Abstract
Situating (My) “Below the Radar” Studies
I am grateful for the invitation to keynote at the inaugural AoIR Flashpoint Symposium in Urbino, in part for the privilege to share my research with scholars across the world, and in part because the eloquently crafted mission of the event led me to seriously reflect upon my body of work thus far. You see, I chiefly study internet celebrities and Influencer cultures, and most people would understandably wonder what these “extremely online” and highly visible actors have to do with being “below the radar.” A popular misconception of Influencers’ use of social media is that they always aim for maximum publicity and attention. Yet, between algorithmic and analogue labor, Influencers balance hyper-visibility and under-visibility to avoid over-exposure in some areas and redirect audience interest to others, and this was a recurring theme that I had newly discovered across my published research in the last 8 years.
My earliest work on the very beginnings of the Influencer industry in Southeast Asia focused on the 2005–2011 cultures of blogshop owners and models chiefly in Singapore, although fieldwork eventually extended to include Indonesian, Malaysian, and Thai sites. These were teenagers and young women who sold used and newly imported clothing online through blog platforms, and who painstakingly cultivated “persona intimacies” (Abidin & Thompson, 2012, p. 472) with viewers-cum-prospective-customers. They did so by engaging in a series of “commercial,” “interactive,” “reciprocal,” and “disclosive” intimacies (Abidin, 2015a) to cultivate an “emotional attachment between blogshop consumers . . . and the models” that led to feelings of community, loyalty, and investment to a blogshop and the popular girls online who represented it (Abidin & Thompson, 2012, p. 472). But much of this “relational labor” (Baym, 2015) where performers invest in strategies to seed feelings of connection with their audience was motivated by an underlying commercial intention—to push for more blogshop sales (Abidin & Thompson, 2012, p. 472). The otherwise clinical and commercial transactions of selling used and new clothes online through blog platforms was softened and masked through strategies like “girl talk” (Currie, 1999; Waff, 1994 in the work of Abidin & Thompson, 2012). Such a deflection is one example of sliding something “below the radar.” Before we get to the social science behind such phenomena in the Influencer industry and other internet cultures, please oblige my anthropological reflexivity in a short personal story.
My anthropological interest in such agile plays in visibility was a condition of my fieldwork: The blogshop industry and subsequently the blogging/Influencer industry of the mid-2000s to early 2010s in Singapore was dominated by young Chinese women (see Abidin, 2019a), and I was a Malay–Chinese person who identified more with my Malay roots but (fortunately or unfortunately) passed as Chinese in these circles—for reasons unknown to me and many other mixed race children, I was naturally tan skinned as a child but became gradually fairer in my early 20s when this research begun; for this, I would like to acknowledge Biology for ushering me into this mini-existential crisis of feeling racially/culturally illegitimate. Many of my fieldwork experiences of “feeling unseen” or “being seen but wrongly” eventually inspired an altogether different stand of research into inter-Asian mixed race young people, focused on ambivalent phenotypic signifiers (Abidin, 2014b) and strategies for passing as “one or the other” in various communities (Abidin, 2017b).
As I continued to study the everyday acrobatics of playing with the radar, I learnt early on to leverage on my intersectional biography and demography to appeal to my informants and gain deeper entry into the field (Abidin, 2020c)—while my Influencer informants often attributed my awkward Mandarin accent to the fact that I had immigrated to Australia, in reality my mixed race parentage and upbringing was designed such that English was the “leveller” in the household, and I only ever used Mandarin sparingly during Chinese lessons. But letting these mundane minutiae pass made way for my informants to see me as an “exotic inbetweener” of sorts (Abidin, 2020c, p. 62), which facilitated my learning of “their world.”
While I am tempted to romanticize the (not too distant) past and claim that I had undergone extremely rigorous, scientific, and systematic training to identify such “below the radar” practices, the fact of the matter is that these sensibilities and sensitivities are built into the flesh of minority persons and people who live on the margins—as Palestinian-American anthropologist Lila Abu-Lughod (1985) suggests, anthropologists who “not only can present themselves as different but can use the difference as a way of stimulating discussion” (p. 18) may find themselves embedding into different circles and gifted new revelations.
