As I put the finishing touches on this paper in September 2020, the short-video app TikTok is officially three-years-old and already among the most downloaded apps, ranking 7th of all-time downloads worldwide in the 2010s (App Annie 2019). TikTok is the international derivative of Douyin (抖音), which was first launched in September 2016 by Beijing-based parent company ByteDance (字节跳动) for the Chinese domestic market. Subsequently in November 2017, ByteDance bought over and integrated a predecessor competing short video app
Companies have reportedly paid TikTokers up to $20,000 per branded post (Influencer Marketing Hub 2020), and industry observers claim that a leaked TikTok pitch deck from June 2019 price in-app advertising at between $25,000 to $150,000 (Hines 2020). High-earning TikTok Influencers have been profiled by the likes of
Thematically, there are emergent but limited academic studies on the digital cultures of TikTok. Research has thus far focused on personality traits (Omar & Wang 2020) and psychology of users (Kumar & Prabha 2019), uses of humour (Wang 2020), subcultures of punk behaviour (Mackenzie & Nicholas 2020), and uses for student engagement (Pavlik 2020) and scientific public engagement (Hayes et al. 2020). Emergent research has focused on privacy issues (Neyaz et al. 2020) and online hate (Weimann & Masri 2020) facilitated by TikTok. Most notably, TikTok has also been credited for facilitating the spread of credible public health messages during COVID-19 (Basch et al. 2020, Chen et al. 2020, Eghtesadi & Florea 2020, Kennedy 2020, Sidorenko-Bautista et al. 2020). However, understandings of TikTok and Douyin are limited, with a handful of these initial studies claiming to focus on one app (e.g. TikTok) when they have in fact studied the other app (e.g. Douyin), resulting in a conflation both apps (e.g. Zhang 2020).
Perhaps the most relevant studies to internet celebrity cultures on TikTok, at present, are: Digital media scholars Kaye et al.’s (2020) comprehensive walkthrough of TikTok’s (and Douyin’s) platformization and features as of September 2019, in which they found that ByteDance is ‘clinging to a gift economy model of monetization, and relies predominately on live streaming and virtual currency exchange’ (2020, p. 20); Feminist media scholar Melanie Kennedy’s (2020) short essay on how the popularity of teen girl TikTokers during COVID-19 underscore the ‘continuation, and indeed intensification, of the construction of femininity and celebrity in contemporary girl culture’; and sports management scholars Su et al.’s (2020) study of how athletes’ used TikTok as a ‘self-branding outlet’ by engaging in viral trends and posting encouraging messages during COVID-19.
In the vein of this very new trajectory of research into internet celebrity cultures on TikTok, this paper offers an exploratory study of TikTok’s ‘attention economies’ (Goldhaber 1997), paying special attention to ideas around practices of visibility labour (Abidin 2016) and internet celebrity (Abidin 2018) as generated ‘on the ground’. This responds to the Special Issue theme of ‘open literacies’ by concentrating on how such actions are ‘bottom-up’ rather than ‘top-down’, with the potential to ‘change the rules of the game’ (Hartley 2019), specifically, the attention economy of TikTok. In this scoping paper, we delve into the methodology undertaken to understand internet celebrity practices on TikTok, with subsequent sections situating internet celebrity on TikTok, mapping out attention economies on TikTok, and investigating visibility labours on TikTok. The paper closes with some considerations on the newest introduction of the TikTok Creator Fund as the next frontier of internet celebrity on the app.
