social environments. Mardinites, indeed, actively shaped online contexts of interactions, and adjusted their presentation accordingly. There is no collapse of contexts on social media in Mardin. This is the consequence of two main factors: social media users massively change privacy settings and have an active role in creating and crafting different online spaces; the public-facing Facebook resembles the traditional social context of wedding ceremonies, and does not mix together previously divided social groups.

The ethnographic evidence illustrated in this article raises important questions related to the relationship between users and platforms. It shows that the uses of a technology intended by its designers are different from the actual uses, Facebook’s architecture is not an immutable and normative structure, and its usage has to be understood within the cultural specificity of a given context. The collapse of context, indeed, has been portrayed as a consequence of platforms’ affordances, defined as the properties of an environment which make possible and facilitate certain types of practices (boyd, 2014). According to Marwick and boyd (2011), the collapse of social contexts that were previously segmented in the offline world is a consequence of social media’s affordances, such as persistence, visibility, spreadability, and searchability. In this article, I argue that the concept of affordance has often been used to describe situated patterns of usage within particular Anglo-American social contexts, as if they were stable properties of a platform. By contrast, my findings advocate for a perspective that views social media as a set of practices that cannot be defined a priori, and are not predetermined outside of their situated everyday actions and habits of usage (Bräuchler and Postill, 2010; Couldry, 2004; Gray, 2015; Moores, 2005, Postill 2010). I suggest that social media studies would benefit from non-media centric approaches (Couldry, 2012; Moores, 2012; Morley, 2009) that shift their focus from the analysis of a specific architecture and social media algorithm, to the practices of usage within situated environments. Non-media-centric approaches understand social media as intricately related to offline everyday life and can grasp the varieties of practices of social media uses, which largely differ across social and cultural contexts. ¹ Therefore, I aim at contributing to the theoretical conversation about social media by proposing a refinement of the often ambiguous notion of affordance. To emphasize the relational component of affordances and highlight their contextual variation, I formulate the concept of affordances-in-practice.

After introducing the context of the research and the methodology used, I provide an in-depth description of the practices of social media usage in Mardin. I first describe how social media users actively change privacy settings and craft different online private and semi-private environments, and I then outline the public-facing Facebook and its similarities with the traditional social context of wedding ceremonies. In the second part of the article, I propose and discuss the concept of affordances-in-practice.

Methodology

The data in this article come from a long-term ethnographic research into the practices of social media usage in Mardin, a medium-sized town located in the Kurdish region of Turkey, and inhabited by a majority of Sunni Muslim Kurds and Arabs. The research was part of the larger WhyWePost project, based at the University College London (UCL) Department of Anthropology, and dedicated to understanding the uses and consequences of social media around the world (see Miller et al., 2016). The research was carried out over a period of 15 months in 2013 and 2014 among people between the ages of 15 and 35 living in a newly developing area of this fast-changing town. Mardin is a peculiar town in Turkey. It is a multi-cultural and multi-religious city that has developed significantly since the late 1990s. Urbanization, economic development, the expansion of nuclear families, and the extension of education to women have transformed people’s everyday life and their social relations. In Mardin, social media are associated with imaginaries of progress, and embed both the positive and negative ideals associated with it (Costa, 2016). For many young Mardinites, mobile phones and the Internet have become a venue to practice modernity and to distance themselves from the elders and the inhabitants of the less developed nearby towns. ²

As part of my research, I carried out participant observation online and offline, I recorded more than 100 in-depth interviews, and I submitted around 250 questionnaires. Over the 15 months of ethnography, I had conversations with thousands of people about their social media usage, and I became close friends with around 10 families, from a total of approximately 100 individuals. We hang out, drank coffee, and dined together. I often visited them at their home or their relative’s home, and on weekends, we took trips together. Around 200 research participants also became my friends on Facebook. I followed their online activities and compared them with their offline routines and habits. I conducted an analysis of their online visual posts and online conversations throughout the entire fieldwork. Thereby, this article builds on the findings enabled by a long-term ethnographic immersion in people’s everyday life. Facebook was the mostly used social media, and for this reason, it dominates the material presented in this article. WhatsApp was also used, but mainly for one-to-one or small-group communication. Other platforms, such as Twitter and Instagram, were used only by a small minority of university students and youth with a higher education, and have not been included in the analysis.

