Studying Instagram Beyond Selfies

Introduction

According to WordStream (2019), of approximately 40 billion photos posted on Instagram to date, only 282 million are selfies—just 0.7%. Thus, for all its zeitgeisty appeal, the selfie is in fact a niche phenomenon in the larger context of Instagram genres. The attention given to selfies is more than understandable, given their importance to brands and influencers as promotional tools (Marwick, 2015). However, such users represent just a small portion of the whole Instagram population. Working through the literature reveals a proliferation of studies on selfies, with three special issues published dedicated to them (Kedzior et al., 2016; Senft & Baym, 2015; Warfield et al., 2016), yet none on Instagram per se. Noting this gap and following similar calls to study Instagram with ad hoc approaches, concepts and methodologies that would be platform-specific (Highfield & Leaver, 2015, 2016; Laestadius, 2017), we identified an opportunity to engage with scholars from all over the world with the challenge of proposing new theories and approaches capable of unlocking the full socio-anthropological potential of Instagram.

This special issue is named after the 2018 Instagram Conference (hosted by Middlesex University London on June 1, 2018) that we organized in our effort to broaden the horizons of research on the platform. The event was sub-titled Studying Instagram Beyond Selfies. Let us be clear, we don’t have anything against selfies, people who take selfies, or scholars who study them. It is not our intention to downplay the selfie phenomenon, nor to dismiss selfies as frivolous acts undeserving of scholarly attention. We are in fact very mindful of their broad sociological import—not least their subversive function and potential—and we remain indebted to those scholars who have brought this phenomenon to wide public attention (Abidin, 2016; Senft & Baym, 2015). Of course, the choice of subtitle was (and is) deliberately provocative.

Theoretical and Methodological Challenges of Studying Instagram

With the impressive amount of 1 billion users worldwide—more than Twitter, Snapchat, and Pinterest combined (We Are Social, 2019)—Instagram has become one of the most important social networking sites globally and in the process has transformed the role of photographs and photography in visual culture. Designed to exploit the affordances of mobile media (Carah & Shaul, 2015) and the immediate and intuitive logic of visual communication, Instagram is notably popular among young people (18–29 years old) (Word Stream, 2019). Instagram is not only huge, it is also dynamic, since, as with other prominent social media platforms, its architecture is always evolving to adapt to the latest market and cultural trends. Launched in 2010 by Kevin Systrom and Mike Krieger as a “free iPhone application designed for sharing pictures with friends” (Laestadius, 2017, p. 573), Instagram has steadily iterated many new features over the years, such as short videos (2013), stories (2016), shoppable posts (2017), and IGTV (2018) (Wikipedia, 2019). Recently, Instagram also changed its algorithm (Instagram, 2016) so that user feeds no longer display posts chronologically but instead according to specific signals—designed to deliver a more “relevant” personalized experience which, of course, includes targeted advertising. Such shifts have radically changed users’ content strategies and, ultimately, how they relate to the platform and each other (O’Meara, 2019).

The phenomenal success of Instagram has not gone unnoticed by brands and influencers who have concomitantly increased their investments and activities on the platform. According to Forbes (2016) the current financial value of Instagram is somewhere between US$25 billion and US$50 billion. Beyond its popularity among users and commercial value to brands, Instagram is an environment but also a device that offers rich possibilities for conducting social research. Most notably, it opens up a number of new pathways for exploring sociocultural processes related to the more mundane and overlooked aspects of “ordinary” lives (Williams, 1989). Instagram enables researchers not only to analyze almost unimaginably vast swathes of visual content in high definition, but also to see through the eyes of social actors, to glimpse into their everyday micro-rituals and private moments. The platform provides privileged access to both back and front stages—the essential material of sociology and anthropology—which were either unavailable or prohibitively expensive to explore at scale previously (Latour et al., 2012).

As the articles collected in this special issue testify, Instagram has an enormous impact on people’s everyday lives on many levels and so clearly it deserves rigorous academic attention. Given the scale and complexity of these impacts, studying Instagram poses different kinds of challenges and many gaps are still to be filled. Let’s consider a few of these.