The research presented in this paper is part of a longer multi-year ethnographic project investigating the role of social media Influencers as conduits of knowledge in the Asia Pacific region (wishcrys 2020), specifically how they are experts in holding followers’ attentions and amplifying specific contents. This project employed three primary methods:
Firstly, digital ethnography was conducted from June 2019 to August 2020 to observe platform and user cultures and norms. This entailed extended online immersion periods on TikTok for at least an hour each weekday, selectively following/unfollowing and liking/unliking posts in order to experiment with the algorithmic triggers, explore various subcultures and genres on the app, and map out trending and viral activity through fieldnotes (e.g. written notes, screengrabs, downloads, etc.). During this period, there were also multiple and frequent spurts of intense activity where daily observations took place for between 3 and 6 hours, usually when there were controversies or time-sensitive incidents unfolding among observed networks of TikTok creators on the app. The screengrabs and mentions of TikTok posts in this paper were included only after a careful evaluation that the creators’ had indeed intended for their posts to be widely circulated, received, and audienced. This was ascertained through a combination of their hashtagging practices that maximized searchability, the tonality in their captions and overlay text that addressed a ‘superpublic’ (boyd 2013) audience, and their profiles that exhibited logics of pursuing ‘high visibility’ (Abidin 2016). For a full playlist of the TikTok posts discussed in this paper, see Appendix 1.
Secondly, traditional anthropological participant observation took place between July 2019 and December 2019, at various fieldsites in Hong Kong, Seoul, Shanghai, Singapore, and Sydney, to observe industry forecasting and strategizing of content generation on TikTok. This entailed immersive in-person participation and observation at various Influencer events, such as client product launches, Influencer road shows, industry panels and conferences, and office visits. Through these activities, I had the opportunity to shadow or accompany Influencers, Influencer managers, or staff from digital media/advertising firms as they went about their work, and visit Influencer agencies and digital media/advertising firms to tour their premise and learn about their business models. Subsequent fieldwork at six further cities to observe industry road shows and conferences about TikTok, Influencer agency workshops for TikTok training, and Influencers’ content generation processes for TikTok have been postponed in light of COVID-19.
Thirdly, interviews took place in person and (in lieu of COVID-19) via phone or video calls between August 2019 and July 2020, lasting between 30 and 90 minutes, with 23 people. This included 3 interviews that took place in pairs, and 18 individual interviews; 1 person was interviewed twice, first in a pair, and subsequently individually. Together, the interviewees represented 14 Influencer agencies or digital media/advertising firms, and 5 Influencer profiles or channels in Australia, Indonesia, Japan, Malaysia, Singapore, South Korea, and Thailand. The interviews that informed the analyses in this paper were conducted in English, transcribed, and coded via a grounded theory schema.
To scope and explore the emergent internet celebrity cultures on TikTok, the data and analysis that I have elected to prioritize in this paper focuses more on the empirical studies via digital ethnography, although the analyses and insights offered are also informed and supported by the traditional anthropological observation fieldwork and personal interviews.
As social media platforms are launched rise, and plateau with each generation, cohorts of users usually emerge as prolific users who are the most visible of the lot: they are often the leaders of trends within subcultures, and perhaps even the highest earners through brand collaborations and ad sales. In the case of TikTok, what does internet celebrity look like, and does this differ from predecessors like Instagrammers, YouTubers, bloggers, and the like?
As I note in prior research (Abidin 2018), the crux of internet celebrity––and consequently Influencers–– is their ability to retain high visibility online. The quality of their internet celebrity is constituted by pairs of elements which can be applied in varying degrees. For instance, internet celebrity can be a result of fame/infamy, positive attention/negative attention, talent and skill/otherwise; the shelf life of internet celebrity too can be sustained/transient, intentional/ by happenstance, and monetized/ not monetized. Most crucially, all of these displays of internet celebrity have to be acknowledged by an audience, the reception of which will also differ based on the attention norms and algorithmic preferences of the platforms on which the viewing and interactions take place, and the cultural ideologies and tastes of the audience. Where internet celebrity represents the quality of an online user’s visibility, an Influencer is a monetizable status and potential career that one can pursue as a result of their internet celebrity. Internet celebrities who progress into becoming Influencers tend to adopt a specific tonality and morality in the element pairs, focusing on fame built on positive attention due to a talent or skill, and that is intentionally aimed to be sustained and monetizable. Considering this, this section aims to understand how internet celebrity is situated on TikTok given the platform’s features, affordances, and norms.
Sustained and long-term fame on the likes of earlier social media like Instagram and YouTube used to be based on a coherent persona or online identity, or the making of an online brand that is memorable and after whom followers would aspire. Followers could subscribe to these Influencers, continuously keep abreast of their updates, feel a degree of interest and investment in learning about their personal lives on and offline, and be persuaded of the (sponsored) messages that they share as a result of the prolonged communicative intimacies (Abidin 2015) that they share. Such was the era of persona-based or profile-anchored fame.