In the rest of this article, I would like to reflect on my experience of researching “below the radar” subjects in my (thus far short) journey as an early career researcher, and make the case for offering a companion framework to “networked publics” (boyd, 2010) (persistence, searchability, replicability, and scalability): A framework of “refracted publics” (transience, discoverability, decodability, and silosociality; see Table 1). The selection of studies presented draws from a long trajectory of my work on the evolution of Influencer cultures: Influencer cultures as a job description and culture of practice; Influencer cultures as a concept and role; and Influencer cultures as an amplification platform. Specifically, I present brief case studies from my various works on social media Influencers, politicians on social media, content creation Facebook pages, K-pop fandoms on social media, Instagram cultures, tumblr cultures, meme factories, and internet paralanguages. All of these subjects are streamlined under the umbrella of (what I have been loosely referring to, and am still developing as) “internet pop culture.” As it is ambitious to attempt an update of technology and social media scholar danah boyd’s (2010) foundational work on “networked publics,” please allow me to first present the framework on “refracted publics” before making the case for it as the article delves into a series of ethnographic case studies.
Conditions and dynamics between “Networked publics” and “Refracted publics.”
“Networked publics” and a companion “refracted publics”
Context
boyd’s (2010) work on networked publics arose from her studies on social network sites that were “popular online hangout spaces” among young people and adults (p. 39), which she defined as publics that are restructured by networked technologies. As such, they are simultaneously (1) the space constructed through networked technologies and (2) the imagined collective that emerges as a result of the intersection of people, technology, and practice . . . the ways in which technology structures them introduces distinct affordances that shape how people engage with these environments [. . . and] introduce new possibilities for interaction . . . Networked publics’ affordances do not dictate participants’ behavior, but they do configure the environment in a way that shapes participants’ engagement. In essence, the architecture of a particular environment matters and the architecture of networked publics is shaped by their affordances. (p. 39)
Networked publics was birthed out of disciplinary training in media and communications and initially focused on platforms, specifically their infrastructure and affordances. In companion, this article’s offer of “refracted publics” arises out of disciplinary training in anthropology and sociology and focuses on (young people’s internet) user cultures, specifically their agentic and circumventive adaptations of what platforms offer them.
Consider my choice of vocabulary (i.e., refracted) a labor of love—it was, after all, inspired by the hundreds of hours spent listening to my fishing enthusiast of a partner muse over strategic theories and folklore. Standing on shore, a fisherman’s view of the catch in the water differs from what is actually happening beneath. Even if armed with the best reels and rods and the knowledge of customizing them, the layer of water functions as a deflective lens or barrier that distorts or impedes a fisherman’s access to the catch. To get around this, a fisherman must acquaint themselves with physics and the nature of light to account for necessary adjustments, or better yet, switch from sight-fishing from shore to sonar-fishing from a boat to increase the success of yield. Refracted publics consider the conditions of spaces as they are manipulated by users to enhance, deflect, or deter detection. I define refracted publics as publics that are circumvented by users. As such, they are simultaneously (1) the space constructed out of the desire for refracted perceptions and (2) the collection of subversive or circumvention practices as a result of analogue and algorithmic manipulations of vision and access. Refracted publics allow users and their content to
boyd (2010) maintains that networked publics are “not just publics networked together, but they are publics that have been transformed by networked media, its properties, and its potential” (p. 42) during the 2000s. In a similar vein, refracted publics are not merely collections of any circumventive, subversive, or off-label practices, but instead are products of their time in a landscape of platform data leaks, political protests, fake news, and (most recently) COVID-19. Refracted publics that have been transformed by an internet culture of the 2010s comprising:
Where networked publics arise from features of social network sites including profiles, friends lists, public commenting tools, stream-based updates (boyd, 2010, p. 42), refracted publics arise from cultures of users who dog whistle (encoding a message to mean one thing to the general public with additional meaning for the target audience), adopt “parallel literacies” (encoding a message to mean vastly different things for different target audiences, Abidin, 2019a, p. 15), and employ social steganography (hiding messages in plain sight; boyd & Marwick, 2011).
Conditions
While the four conditions of networked publics are “persistence,” “searchability,” “replicability,” and “scalability” (boyd, 2010, p. 45), the corresponding ones of refracted publics are “transience, “discoverability,” “decodability,” and “silosociality.”