In contrast to other notable photo-sharing websites (like Flickr), Instagram is explicitly designed to encourage ordinary users to share their amateur pictures with friends. Moreover, as it is principally a mobile application first and foremost (e.g., it is not possible to upload files through its website interface), it invites users to capture mundane moments in everyday life and share the more intimate, minute, and granular details of lives lived “in the moment.” In this way Instagram has developed a vernacular visual culture (Burgess, 2006), characterized by a specific language; that is, a “system of conventions and techniques that define[s] the subjects, narratives, editing, compositions, lighting, sequencing, and other image characteristics” (Manovich, 2017, p. 18). An example of this can be seen on “@insta_repeat,” an Instagram channel that collects repertoires of such (quite repetitive) Instagram visual languages. The idea that Instagram entails a specific visual culture is well captured by the concept of platform vernacular (Gibbs et al., 2015). A platform vernacular “refers to the different narrative patterns that shape content and the flow of information” (Niederer & Colombo, 2019, p. 55) across a given platform, which are not solely driven by the creativity of individual users but rather shaped by the “specificities of the platform, its material architecture, and the collective cultural practices that operate on and through it” (Gibbs et al., 2015, pp. 257–258). As is well documented by studies of Twitter (Caliandro, 2018), it is not sufficiently accurate to speak of platform vernaculars in general. In fact, it is empirically impossible to identify a single language for a single platform—(as the phenomenon of selfie seems to suggest). Conversely, different languages emerge according to the different social formations (crowd, communities, or publics) inhabiting the platform and different topics around which they converge (Bruns et al., 2016; Marres, 2015). Unfortunately, and despite how promising the concept of platform vernacular might be for sociological enquiry, very little research on specific Instagram sub-vernaculars has been carried out so far. We therefore suggest that a greater and more systematic attention to Instagram’s vernaculars could open new avenues of research, both within and across disparate fields such as politics, health, consumption, aging, art, and so on. Moreover, this project will necessarily require the elaboration of new theories and methodological approaches that will in turn reveal and open up new domains of study.

Alongside the theoretical question, there are methodological and technical challenges to take into consideration too. The first one concerns liberation from what might be termed “Twitter-dependency.” Instagram is composed of a set of devices, like hashtags, mentions, likes, captions, geotags, which are very similar to those found on Twitter, but have been in fact been designed for different functionality and are employed by users with different purposes in mind. Let’s consider the hashtags: on Twitter they serve mainly to aggregate conversations; on Instagram they specify the content of pictures as well as connect to ad hoc communities (Caliandro & Gandini, 2017). Thus, as Highfield and Leaver (2015) stress, when exploring Instagram we must restrain ourselves from using Twitter as a methodological blueprint. There is, therefore, a pressing need for Instagram-native research strategies which exploit the specific methodological potentials of Instagram’s hashtags, mentions, likes, captions, and geotags to enable in-depth investigations.

An equally important methodological challenge concerns the most crucial pieces of content on Instagram: images. A lot has still to been done in terms of deploying and developing techniques to collect, organize, analyze, and ultimately make sense of the enormous collection of images populating Instagram—from both qualitative and quantitative perspectives. Pioneering examples here include computational approaches that, for example, use automated image recognition to explore huge quantities of Instagram images and uncover macro-visual patterns of colors, photo filters used, or selfie styles (Manovich, 2017). On the qualitative side, the networked content analysis developed by digital methods scholars (Niederer, 2016; Niederer & Colombo, 2019) has given fresh and useful inputs for the interpretation of Instagram content, exhorting researchers not merely to consider an image as a standalone piece of content but rather as a part of a network of metadata that contribute to defining its meaning and social uses. These studies have paved the way, yet both computational approaches and networked content analysis are still underemployed in social research on Instagram. Indeed, we believe that the combination of computational approaches and networked content analysis is desirable for the future, especially for supporting scholars in exploring meaningful and emerging social phenomena on Instagram.