On a space like TikTok, the nature of fame and virality has shifted, and tends to be based on the performance of users’ individual posts. Most TikTok users strive to have individual posts accumulate ‘engagements’ in the form of views, comments, and shares, as encouraged by TikTok’s culture of aiming to be picked up for and catalogued by the For You Page. In this way, TikTok users and internet celebrity aspirants do not always conscientiously maintain a single singular coherent persona or style, but instead are actively and very quickly adapting from the latest trends and viral practices on TikTok, to attempt varieties of styles––across hashtags, keywords, filters, audio memes, narrative memes––to aim for the For You Page, or the ‘golden ticket’ that would allow one to gain an immense number of followers overnight. The platform logics of TikTok force internet celebrity aspirants to actively seek out, learn, participate in, and engage in these what is ‘going viral’ at the moment in order to remain visible to others on the app, unlike earlier platforms that may accommodate esoteric subcultures and communities who can find their footing and niche regardless. This has in turn sparked off the norm of TikTok users who attempt to introduce new trends that may fly or flop, wax and wane, depending on the flavour of the moment. Such is the era of post-based fame.
Where visual social media such as Instagram tend to focus on images over text (Leaver et al. 2020), TikTok privileges sounds over images. A feature on all TikTok posts is a round rotating button on the bottom right hand side, which clicks through to a catalogue that pins to the top the original video that first debut the audio clip on TikTok, an option to ‘use this sound’, and a list of all the posts that have ever used the same clip. Clicking on ‘use this sound’ enables users to create their own video with the same background audio template, which they are able to embellish by adjusting the volume or overlaying their own dialogue when creating their post. The ‘templatability’ (Leaver at el. 2020) of background audios have been central to viral trends on TikTok, and are perhaps the most novel feature on the app. It can be understood as an ‘audio meme’, and is the driving template for content production on TikTok.
Audio memes are not limited to an actual snippet of a song or spoken dialogue, but can include variants based on remix and delivery styles as well as ownership of the template. Audio memes are an organizing principle for how content is catalogued into repositories on TikTok, and how users navigate the platform to seek new trends and contents. In addition to reusing and remixing audio clips and music, TikTokers also engage with other original audio templates through ‘duets’. Duets juxtapose videos side by side and can be viewed in tandem, and are often used by TikTokers to ‘react’ to or ‘reply’ to an original video, whether replicating it for comparison, or adding commentary as compliment or critique. In some instances, TikTokers also have dedicated accounts to generate original audio clips with the intention of having others use them in their videos.
On TikTok, audio memes and the texture of sound take on an intimate disposition, requiring care, tact, and wit to situate and decipher: TikTokers often rely on the lyrics of specific songs to tell a story (e.g. when the lyrics are central to lip-syncing, when the punch line of a video is a specific lyrical line in the song); consider the musical and rhythmic shape of a tune to advance the storyline of their video (e.g. when a beat ‘drops’, when a song transitions to ‘bad recorder playing’); complement or juxtapose audio memes against video content and textual captions (e.g. song to provide ambience for the storytelling, song to change the tonality of storytelling to sarcasm or parody); or organize and streamline content into specific silos.
However, accompanying the complexity of these audio memes is often sophisticated and clever editing work that TikTokers employ to drive the visual narrative of their post. On TikTok, these edits and cuts are known as ‘transitions’, and TikTokers who display this technical expertise are often lauded and celebrated on the app (see also similar observations on Vine star Zach King (Mattise 2020; Rayner 2020)). The most popular of these transitions (at the time of writing) are fashion transitions where TikTokers quickly switch outfits (e.g. @the.navarose), and optical illusion transition videos where TikTokers engage in magic tricks (e.g. @amen_716; @zachking). This has sparked off a now-established routine of virality on TikTok, where every time a new video transition (and accompanying audio meme) is introduced and popularized on the app, some TikTokers would swiftly follow up and make ‘tutorial’ versions of the transition to teach others how to replicate them, and in turn attain high visibility and virality for their posts.