Crafting online contexts: multiple accounts, fake profiles, private chats, and public Facebook walls

In this section, I illustrate why context collapse is absent from social media in Mardin. I first describe the active molding of private and semi-private online contexts, and then present the public-facing Facebook as the reproduction of a traditional offline environment. Changing privacy settings, opening multiple Facebook accounts, creating anonymous and fake profiles, forming different closed groups within the same account, unfriending people, blocking undesired acquaintances, and largely using the private chat are the natural and presumed way of using Facebook in Mardin. These are not practices of resistance to a rigid architecture that constrains and limits its users. The majority of my research participants didn’t envisage any other ways of using this platform. Consequently, the default settings and the intentions of the designers did not coincide with the actual uses. Mark Zuckerberg’s model of radical transparency (Kirkpatrick, 2011) encouraged people to have only one identity online. He argued, “The days of you having a different image for your work friends or co-workers and for the other people you know are probably coming to an end pretty quickly” (Kirkpatrick, 2011: 199). The platform of Facebook was designed to be public by default, but Mardinites appropriated it in such a way that different social contexts were kept strictly divided. By integrating social media into their everyday lives, Mardinites have produced patterns of usage that reproduce longstanding boundaries between different social contexts of the offline world, and local meanings of private and public. Facebook users have actively constructed online social settings and adapted their self-presentation, language, and content to different audiences. In Mardin, the active molding of online spaces is the “natural” way of using social media. For these reasons, the theory of context collapse does not account for my ethnographic data, and my research findings can be explained only by decentralizing social media in relation to people’s everyday uses. The ethnographic evidence shows that social media exist only through people’s practices, and that digital architectures and their properties cannot be studied, described, and understood outside their situated practices of usage.

In the following pages, I present three stories that illustrate how people use social media in Mardin and take for granted a specific genre of usage (Horst and Miller, 2012) that does not satisfy the theory of context collapse. These stories show that young Mardinites on social media create separated social contexts inhabited by different audiences. Far from being representative of the entire population and their diverse social media practices, these three case studies, however, exemplify common patterns of social media usage. Fatih is a 35-year-old Kurdish man, married with two children, and owner of a small home furniture shop. When I asked him to tell me about his Facebook habits, Fatih appeared impatient. He was eager to reveal to me that he was simultaneously active on 12 different Facebook accounts: one was used for relatives from his hometown, one with work friends, one with friends from school and hometown, one with female friends, one with intimate female friends, one with foreign girls, one was used for online gaming, one for business, one was used under the name of his 4-year-old daughter, one under the name of his 6-year-old son, and two more accounts were allegedly used for politics but he didn’t want to say anything more about it. Fatih used different username and profile pictures with different levels of anonymity in each Facebook profile. He was a master in changing his Facebook’s privacy settings, and he cared deeply about keeping different spheres separated. This case study not only shows that Facebook has not led to the collapse of multiple audiences into a single context (Marwick and boyd, 2011), but it has also made the division of contexts and audiences neater than in the offline world. Fatih used Facebook precisely to keep different social circles apart, and to conceal and display different aspects of himself to different audiences. In the offline world, he would not have been able to create so efficiently such different and contradicting presentations of himself within divided groups of people. Whereas previous works on social media (boyd, 2002, 2008a; Marvin, 2013; Marwick and boyd, 2011; Marwick and Ellison, 2012; Vitak, 2012; Wesch, 2008, 2009) highlighted the potentially limitless and porousness of online audiences, Facebook usage by Fatih shows that this platform can be used to maintain different audiences strictly apart when this is not possible in the offline world.