Finally, a major issue concerning the very possibility of conducting social research on Instagram arises from the fact that in 2016 the platform heavily curtailed access to its feed API for non-commercial partners, rendering it particularly difficult for many academics to access for research purposes (Puschmann, 2019). This situation stems from the Cambridge Analytica scandal and the consequent social media APIs curtailments undertaken by Facebook and Twitter, what Axel Bruns (2019) in a recent critical article termed the “APIcalypse.” Regardless of its causes, this situation poses serious limitations to the possibility of developing sustainable and effective strategies for collecting and analyzing Instagram data. And yet this challenge also presents an invaluable opportunity to researchers. Just as Instagram continues to evolve, we hope that the innovative approaches offered in this special issue not only provoke discussion but also stimulate others to continue experimenting with new methodological approaches for exploring the platform and its role in our lives.

With this special issue we don’t claim to resolve all the issues noted above. More modestly, our goal is to kick-start a fruitful conversation among social media scholars interested in furthering Instagram studies. In this, we aim at contributing to a dedicated stream of Instagram-centered research by building on the work of those innovative scholars who have come before us (see Gibbs et al., 2015; Leaver et al., 2019; Manovich, 2017; Niederer & Colombo, 2019). This would not be possible, of course, without the invaluable contributions of the participant in this special issue.

In their article on “‘Visual Affluence’ in Social Photography: Applicability of Image Segmentation as a Visually-Oriented Approach to Study Instagram Hashtags,” Chamil Rathnayake and Irida Ntalla elaborate a methodological approach that addresses an aspect of social photography that has hitherto gone under-researched, namely, the relative visual “affluence” of images that are identified by a given hashtag. While close readings of posts of the kind offered by Bozzi in this volume can provide invaluable ethnographic insight, such data can be limited in its generalizability. Indeed, the subjective meaning-making among individual users and broader communities is intrinsic to what Rathnayake and Irida identify as an “internal incongruence” that is typical of content within hashtags, an argument they illustrate by analyzing content within hashtags for #Trump and #Graffiti. Relatedly, where quantification can be used to measure engagement via a platform’s native “vanity metrics” (Rogers, 2018), such approaches only start to scratch the surface of the data contained in posts and networks. The limiting factor in all of these examples is the central concern within social science research for representation and meaning-making. While we have witnessed a visual turn in social media research in recent years, this has tended to prioritize the social over the content and affordances of the visual as such.

Moving beyond the strictly “social” dimensions of social photography—that is, the local significance of posts that are meaningful for communities of users connected via hashtags—Rathnayake and Ntalla propose that the affluence of visual features in images become the focal object of study, and that automated image segmentation techniques enable their meaning-independent analysis. This quantitative approach references cognate fields of vision research—into visual clutter, complexity, and entropy—but differs from them insofar as the segmentation technique identifies the color regions in a variety of objects and surfaces in images in a manner that captures their “visual richness.”

Applying this approach to a sample of five Instagram hashtags, #food, #nature, #graffiti, #minimalism, and #instagood, Rathnayake and Ntalla show how hashtags can contain differing levels of visual affluence. This not only challenges much of the received thinking about hashtags as such, it also opens new avenues of enquiry into how different levels of visual affluence might appeal to different personalities and how this might in turn impact on user and consumer attitudes and responses.

Just as Rathnayake and Ntalla challenge assumptions around what might be considered meaningful hashtag content, in “Digital Methods for Hashtag Engagement Research,” Janna Joceli Omena, Elaine Teixeira Rabello and André Goes Mintz similarly call into question received ideas around the definition and scope of hashtag “engagement”—the metrics-based analysis of which provides what is currently the dominant approach to understanding user and consumer attitudes and responses on the platform. Their contention is to outline a methodology that moves hashtag engagement analysis beyond a simplistic approach that frames engagement simply in terms of vanity metrics, that is, “the ‘most engaged’ list.” To address the complexity of hashtag engagement involves critically accounting for the relationship between hashtags and what the authors term their “forms of grammatization.”