Like the audio meme catalogue, visual filters and hashtagged words in the captions of TikTok posts also sort posts into streams. These lists are roughly ranked by the number of ‘engagements’ on the post, measured by views, comments, and shares. The ranking is not strictly chronological or sequential, but tends to group videos by ‘batches’ of engagement counts, with the videos raking in the high millions in engagements, followed by the low millions, high 100,000s, low 100,000s, high 10,000s, low 10,000s and so on. From the perspective of meme ecologies, the ability to trace the history and use of a template is an unprecedented record of documenting linearity in meme cultures.
However, there are often contestations over the ownership of audio memes, and each stream is at best one account of an audio meme’s authoritative origin and often exclude spin-offs. This is because TikTok users have been known to replay, re-record, or re-upload the same audio meme under a different title under their accounts, replicate someone else’s voiceover without attribution, or intentionally misattribute the ownership of an audio meme. Some TikTokers have also circumvented being catalogued into the platform-automated streams by strategizing around refusal, manipulation, soundjacking, and other practices.
Further, there are often competitive ranking sprees known as ‘chart jacking’ wherein TikTokers encourage others to engage with specific posts to get them to ‘climb’ the ranks on a stream, to overshadow posts that they want to suppress, or to negotiate the ‘mainstream’ use of a specific audio meme or filter. These ‘legacy wars’ demonstrate the importance that TikTokers place on the ‘tonality’ assigned to each of these audio and filter templates, and the complexities of attention economies around templates, and will be elaborated on later in the paper.
In addition, it is important to note that many of TikTok’s newly introduced and updated features and affordances mimic after its predecessor sister-app Douyin, also managed by parent company ByteDance. In this project’s longer-term digital ethnography of both Douyin and TikTok, it has been observed that several of the new features recently added to TikTok––such as livestreaming (Delfino 2020)––were features carried over from the original sister app Douyin. As such, in this section, I briefly review the distinctions between both platform’s features and norms, and highlight some of the distinctive features currently active on Douyin that may be incorporated into TikTok in time to come. To complement Kaye et al.’s (2020) September 2019 walkthrough (Light et al. 2016) of TikTok, this section similarly adopts an in-depth walk-through of Douyin and TikTok, but elects to highlight only the key differences in platform features and norms at five key stages of use:
When
When
When
When
When
In summary, while the features of Douyin encourage positive reputation building, the localization of content, commercial exchanges, curation of geo-location destinations and shopping list, and groupchats; those of TikTok (at the time of writing) encourage a degree of pseudonymity, ‘discoverability’ (Lobato 2018) of content based on an ‘algorithmic blackbox’ (Bucher 2017) of user preferences and history of actions, a focus on content creation rather than commerce, a more in-depth exploration and use of the in-app features, and dyad messaging. It can be expected that TikTok’s anticipated features will adopt from Douyin’s established features, facilitating the ease of TikTok users to grooming their internet celebrity and progression into potential Influencer careers.
Having established the norms of internet celebrity on TikTok, in this section we consider some of the key attention economies on TikTok. Physicist Michael Goldhaber (1997) first popularized the notion of ‘attention economies’ to explain that in the new realities of information overload, human concentration is limited, and attention is a form of cumulative wealth. He explains that ‘getting attention’ requires one to ‘build on the stock you have every time you get any, and the larger your audience at one time, the larger your potential audience in the future’ (1997: n.p.). In brief, Goldhaber offers that people with the ability to attract and sustain the attention of others are ‘in a preferred position to get anything this new economy offers’ (1997: n.p.).