Besim is an 18-year-old male student in the first year of a vocational school. Besim has three Facebook profiles and one WhatsApp account. In his first Facebook account, the content is visible to more than 1000 Facebook friends: school friends, neighbors, relatives, and strangers he met online while playing games. This profile has his real name and profile picture, and is used to achieve fame and popularity among his friends and family members. He shares a lot of photographs, usually portraying himself alone or with male friends during holiday trips, special dinners, and on special occasions. Images here are normative and conventional: several self-portraits in which he aims to portray himself looking handsome, happy, well-dressed, and surrounded by many good friends. The second account is held under a fake name and is used exclusively with a few hundred school friends, male and female. He changed the privacy settings as not to be found by anybody, and to be free to share funny YouTube videos, anecdotes, and jokes. He uses this account to share memes, songs, music, and to freely enjoy the company of his friends, secured by the privacy assured by the fake name and profile picture. He does not share any photos to avoid being recognized by relatives, acquaintances, and girlfriends although all his Facebook friends know who he is. He shares amusing photos with close and intimate friends on a couple of WhatsApp groups of school friends with whom every day he shares numerous pictures and videos that are funny, irreverent, and disrespectful, such as images of himself smoking, or grimacing under the strain of holding up a dumbbell. When he went for a holiday to the seaside of western Turkey, he shared pictures of himself wearing swimming trunks. These images would have been considered unacceptable and highly disrespectful toward his relatives, if posted on the public Facebook. The third Facebook account is used with a few other close male friends and his girlfriend, who is unaware of the second profile, as she does not approve of him having other girls as Facebook friends. On this account, Besim privately shares attractive self-portrait pictures with his sweetheart, who in turn reciprocates by sending him her self-portraits. Besim changes the privacy settings before posting any photo, as he wants to keep the image with his girlfriend completely private; and, as is often the case, flirtatious relationships are often maintained through the private exchange of best portraits and pictures. This case study also shows the extent to which young Mardinites can easily use Facebook to interact and communicate with people within different and divided social contexts.

Işılay is a 21-year-old unemployed woman who is waiting for a man to ask for her hand in marriage. She is a loyal Facebook user. She spends many hours every day on her small Smartphone talking with friends and strangers, while sitting at home with her mother and younger sisters. She uses Facebook under a fake profile. Her profile picture is the image of a baby she found on the Internet, and her name forms the acronym of a famous Turkish pop singer she is a fan of. Her Facebook friends, who are mainly peers from school and from her neighborhood, know about her real identity, but her parents, relatives, and family’s friends don’t have any clue about her being online. Her parents think that social media are inappropriate for unmarried women. They fear that social media can harm the reputation of their daughter and consequently that of the whole family. To keep her public profile anonymous, Işılay does not post images of herself or of her friends on the wall potentially visible to all. She is also very cautious not to write comments that might reveal her or her friends’ identity, although she has created a closed Facebook group to communicate and share photos with a few of her best friends. She also continuously uses the private chat to talk with them and with strangers from other towns in Turkey who she met online. She loves looking at her relatives’ and friends’ images, and she likes gossiping about their lives. She uses Facebook to gaze at other people’s lives, but she does not want to be gazed by anyone. Işılay, like many other young women from more conservative and religious families, used Facebook to interact with a closed group of friends and to keep her conversations and profile completely private.

In Mardin, a person might have up to 12 Facebook profiles, but people more commonly use two, three, or four Facebook accounts, or create different closed groups within the same account in order to create divided and delimited social contexts inhabited by different audiences. Social media users in Mardin very rarely accept the default privacy settings proposed by Facebook, which promote the public sharing of personal information (Kirkpatrick, 2011). By contrast, they actively create and mold different online spaces and keep different groups of people apart. They produce a configuration of different online contexts that vary in number and kinds of people that inhabit them. They sometimes turn their Facebook wall into a private space to store and share confidential images with their sweetheart or very few selected friends, and largely use the private Facebook chat for one-to-one communication, or create (semi) anonymous Facebook profiles with fake names and fake profile pictures. In so doing, young women from rural or conservative backgrounds escape the control of parents and family; young men better court (and harass) women without being recognized; supporters and sympathizers of leftwing Kurdish groups access political websites and share political content escaping strict State and social surveillance. In Mardin, social media practices vary significantly among genders, ages, and social classes. Yet, everybody is concerned about maintaining an appropriate presentation of themselves in front of different audiences, and is actively involved in following the social norms ruling these different online settings.