Grammatization refers to the process whereby the specific affordances of socio-technical platforms—in this case, the natively digital object that is the hashtag—facilitate user activity in ways that enable, structure, and ultimately formalize actions. In the case of the hashtag, it is a mechanism whereby user actions—engagement—can be traced and rendered calculable. However, hashtags acquire meaning and function differently according to how they are used by social actors in a context where the architecture of the platform continues to evolve. As we shall see in some of the other contributions to this special issue, the socio-technical element of grammatization is therefore intrinsic to the commercial and political—as well as social and aesthetic—dimensions of the platform. Just as the standardization and “metrification” of engagement provides the means for certain types of users to operationalize these grammars in the service of whatever their communication goal might be—setting a political agenda, reaching a very particular demographic, and so on—the complexity of the process also allows for heterogeneity and creativity, facilitating the emergence of new kinds of aesthetic and political imaginaries.

In their analysis of hashtag grammatization, the authors posit a three-layer (3L) approach to investigate the ways in which hashtags are used in relation both to their content and range of implicated actors. Layer 1 of the analysis involves distinguishing between “high visibility” and “ordinary” users. Layer 2 focuses on hashtagging activity as such, and Layer 3 factors in the visual and textual content. As illustrated in the case study of political protests in Brazil in March 2016, the key to this approach is that it distinguishes between dominant and ordinary groups of users by drawing a contrast between “high-visibility” and “ordinary” grammars of engagement. Applying the 3L approach to hashtags associated first with the protest to support the parliamentary process to remove president Dilma Rousseff from office, and then from the subsequent protest against the same, most notably the authors identify different “cultures of appropriation” between high-visibility and ordinary actors. In the course of just a few weeks certain patterns and tropes emerged on both sides of the political debate, but also very particular shifts in meaning occurred within hashtags within these distinctive groups.

Such forms of platform-specific cultural production do not of course occur in vacuums. Terri L. Towner and Caroline Lego Muñoz’s article “Instagramming Issues: Agenda Setting during the 2016 Presidential Campaign” reminds us that the aesthetic and political imaginaries afforded by the forms of engagement discussed by Omena, Rabello, and Mintz are framed by external events and the broader media ecosystem of which Instagram is an increasingly important component. Their concern is to investigate precisely how important it has become by deploying a comparative content analysis of campaign coverage on the platform and in newspapers to determine the role and scope of the platform in “agenda setting” in the 2016 U.S. presidential election. Introduced as a contribution to agenda-setting theory, the article is concerned with the role played by citizens—ordinary users—in comparison with the journalists and associated intermediaries who have typically dominated this area in the mainstream media.

Instagram is becoming increasingly important to journalists, both as a space to disseminate their work and as a research tool for identifying issues and monitoring conversations. However, despite the growing use of the platform among the public, and of the much more direct and transparent relationship between journalists and the public that it affords, evidence of the platform’s influence in setting the mainstream media’s news agenda is mixed. Through their comparative content analysis, Towner and Muñoz identify several interesting and significant trends. Primary among these is that for some of the particular demographics sampled during the election campaign—young adults and minorities—Instagram appears to be replacing mainstream media as a space for publicizing and discussing the most relevant issues. Foremost among these are the eponymous hashtag movements #blacklivesmatters and #metoo. The analysis reveals a very weak relationship between the issue agenda of newspapers and that of posts on Instagram, yet there is some evidence of the opposite—that Instagram posts do have a degree of influence on newspaper agendas. But despite there being some interrelationship, tellingly, there is a notable time lag between the posts and newspaper coverage of the issues, suggesting that while its use among journalists is increasing, Instagram does not yet occupy the same space as either Twitter or Facebook in contemporary journalism. However, in acknowledging the pace of change in the digital news environment, the authors note that a growth in intermedia influence is very likely, as well as the necessity for studies that focus on the increasing importance of the visual in news consumption.

While both quantitative and qualitative methods provide researchers with the means of exploring the socio-technical affordances and effects of Instagram in ways that would be unimaginable even 20 years ago, Elisa Serafinelli’s article, “Networked Remembrance in the Time of insta-memories,” illustrates how, in some senses, the platform merely extends and amplifies a long-established use of photographs: what she terms, following Susan Sontag (1979), their remembrative function, which is to say their capacity to facilitate memory.