In persistence, “online expressions are automatically recorded and archived” (boyd, 2010, p. 45), whereas in
In searchability, “content in networked publics can be accessed through search” (boyd, 2010, p. 45), whereas in
In replicability, “content made out of bits can be duplicated” (boyd, 2010, p. 45), whereas in
Finally, in scalability, “potential visibility of content in networked publics is great” (boyd, 2010, p. 45), whereas in
Dynamics
While the three dynamics of networked publics are “invisible audiences,” “collapsed contexts,” and “blurring the public and the private” (boyd, 2010, p. 48), the corresponding ones of refracted publics are “impactful audiences,” “weaponized contexts,” and “alternating the public and the private.”
In invisible audiences, “not all audiences are visible when a person is contributing online, nor are they necessarily co-present” (boyd, 2010, p. 48), whereas in
In collapsed contexts, “the lack of spatial, social, and temporal boundaries makes it difficult to maintain distinct social contexts” (boyd, 2010, p. 48), whereas in
In blurring the public/private, “without control over context, public and private become meaningless binaries, are scaled in new ways” (boyd, 2010, p. 48), whereas in
Evolution of Influencer Culture Studies in the Last Decade
In this section, I present a brief overview of my studies on the evolution of Influencer cultures from which the framework of refracted publics and their repertoire of strategies were drawn. Having first developed original research on the earliest iterations Influencer and internet celebrity in 2009 (Abidin, 2018b), the vernacular vocabularies and concepts associated with the field has evolved and rapidly progressed alongside a burgeoning internet culture (Abidin, 2018a). As such, I categorize these studies into three stages of research and fieldsite maturity:
Influencer cultures as a job description and culture of practice;
Influencer cultures as a concept and role;
Influencer cultures as amplification platforms.
Influencer Cultures as a Job Description and Culture of Practice
My early definitional work on Influencers built on cornerstone microcelebrity studies at that time (Marwick, 2013; Senft, 2008), and had focused on their everyday, ordinary Internet users who accumulate a relatively large following on blogs and social media through the textual and visual narration of their personal lives and lifestyles, engage with their following in “digital” and “physical” spaces, and monetize their following by integrating “advertorials” into their blog or social media posts and making physical appearances at events.
A range of studies uncovering “below the radar” practices include the following: How Influencers used persona intimacy to mask the commercial transactions of their online socializing (Abidin & Thompson, 2012); how they allow their use of selfies to be parsed as vanity while quietly profiting off them (Abidin, 2016b); how they hijacked personalized hashtags belonging to prominent competitors to wrestle for eyeballs (Abidin, 2014a); how they design filler content to play-up their everyman amateur persona and play-down their rapid professionalization and commercial success to stage impressions of authenticity (Abidin, 2017a); how they used body language and narratives of romantic vulnerability to cultivate over-protective fans (Abidin, 2016a); how they staged “leaked” sex tapes to fan online gossip and bait attention before revealing the ruse in partnership with a sponsored message (Abidin, 2017c); how they engaged in provocative topics such as (unverified) under-aged sexual acts to generate controversy and retain their audience’s interest (Abidin, 2017c); how they incorporated “grief appropriate” cosmetic and fashion tutorials during a period of national mourning to maintain their content production without rousing penalties from the authorities (Limkangvanmongkol & Abidin, 2018); and how their agencies coordinated inauthentic Tweets to sully the reputation of competitor clients (Abidin & Ots, 2016).
Throughout these studies, Influencers adjust their self-presentation and contents to
Influencer Cultures as a Concept and Role
My subsequent work looked to Influencer cultures and how they are more deeply situated within an industry, their extended impact in society beyond fostering feelings of relatability to facilitate advertorials, and focused on
These studies took a meta step to assess and situate Influencer cultures within a much broader framework of “internet celebrity,” which are: all media formats . . . that attain prominence and popularity native to the internet . . . mainly known for their high visiblity, whether this be attributed to fame or infamy, positive or negative attention, talent and skill or otherwise, and whether it be sustained or transient, intentional or happenstance, monetized or not. (Abidin, 2018a, p. 16)
This update emphasized that on the other side of performers are viewers, and as such internet celebrities had to be “received, watched, and acknowledged by an audience,” even while their popularity among these audiences depend on “the platforms they use and the cultural ideologies and tastes of their intended audience” (Abidin, 2018a, p. 16).