There is no collapse of contexts in Mardin because social media users change privacy settings and have an active role in delimiting different online environments. Yet, in the following pages, I show that also the public-facing Facebook, which is usually inhabited by the general public of relatives, family’s friends, and neighbors, does not lead to collapse of contexts. Here, social media users do not flatten pre-existing offline contexts into the same space (boyd, 2010; Marwick and boyd, 2011; Marwick and Ellison, 2012), but rather intentionally re-create a public environment that already existed in the public spaces of the town and people’s imaginary. In the “public” Facebook account, the imagined and actual audience is the online reproduction of a public that has always existed offline, whose practices reflect longstanding habits and meanings of public visibility that are quite different from those existing in Western contexts (Gole, 2002). On the “public” Facebook wall, social media users address the “community” formed by the extended family and its friends, who often gather at weddings ceremonies. Indeed, the conversations and interviews I had with hundreds of social media users show that people have a clear understanding of the audience they are communicating to on the “public” Facebook wall. And the visual analysis of photographic images shared by 200 research participants demonstrates that the online presentation of the self reproduces that same formality of posture and dress that people have at weddings.

Young people post a lot of individual portraits, and love regularly changing their profile pictures every month in order to increase their fame and popularity, and have a high number of “likes” (Costa, 2016). Images are formal, people keep conventional poses, and look straight into the camera lenses. The goal is showing beauty, social and economic success, and conforming to ideal norms of public presentation. Pictures portraying groups of friends are less frequent, and sentiments are rarely expressed, with few exceptions, such as the case of women hugging babies who are members of their own family. People prefer to share informal and spontaneous images with friends in the private space of Facebook and WhatsApp. On the public-facing Facebook, tagging others is not common because it often requires an explicit permission from the person to be tagged. Formality, conventionality, and frontal postures are the result of the attention and care that social media users dedicate to their online public appearance. This aims at impressing others, reinforcing social status, and boosting their reputation.

The public presentation of the self on online spaces, just like at wedding events, produces gossip and comments that have their own life outside the control of their owner and beyond defined places and specific times. Therefore, people aim at creating an image that conforms to the expectations that society holds about the social lives of specific individuals. The imagined audience on “public” social media is not different from the imagined audience at weddings, which for many years has been the only public happening where women and men meet, and where different extended families and networks of friends gather together. They were the only events where women had the opportunity to show themselves in public and where they could be chosen as brides by young men’s mothers, aunts, or older sisters. Even now, with the expansion of new mix-gendered public spaces, such as shopping malls and modern cafes, weddings continue to maintain a crucial social role (Costa, 2016). In the summer, people attend up to 10, 12, 20, or even more weddings where participants like to both gaze at others and be observed. This is the occasion where people get to present an embellished self in public, to appear in fancy clothes, and for women, the place to display sophisticated hairstyles and expensive jewels. At weddings, men and women compete for social recognition and prestige while following traditional codes of behaviors, exactly like on Facebook. My ethnographic data show that in Mardin, weddings constitute a distinct social context on their own, and are not the results of the collapse of delimited social environments as argued by social media scholars (boyd, 2010; Davis and Jurgenson, 2014; Marwick and boyd, 2011; Marwick and Ellison, 2012). Therefore, social media have not led to the raise of context collapse not only because users largely change privacy settings and have an active role in creating and crafting different online private and semi-private environments, but also because the public-facing Facebook is a reproduction of a traditional public context that has existed well before the diffusion of social media.