Serafinelli points out that since its invention photography has been used as a tool for remembering by offering the means both to capture a moment in time and then later to retrieve, review, and share that moment. In such ways, photography can be said to keep memories alive. By acting as a “digital memory box,” Instagram extends this function into a context where the practice and role of photography in our lives has changed dramatically. In an era of social photography—where the taking and sharing of photographic images via connected mobile devices is instantaneous and ubiquitous—the remembrative work of photographs has become interwoven with myriad social and communicative functions. The memories captured in photography have become distributed, networked, and connective. Most strikingly in this context the remembrative function is collectivized, in part due to the public nature of the platform, but intriguingly this is not necessarily oriented to publicity as such.

Using a qualitative approach informed by netnography, online participant observation, and visual content analysis, Serafinelli discusses examples that illustrate the diversity of ways in which Instagram is used as a “memory tool.” Alongside this diversity, however, there are distinctive trends, most notably—and somewhat contradicting arguments concerning the instaneity of photosharing—the considered selection of images to be shared. Among the participants sampled in the study the trend is to select the “best possible” photographs, but conversely, there is a notable reluctance to post intimate (especially family) images to the platform. Instagram, therefore, has become a central agent in the transformation of photography and its remembrative role in our lives. As a memory archive, however, it has its limitations, as users seek to protect their most intimate memories from the consequences of what Marwick and boyd (2011) have termed “context collapse.”

Just as digital networking technologies have given rise to new and emergent forms of communication and sociality, many of which are discussed by contributors to this special issue, they have also liberated certain kinds of workers from traditional work spaces. According to Bozzi, in his “Cultural Critique of the Travelling Entrepreneur on Instagram,” this particular platform plays a vital role in such developments as it serves as the primary social imaginary for such “digital nomads.”

While the figure of the digital nomad has been with us since the emergence of the World Wide Web in the 1990s, the increasing dominance of neoliberalism coupled with the ubiquitous connectivity of mobile devices has seen the entrepreneurial promise of remote work become increasingly mainstream, what Bozzi terms a “quasi-utopian horizon.” Discussing the imaginary appeal of the visual tropes associated with the hashtags #digitalnomad, #solotraveller, #remotework, and #4hourworkweek alongside the geotagging behaviors of their users, Bozzi identifies a depoliticized aesthetic that stands in stark contrast to the very real effects of the globalized gentrification of which the digital nomad is symptomatic.

Yet, despite the hollowness of this figure (permanently on the lookout for “authenticity and wi-fi”; the image of a “laptop next to a cappuccino on a wooden desk” their signature post), Bozzi identifies in their use of Instagram possibility as well as contradiction. Most notably, the ways in which work is aestheticized by digital nomads through their creative use of the platform’s techno-poetic affordances signals the potential for alternative cultural narratives and new political categories. In this reading, the “travelling entrepreneur” is construed not so much as an avatar of neoliberalism, but rather of post-work. This is a scenario where the redistributive potential of digital work might be alloyed with the communitarian potential of social networks—a quasi-utopian horizon that is no less speculative, and indeed ideological, than the present (neoliberal) digital nomad social imaginary as it is outlined in the article.

However, in line with other contributors to this volume, this speculative conclusion is arrived at via a methodological approach that seeks to decenter currently dominant paradigms—in this case those that take the self and the network as primary objects of study. In a related vein to Omena, Rabello, and Mintz, it does so by drawing attention to the heterogeneous ways in which users deploy and engage with hashtags and geotags associated with digital nomadism. In this regard, what Bozzi says of the digital nomad might be more widely generalizable, whereby the practice of tagging informs a collective cultural production wherein the “impersonal quality and the visual character of this gesture go hand in hand . . . so the user is at once pulling themselves towards that imagery and drawing from it to assemble part of their own social media self.”

The extent to which self-representations on Instagram as such can be deemed political is the subject of Caldeira, De Ridder, and Van Bauwel’s article “Between the Mundane and the Political: Women’s Self-Representation on Instagram.” Focusing their research firmly on “ordinary” female Instagram users, the authors offer a detailed textual analysis of a very broad cross section of such users in order to gain a sense of the enormous range and variety of visual tropes through which women perform identity through the platform. While less than 2% of the posts in their sample could be identified as overtly political, the authors trace myriad examples in which users articulate tangential political issues and themes through what they describe as the characteristically mundane self-representations that have become intrinsic to what is understood as “photographable” on Instagram. Taking this concept from Bourdieu’s (1965) early work on the subject, the authors note how practices of self-representation on Instagram (in particular selfies) are imbued with social, aesthetic, and ethical values and as such define what is worthy of being photographed in a given context. Moving beyond a simplistic notion of the selfie as being merely a narcissistic status-seeking performance, such practices have the effect of expanding the “realm of public visibility, granting both aesthetic and political value to moments and people otherwise symbolically annihilated.”