My studies on Influencer cultures continued to develop and focus on how they construct and are situated with their own “ecology and economy” (Abidin, 2016b), and how they are epitome of internet celebrities for “rapidly professionalising in production standards, aesthetic ecologies, and financial and socio-cultural capital” (Abidin, 2019a, p. 1). The studies that spoke to “below the radar” practices include the following: How Influencers incorporate dissent and social commentary into their content through parody and satire (Abidin, 2019a) that also allow them a degree of plausible deniability; how they champion specific politics pertaining to anti-Racism (Abidin, 2019a), LGBTQ representation and advocacy (Abidin, 2019c; Abidin & Cover, 2019), and sexual health and confidence (Abidin, 2017c), and cyber-bullying (Abidin, 2019b). These works considered Influencers as “opinion leaders for young internet users,” who “communicate with followers through their effective digital strategies applied across a variety of potentially-integrated digital platforms,” and sit at “the intersection of relatability politics, attention gaming economies, and self-branding culture” (Abidin, 2019b, p. 199).
Another strand of studies looked into the larger culture of internet celebrity, including the following: Politicians’ adoption of Influencer strategies and selfies for stealth campaigning (Abidin, 2017d); meme cultures and celebrities who use popular memetic formats to embed social commentary (Abidin, 2018a, 2020d; Ask & Abidin, 2018); brands and small businesses who advertise or conduct sales on Instagram by attuning to Influencer and algorithmic cultures (Abidin, 2016c; Leaver et al., 2020); vigilante activism through public shaming and “CSI-ing” as instruments of justice-seeking reparation on Facebook pages (Abidin, 2018d); and contentious practices on tumblr that evade new content moderation guidelines (Abidin, 2019h; Tiidenberg & Abidin, 2018).
Influencer Cultures as Amplification Platforms
My newer strands of research have turned to studying the impact of Influencers on the information economy, to focus on how they are
By this time, the most mature Influencer industries were around 15-years-old, and the phenomenon of Influencer cultures had spread across the globe. Influencer cultures have evolved so much that it was no longer viable to conceptualize a singular archetype of Influencer. Instead, they are now distributed across (at least) five classes. The rapidly blossoming Influencer industries worldwide have demanded different categorical boundaries for determining the class of an Influencer, namely, country, culture, platform, and genre. For instance, in my research on Influencer cultures in the Nordic, it was important to account for specific “classes” of Influencers scaled to the population of the country (Abidin, 2019g). In countries like Denmark (population ~5.8 million) and Sweden (population ~10.2 million), a “mega-Influencer” is one with over a million followers, whereas newbies and aspirants with fewer than 1000 followers can also be considered “nano-influencers” (see Table 2). Subsequently, an Influencer’s class determines their primary role in the information ecology, whether this be serving as amplifiers of information (akin to broadcast media), opinion leaders, or persuasive converters (~word of mouth marketing). In place of country/population, we may also assess the class of Influencers by other culturally bounded categories such as language group (e.g., across the ~477 million Spanish-speakers worldwide, mega-Influencers may need to command at least 10 million followers).
Influencer Classifications for Denmark and Sweden.
Third, we can assess an Influencer’s class by the basal viewership on the platforms where they are hosted. For instance, on mainstream and widely used platforms like Instagram with more than 1 billion monthly active users, the figures in Table 2 will need to scale up drastically; whereas on lesser known or newer or more niche platforms like BIGO live with over 22 million monthly active users, the figures in Table 2 will need to scale down slightly. Finally, Influencers can also be classified by their genre. The threshold for mega- and macro-Influencers will be significantly higher for popular categories such as gaming, and significantly lower for newly emerging categories such as slow-living silent vlogs (Abidin, 2020b). Across these studies, the focus in Influencer cultures has shifted from their persona strategies and self-branding, to their ability to cut through the white noise of an increasingly saturated internet space to amplify a message and deliver it to specific audiences.