Affordances-in-practice

In this article, I’ve showed that context collapse, seen as the flattening of multiple audiences into one (Marwick and boyd, 2011), is presented as a theory for understanding the effects of intrinsic properties of a platform and its architecture; however, it is a Western-centric generalization. It has risen from research within particular Anglo-American contexts and has been used to describe universal properties of a social media platform. Yet, these findings also suggest that we need a more accurate and nuanced understanding of affordances. In the literature on social media, this concept has been used to emphasize the power of the architecture and to minimize the agency of the users:

Maintaining distinct contexts online is particularly tricky because of the persistent, replicable, searchable, and scalable nature of networked acts. People do try to segment contexts by discouraging unwanted audiences from participating or by trying to limit information to make searching more difficult or by using technologies that create partial walls through privacy settings. (boyd, 2010: 50–51)

The affordance of visibility (boyd, 2014) does not account for the uses of Facebook in Mardin. Here, social media users can use Facebook to increase their visibility, or can take advantage of the online private spaces to communicate with illegitimate friends and sweethearts out of the gaze of neighbors and relatives. Private chats, fake and anonymous profiles, and closed groups constitute the ordinary way of using social media in Mardin. The statement “Public by default and private through effort” (boyd, 2014: 12) does not describe the logic of social media in my field-site. People are private by default, and go public through big effort. In Mardin, social media users change privacy settings and use Facebook mostly as a private or semi-private space. Being visible and public is often more laborious than being private because it implies a continuous monitoring of the public Facebook page and a carefully crafted performance of self. The affordance of persistence, defined as the durability of online expression and content (boyd, 2014), is only partial in Mardin because social media users frequently open and close accounts. When young people change partner, quarrel with a friend, or want to conceal something, they often close their old account and open a new one. Also, social media users rarely share their friends’ photos and images; consequently, when an account is closed down, images and posts are lost. Content posted on the Facebook wall is temporary and ephemeral. Searchability does not account for those many Facebook profiles with fake names and anonymous profile pictures.

Therefore, the affordances of visibility, persistence, and searchability are specific to a given social and cultural context. Social media technologies are not neutral and do contribute to shaping social interactions and communications, but users actively appropriate and adapt digital technologies to better reflect their own goals and lives. Consequently, affordances take shape only through specific material, and social and cultural circumstances (Davis and Chouinard, 2017). However, in social media studies, the agency of users has often been overlooked (Chambers, 2017). In my ethnographic data, there is evidence that people use the platform in creative and active ways that both designers and social media scholars have not envisaged. This ethnographic evidence shows that it is time to rethink the relationship between user agency and the structuring force of social media platforms, and to propose a more nuanced notion of affordance.

Social media studies have been largely dominated by platform-centric approaches that emphasize the role of architecture in shaping uses and content. For example, Van Dijck and Poell (2013) focused on the logic of social media, defined as “the strategies, mechanisms, and economies underpinning these platforms’ dynamics” (p. 3). They emphasized how properties of social media platforms shape contents and usage. Facebook’s business model promotes openness and the gathering of user’s data and the disclosure of friendships and connections (Van Dijck and Poell, 2013). Other scholars (see, for example, Bucher, 2012) highlighted the role played by software and algorithms in shaping friendships and sociability. The overall emphasis given to features and properties of the platform has developed together with a specific usage of the concept of affordance, which emphasizes the power of social media architecture, and overlooks the relational component and user agency. Davis and Chouinard (2017) argue that definitional confusion and failure to account for diverse subjects and circumstances are two of the main shortcomings in the uses of affordances. They argue that the concept of affordance has not adequately captured the relational component. “Affordances are largely defined ecologically, but in practice, [are] reduced to a ‘homogenous block’” (Scarantino, 2003: 961 in Davis and Chouinard, 2017).

The notion of affordance was initially formulated by Gibson (1966) to name the action possibilities that are available in different environments. The concept was then appropriated by Norman (1999) who brought the concept into design studies. Norman (1999) emphasized the role played by the perception of the users and distinguished between real and perceived affordances. More recently, the concept has been embraced by social scientists and media and communication scholars to describe the relationship between the properties of technologies and the structure of social relations, and to point out the technological qualities that are subsumed by users’ practices (Hutchby, 2001, 2003; McVeigh-Schultz and Baym, 2015; Medianou and Miller, 2013; Nagy and Neff, 2015). Scholars have conceptualized the concept of affordance to overcome deterministic accounts. Hutchby (2001) emphasized the relational component of affordances, which have been defined as the relationships between materiality of technology and practices of communication. Faraj and Azad (2012) advocated a focus on relationality between actions that occur among people and technologies. Madianou and Miller (2013) used the term affordance to address the contextualized user’s perceptions of the platforms. In their ethnographic study on new media and migration, they defined “new media as an environment of affordances” (Madianou and Miller, 2013: 169), and place emphasis on the social and emotional consequences of choosing between different media’s properties. The authors viewed affordances as communication opportunities defined by the relationship between them, and the relationship between these and the users. McVeigh-Schultz and Baym (2015) focused on the process of sense-making in shaping vernacular affordances. Nagy and Neff (2015) studied the imaginary dimension of the affordances of digital technologies. In all these studies, affordances are understood as a part of language, the imaginary, and practices of social media users. Here, the concept of affordance is conceived as relational, and is used to develop a middle ground between technology determinism and social constructivism, and overcome the duality of objects and subjects.