It is in this expanded realm of public visibility that the authors identify, for example, a diversity of femininities rather than a singular or homogeneous gender identify that might have been mediated via broadcast media in the past. Such expressions evidence feminist political potential, yet many examples are also, the authors acknowledge, symptomatic of postfeminist sensibilities where the political horizon is limited to a concern with individual choice and empowerment, rather than a broader collectivist concern with systemic inequality.

In “Lifestyle Enclaves in the Instagram City,” John D. Boy and Justus Uitermark present a more empirically focused study of the role played by Instagram in the creation of subcultures and social divisions in the city. Situating their study within the sociological tradition of Gemeinschaft and Gesellschaft, the authors draw an analogy between the role of the city and by the social media platform in determining the underlying character of social relations. On one hand this is observable in the amplification of those forces and forms of social fragmentation familiar to urban sociologists through the selective presentation of city spaces by Instagram users. Much like the role played by “digital nomads” in Bozzi’s discussion of gentrification, spaces of (or associated with) aspirational consumption are typically more “instagrammable” than quotidian urban spaces, leading to the construction of the “lifestyle enclaves” referred to in the article title.

Without discounting the social consequences of this particular manifestation of fragmentation, Boy and Uitermark argue that there is an equally significant—and in some ways no less problematic—countervailing process of integration at play. Here, integration describes the actions and outcomes of users connecting around shared interests but also surveilling each other when and by doing so. While such actions and activities facilitate community—which it goes without saying is especially important for minorities who might otherwise face exclusion—an unintended consequence is to suppress difference and instead foster conformity. The authors draw on studies that show how the decline of certain urban subcultural milieus is concomitant with the growth of social media, before offering an empirical analysis derived from a city-wide dataset using multiple methods that show the ways in which Amsterdam is indeed “segregated” on Instagram—but also the ways in which communities constellate and frequently intersect in “zones” at what the authors term the “the online–offline interface, i.e., symbolic and material domains that serve as stages for the enactment of identity and the performance of status.”

One of the biggest challenges facing social media researchers working on Instagram is in developing methods that can adequately address the scale and scope of the available date without losing the capacity for analyzing the more granular aspects of participatory cultures, especially as these evolve over longer durations. In her discussion of “Self-Tracking Health Over Time: From the Use of Instagram to Perform Optimal Health to the Protective Shield of the Digital Detox,” Rachael Kent demonstrates how this challenge can be met via an innovative ethnographic approach that triangulates a textual and thematic analysis of online content with reflexive diaries and semi-structured interviews. While the results of her study do not have the same scope for generalizability as some of the more data-focused studies in this collection, they offer an invaluable insight into the “online-offline interface” of a particular community of users and in so doing broaden the horizon of digital research methods more generally.

Kent’s article investigates the ways in which Instagram users perform “optimal” health identities through their representation of health-tracking and associated self-monitoring practices. By analyzing these representations but also contextualizing them with the diaries and interviews, Kent shows how surveillance and feedback from the community over time influences individual health management and decisions in their offline everyday lives. At the outset of the study the behavior of participants all broadly accord with critical accounts of the “quantified self,” whereby self-optimizing technologies produce ideal neoliberal citizens. Over time, however, a number of participants demonstrated a growing awareness of and resistance to such subjectification—specifically, to the negative impact on mental health it gives rise to—leading in some cases to users “detoxing” and even quitting the platform entirely. While the “friendly” surveillance afforded by Instagram is the reason it proves so popular with users concerned with health-optimization, the “integration” that Boy and Uitermark speak of in another context manifests here as pressure to both conform and perform in ways that can become unsustainable and counter-productive.