Alongside social media Influencers, other varieties of internet celebrity are also functioning as amplification platforms. Two further examples from my newest research are “meme factories,” which are a coordinated network of creators or accounts who produce and host content that can be encoded with (sub)text and parsed into new contexts across multiple organizational structures (Abidin 2020d; Abidin & Cheong, 2019), and K-pop fandom networks on social media, who often deploy analogue and algorithmic baiting strategies through large-scale but informal coordinated action to suppress some news (e.g., idol scandals) and magnify others (e.g., idol promotions; Abidin, 2019d).
Refracted Publics Strategies
In this final section, I draw from the selection studies above to briefly highlight some strategies utilized in cultures of refracted publics. It is hoped that future work on refracted publics will situate these strategies within legacies of scholarship on human interventions and circumventions of algorithms and algorithmic imaginaries (Bishop, 2019, 2020; Bucher, 2017, 2018; van der Nagel, 2018; Willson, 2017), content moderation practices and circumventions (Gerrard, 2018; Gillespie, 2018; Parks & Mukherjee, 2017; Roberts, 2019), and analogue visibility labor practices (Abidin, 2016c; Baym, 2018; Duffy, 2016; Neff et al., 2005; Wissinger, 2015).
This list is non-exhaustive nor exclusive to Influencer cultures, and expected to grow as studies of “below the radar” cultures develop. It includes:
Self-Amplification Groups: Instagram Pods, Twitter Decks, and Facebook Circles
As platforms like Instagram, Twitter, and Facebook no longer display all posts in our home feeds nor organize them chronologically by default, Influencers are struggling to maintain engagement rates on their posts. Furthermore, bots and bought accounts have been disallowed by platforms like Instagram, with such automated coordination also denounced by the Influencer industry at large. To cope, Influencers have formed
Hashtag Jacking: Content Spamming, Grief Hypejacking, and Manufacturing Trending
As attention is a scarce commodity on the internet (Goldhaber, 1997), users may use
Among Instagram cultures, a version of hashtag jacking known as “grief hypejacking” is especially popular during global grieving events, where in Instagrammers “bandwagon[n] on high visibility hashtags or public tributes [. . .] to misappropriate highly public channels of collective grief for self-publicity” (Abidin, 2018c, p. 169), such as promoting their microcelebrity accounts, pushing advertorials, or hawking wares.
K-pop fandoms on Twitter are also masters of manufacturing trending hashtags and keywords. In the lead-up to an idol’s debut, comeback, birthday, or other special occasions, it is not uncommon for fan networks to distribute detailed schedules of and by-the-hour timelines of planned social media posts comprising specific keywords, phrases, and unique hashtags. To maximize the trendability of hashtags and ensure that they are not prematurely spoiled, instructions usually comprise small circumventions such as symbols and spaces prior to the debut of a hashtag. An example of these calls is: “Please use ‘#/WelcomeBackOurBestIdolXX’ for our Tweets on DDMMYY at HH:MM GMT+8 to celebrate Best Idol! Remember to remove the slash before posting, and do not post before the exact time so that we ensure that we trend!” At times, these campaigns also involve some playful deception through hoaxes and manipulated images—such as the Lady Gaga and BLACKPINK Starbucks hoaxes which circulated fake news that streaming a new song would entitle listeners to free Starbucks drinks (Abidin, 2019d)—to trick curious K-pop fans or unwitting Twitter passersby into amplifying information or boosting the visibility of hashtag trends.
SEO Spotlighting and Shadowing: Witching Hour, Screengrabbing, Internet Paralanguages
Influencers also
The second method is screengrabbing, where followers may capture an image of a post and circulate/repost the image for others to access without the need to drive traffic toward the source, akin to social media scholar Emily van der Nagel’s (2018) work on “voldemorting,” or the refusal to mention “words or names in order to avoid a forced connection.” This is usually done by anti-fans who engage in lengthy discussions of Influencer cultures on internet forums.
In instances where platforms and governments censor specific keywords, users may also replace words with emoji or substitute blacklisted words with homonyms to evade detection. While there are niche examples across the Influencer subcultures I study, my favorite example is a study of the #MeToo movement in China by digital media scholar Meg Jing Zeng (2018), who traced how Chinese social media users replaced “#MeToo” with translated homonyms, such as “#米兔” (mi tu, lit. translation “rice bunny”), and emoji such as “#[rice][bunny]” to avoid detection.