However, in social media studies, affordance has often been used to refer to properties and features of a technology that are separated from the social context and the work of human users. Most of the research on social media has been carried out in Anglo-American and European contexts and make generalizations that overlook the role of social and cultural contexts (e.g. Aakhus, 2007; boyd, 2008, 2010, 2014; Papacharissi and Yuan, 2011; Postigo, 2014; Trepte, 2015; Vitak, 2012, 2015; Vitak and Kim, 2014; Wellman et al., 2003). The affordances of persistence, replicability, scalability, and searchability described by boyd (2010) overlook the relational component and do not clearly state that they are the results of the encounters between the platform and specific users in a given socio-cultural context. Similarly, Papacharissi and Yuan (2011) introduced the affordance of shareability, defined as a property of the architecture that invites people to share. Vitak and Kim (2014) argued that social media—called social networking sites (SNSs)—are distinguished from other media by their unique affordances, including the “visibility and persistence of content, the association of connections, and the editability of content” (p. 472). In this statement, affordances are defined as stable features, and the role played by socially and culturally situated human actors is not taken into account. Similarly, Trepte (2015) understands affordances as the inherent features of social media and not as a relational process. Postigo (2014) uses the concept of affordance to name the technological features that are designed into YouTube to serve business interests, and which “create a set of probable uses/meanings for YouTube, most of which are undertaken as social practice” (p. 333). These accounts do not take into consideration that it is the users’ action that makes these features possible; as a result, they fail to account for the diverse and multiple object–subject relations that can take shape in different social contexts. These accounts describe social media as platforms that embody stable architectures, and tend to simply identify or list affordances, operation that provides limited theoretical value (Evans et al., 2016).

The term affordance should inscribe the possibilities not actualized by media users (Have and Pedersen, 2015), but it fails to do so because the properties that shape, constrain, or generate practices cannot be known outside their actual situated uses. I can describe only actualized and enacted affordances. Affordances are never specific to a platform only, but are always specific to the relation between the platform and the situated users. In his research on e-learning technologies, Oliver (2005) argued that measuring affordances “becomes speculative rather than analytic” (p. 401), and affordances are “redundant as analytic objects” (p. 406). There is a methodological issue at stake: we cannot acknowledge and describe unanticipated uses. But there is a theoretical shortcoming, too: the ethnographic data presented in this article highlight the impossibility of acknowledging all the potentialities of a social media platform. The ethnographic evidence urges a reconsideration of the concept of affordance.

Drawing on Orlikowski’s ³ (2000) concept of “technologies-in-practice,” I propose the concept of affordances-in-practice, the enactment of platform properties by specific users within social and cultural contexts. Orlikowski draws on Giddens’ (1979, 1984) notion of structure to focus attention on the ongoing enactments of a technology, and the work of humans who through their social practices produce and constitute technology’s structures (Orlikowski, 2000: 407). Technologies-in-practice are enacted structures. They are “the sets of rules and resources that are (re)constituted in people’s recurrent engagement with the technologies at hand” (Orlikowski, 2000: 407). Likewise, social media affordances are the results of the repeated interactions between humans and platforms. They are the results of social and material enablement. As such, they are not fixed and stable properties but are implicated in different ongoing processes of constitution, which may radically vary across social and cultural contexts. The concept of affordances-in-practice avoids theoretical reductionism by stressing the fact that affordances take shape only through specific material and social practices. Concentrating on affordances as culturally and socially bound, I counter the idea of social media logic (Van Dijck and Poell 2013) and view context collapse as the result of situated practices within Anglo-American contexts.

In the last two decades, media scholars have increasingly focused on media practices (e.g. Barassi, 2015; Bräuchler and Postill, 2010; Couldry, 2004; Moores, 2000). Practice has become an emerging theme in media studies, to escape the constraints laid by an emphasis on text (Couldry, 2004). Theories of domestication in relation to new media and technologies (among others, see Hirsch, 1992; Silverstone, 1994) have also formulated theoretical frameworks that emphasize the role of practice and processes of technological appropriation. There are many ethnographic studies on everyday practices of Internet usage and new media. ⁴ By contrast, studies on social media are strongly anchored to concepts of social media logic, architecture, and algorithm. The field is dominated by research that builds on software-centered approaches, which emphasize the role played by social media architecture in shaping and constraining the practices of its users. This trend goes along with a misappropriation of the concept of affordances and a lack of studies on social media and everyday life in non-Western countries.

Little ethnographic research on social media practices escapes the constraints posed by the architecture and social media logic paradigms. Among others, such is the comparative project WhyWePost ⁵ (Costa, 2016; Haynes, 2016; McDonald, 2016; Miller, 2016; Miller et al., 2016; Nicolescu 2016; Wang, 2016). In the manifesto of the journal Social Media + Society, Mary Gray (2015) calls for a curatorial theory of social media. She argues, “Social media scholarship could challenge the assumption that there is something settled or predetermined—predictable—about the meaning or potential of communication and information outside of their practiced, situated deployments” (p. 2). Yet, studies on social media are far from challenging the assumptions that platforms exist outside of their practices of usage.

Conclusion

Anthropological research on media (Bird, 2003; Ginsburg et al., 2002) has contributed to shifting the paradigm of media research. In the early 2000s, a new research approach that sees media as practice moved beyond the ongoing debates on text and production economy. Also, Nick Couldry (2004) proposed a different paradigm that “treats media as the open set of practices relating to, or oriented around, media” (p. 117). He aimed at overcoming the constraints posed by the main approaches that have influenced media research, such as media texts and production structures and paradigms that have dominated the field of media studies, but did not address the variety of ways in which media “are embedded in the interlocking fabric of social and cultural life” (Couldry, 2004: 129). Similarly, anthropological research on social media (Costa, 2016; Miller et al., 2016) has emphasized user agency, challenging the idea that social media have a stable architecture and logic that can be studied out of situated practices of usage.

The ethnographic data and examples presented above have demonstrated that context collapse is the result of situated practices of social media usage within Western Anglophone contexts, and not the effect of social media logic and stable properties of the platform. I have argued that these findings urge a refinement of the concept of affordances, which in the literature on social media has been used in incoherent and contradictory ways. Although this concept has been formulated to “focus on questions of the use-in-situated-interaction of technological devices” (Hutchby, 2003: 582) and to recognize the constraining and enabling properties of technologies and artifacts, its current uses in social media scholarship often refer to stable and inherent properties of social media platforms (e.g. Aakhus, 2007; boyd, 2008, 2010, 2014; Papacharissi and Yuan, 2011; Postigo, 2014; Trepte, 2015; Vitak, 2012, 2015; Vitak and Kim, 2014; Wellman et al., 2003). Persistence, searchability, scalability, and replicability (boyd, 2010) are not intrinsic and stable qualities that “emerge out of the properties of bits” (boyd, 2010: 7), but rather the accomplishments of technical and material potentialities within specific situated contexts. Yet, they have not been recognized as such. In order to escape this shortcoming, I propose the term affordances-in-practice, which can account for the multiple and varied realizations of the social technical potentialities of social media in different places and social groups around the world. Affordances-in-practice stress the idea that affordances are not intrinsic properties that can be defined outside their situated context of usage, but ongoing enactments by specific users that may vary across space and time. Such refinement should encourage academic discourses on social media to take into account the diversity and multiplicity of mediated practices around the world, and to contribute to de-Westernizing social media studies.