Social Steganography: Vague-Booking and Insta-Vagueing, Dog Whistling and Parallel Literacies, Breadcrumbs and Easter Eggs, Memes
Dog whistling and parallel literacies involve crafting messages that are encoded with multiple meanings. While the former results in messages that mean one thing to the general public with additional meaning(s) for the target audience, the latter is a practice of deliberately encoding a kaleidoscope of messages such that a single piece of content will be interpreted in vastly different and specific ways for by different target audiences (Abidin, 2019a, p. 15). An exemplar of parallel literacies is the body of work by Singaporean Malay YouTube Influencers MunahHirziOfficial whose Beyoncé parody videos are painstakingly conceptualized to include different subtexts for the minority group of Malay Muslim viewers, the predominantly Chinese Singaporean majority, and the global network of Beyoncé fans at once to maximize traction (Abidin, 2019a).
To generate buzz and activate speculative conversation that is usually active on internet forums and comments sections on social media, many Influencers also take to leaving breadcrumbs and Easter eggs in their contents for followers to uncover. Breadcrumb trails involve leaving disparate hints across various social media estates that send followers on a hunt to collect and corroborate information to decipher a message. Easter eggs involve leaving recurring hints across a series of posts on one social media platform that serve as teasers to a big reveal. Among my informants, breadcrumbs are usually used to soften breakup announcements through the conspicuous absence of an Influencer’s romantic partner in social media updates over a period of time, and Easter eggs are usually used to build anticipation toward new releases or pregnancy announcements.
Among vigilante activist groups and politically active citizens in my research on politicians’ adoption of Influencer techniques (Abidin, 2017d; Leaver et al., 2020, pp. 154–157), memes are a popular way of encoding criticism toward politicians and the government with the allowance of plausible deniability to worm one’s way of out punitive sanctions. For instance, in the wake of the terrorist attacks in Paris in November 2015, Singaporean Member of Parliament Baey Yam Keng posted “a rather gratuitous picture of himself posing with arms akimbo next to the Eiffel Tower” on Facebook to accompany a caption expressing condolences (Leaver et al., 2020, p. 155). Citizens felt his “superhero power pose” was disingenuous and memed him into backdrops of popular tourist destinations worldwide in the vein of “Where’s Wally” templates to express displeasure. Influencers also partook in the meme creation and amplified the incident to the consciousness of the general public, which eventually resulted in a sincere apology from the minister.
Sentiment Seeding: Meme Factories, Astroturfing, Internet Brigade
A popular mode of “below the radar” practices that has long been covered in popular media articles and marketing research is content farming (e.g., Bercovici, 2013; Buzz, 2015; Chace, 2011; Gross, 2011), where companies employ freelancers to produce content in mass quantities intended to optimize algorithmic visibility and rank higher in SEO indexes. However, unlike intentions to flood or manipulate search indexes,
Meme factories, which are a coordinated network of creators or accounts who produce and host content that can be encoded with (sub)text and parsed into new contexts across multiple organizational structures (Abidin, 2020d; Abidin & Cheong, 2019), are also effective vehicles of sentiment seeding. They can occur as online forums (Borgerts & Fielitz, 2019; Cohen & Kenny, 2020), such as Reddit (Donovan, 2019) and 4chan (Knuttila, 2011), meme-aggregator websites (Chen, 2012, p. 13), and networks of actors, such as Anonymous (Jarvis, 2014). In my fieldsites of Singapore and Malaysia, sentiment seeding through meme factories can occur on Instagram and Facebook through formal and commercial meme businesses, individual or small collectives of hobbyist memers, or large-scale crowd-sourced user-generated memes (Abidin, 2020d). Meme factories use strategic calculation to obtain virality or active a call to arms to seed decision-swinging discourses, at times with the potential to commercialize their meme contents for sponsors. From my fieldnotes, examples include the (sponsored) use of Oreo cookies in YouTube Oreo challenges to promote the product, and the use of comics and illustrations to promote mask-wearing during COVID-19 (Abidin, 2020d).
Another sentiment seeding strategy is astroturfing, or “the orchestrated promotion of sponsored messages in the guise of unsolicited, non-commercial casualness, and without any indication or disclosure that these sentiments and messages are in fact calculated campaigns” (Leaver et al., 2020, p. 113). Among Influencers on Instagram and Twitter, this has taken the form of coordinated inauthentic action in which posts maligning a client’s competitor are crafted to pass as organic sentiment (Abidin & Ots, 2016). Among revenue-earning users on tumblr, this has taken the form of coordinated Asks and conversational threads that promote specific brands and products (Abidin, 2019e).
Among politically inclined users in Singapore, an “internet brigade” refers to a coordinated group of internet users who manage fake profiles and provide “counter viewpoints online that disagree with or criticise government policies” (Lim, 2018). While these groups are usually secretive and operate pseudonymously, it has been documented that they have been active since at least since 2007 (Romero, 2018) and became more public in 2012 (Low, 2018; Xu & Lee, 2018) to “giv[e] an artificial impression that [Members of the Parliament in the incumbent party] are well-liked, and that [the incumbent party’s] policies are well-loved by its citizens” (Low, 2018). It is unclear whether these “militant” and “counter-insurgency” groups are paid actors (Xu & Lee, 2018). Similar groups have been studied around the world, including the 50 Cent Party in the People’s Republic of China (Jing, 2016; Keating, 2011; King et al., 2017; Yang et al., 2015) and Cyber-troopers in Malaysia (Guest, 2018; Hopkins, 2014; Johns & Cheong, 2019; Wahab, 2019).
Clickbait: Controversial Subjects, Optical Illusions
Another analogue strategy is the use of
In my fieldwork, clickbait generally takes two forms. Among Influencers, topical clickbait constitutes the use of sensational topics to lure viewer interest, such the use of sex-related discussions and images as “sex bait” (Abidin, 2017c), the use creation of “are they, are they not” suspense in playfully hinting at non-heterosexual sexualities as “queer bait” (Abidin & Cover, 2019), and the use of evocative language to spark viewer interest and generate public debate in “call out cultures” (Abidin, 2018d).
A second form is platformed clickbait in which users rely on small optical illusions to deceive viewers into generating traffic. In my research on content creation Facebook pages, three optical illusions are especially prominent: First, the use of fake video play buttons, wherein play buttons are superimposed onto static images to give the impression of preview video thumbnails. These encourage users to click on the fake play button to start the video, only to lead them to external websites and add to ad revenue (Constine, 2014, 2017; Hutchinson, 2017). Second, passing off GIFs as live videos when embellished with fake live buttons, fake viewer counters, and audio files playing on loop. These extend and retain viewer traffic and augment engagement statistics (Grigonis, 2017; Mezzofiore, 2017). Third, the creation of floating videos wherein an overlay of translucent floating triangles or arrows (also referred to as “dust sparkles” by my informants) are superimposed on static images to trick algorithms into parsing them as videos. These increase the visibility of such posts as videos are more algorithmically visible than static images (Charles, 2017; Lorenz, 2018).
Conclusion
Refracted publics are vernacular cultures of circumvention strategies on social media in response to both analogue and algorithmic vision and access. They have been mobilized to avoid detection, promote deflection, and facilitate the dissemination of specific messages away from or toward target audiences. Refracted publics arise from the 2010s media culture of perpetual content saturation, hyper-competitive attention economies, gamified and datafied metric cultures, information distrust, and infodemic fatigue. They facilitate and maintain the conditions of transience, discoverability, decodability, and silosociality, under the dynamics of impactful audiences, weaponized contexts, and alternating publics and privates. They present as self-amplification groups, hashtag jacking, SEO spotlighting and shadowing, social steganography, sentiment seeding, and clickbait. Refracted publics are cultures of strategies that are organized by public/pseudonymous/anonymous users in concealed spaces “below the radar,” and orchestrate seemingly organic results from “below the radar.” They allow cultures, communities, and contents to avoid being registered on a radar, register in misplaced pockets while appearing on the radar, or register on the radar but parsed as something else altogether. They are the strategies of private groups, locked platforms, or ephemeral contents that will continue to thrive alongside the internet for decades to come.
Footnotes
Declaration of Conflicting Interests
Funding
ORCID iD
Author Biography
.