In subsequent work, he reiterates that the attention economy is the ‘natural economy of the internet’ and ‘increasingly all of society’ (Goldhaber 2006: n.p.). Goldhaber asserts that the attention economy is a system that involves ‘paying, receiving, and seeking… the attention of other human beings’, which is ‘intrinsically limited and not replaceable by anything else’ (2006: n.p.). Since the widespread uptake of his concept, Goldhaber clarifies that while many readers have adopted a business perspective to his work and conflated the attention economy with ‘money, markets and standardized industry’ (2006: n.p.), this is simply an extension or application of this theory, and trading in attention is ultimately about ‘activating’ the consciousness of other people to associate and connect ‘memories’ and ideas to oneself. He offers that ‘true modesty or humility’ is difficult to maintain in the attention economy where people engage in displays to be noticed by others, even when anonymity is deployed.
While many scholars have since taken up the concept and studied it in various contexts, sociologist Zeynep Tufekci’s (2013) study of social movements offers that in the social media age where attention can be produced and distributed widely through participatory media, there are also specific ‘pathways’ to acquire attention. In this vein, this section reviews three broad inclinations on TikTok which govern attention spans from users on the app.
The rapid ascend of TikTok––especially in the wake of the mass influx of new users during increased internet use over the COVID-19 self-isolation periods––seemed to be received in trepidation by competitor apps. Prior to TikTok’s launch, Instagram was once upon a time the most used app among teens (Wolfe 2018). In response to their fast competition, Instagram began to profusely advertise itself on TikTok through a range of sponsored ads, targeted at luring users from TikTok (back) to Instagram. These included ads in the guise of TikTok posts that promoted the Instagram feature of Stories (
Figure 1
(L–R)

This corporate luring and siphoning of users from TikTok (back) to Instagram is further unscored in the design of TikTok in-app advertising: Ads embedded within one’s FYP would include conspicuous banner pop-ups in the footer (
Figure 2
(L–R)

In the wake of TikTok’s massive growth, rival apps soon launched competitor features, including Instagram’s Reels (Alexander 2020a) and YouTube’s Shorts (Perez 2020) – the former of which explicitly began roping in prominent TikTokers to seed content on and popularize the use of Reels among their followers. This quickly eventuated in serial crossposting and duplication across apps, wherein TikTokers would simply repost the same TikTok videos on Instagram Reels and YouTube Shorts, resulting in a swift pilfering of iconic TikTok content genres and aesthetics to Reels and Shorts.
The emergence of a class of internet celebrity and Influencers on TikTok––facilitated by its platform logics and features of visibility labour and reputation building––seemed to usher in a new frontier of social media. The way attention works online was undergoing a seismic shift, with TikTok Influencers pioneering new formulae for how one might ‘qualify’ to be a highly visible and thus successful creator on the app: The ‘staging’ of an ‘Instagrammable’ lifestyle that was aspirational and pristine, seemed to give way to the ‘crafting’ of a relatable performance that was entertaining and accessible. Further, the explosion of diversity in content genres on TikTok tended to favour discursive content and performance talent over the Instagramesque physical appearance of simply ‘posing for a photo’, which swiftly diluted the industry’s fixation over the archetypal ‘young, skinny, beautiful woman’ as the ‘default Influencer’. In short, TikTok seemed to be facilitating a wide-scale demoticization (Turner 2010) of the Influencer industry. But as the COVID-19 season unfolded, these shifts seemed to spread beyond TikTok to apply to the Influencer industry at large.
In light of curtailed international and localized travel during COVID-19, the decline of in-person Influencer events, and overall budget cuts on Influencer marketing spending (Abidin et al. 2020), Influencers found themselves strapped for resources and needing to creatively pivot or adapt to home-based activities to generate content. While many of these norms were already beginning to be established on TikTok, they swiftly emerged on the likes of Instagram Stories and were accelerated during the season of COVID-19. In general, three major industry and systemic shifts were observed (Abidin forthcoming;
Figure 3
Systemic shifts in the Influencer industry with the emergence of TikTok, and subsequently in the industry at large during COVID-19. Image by the author.

First, the parasociality or ‘communicative intimacies’ (Abidin 2015) between Influencers and followers, which is negotiated through audience feedback, has become more reliant on performances of relatability from Influencers rather than their incitement of feelings of aspiration among followers. As observed in the comments section in various Instagram threads observed during longitudinal digital ethnography, followers are craving for more personalized interactions, and Influencers are emphasizing their ‘calibrated amateurism’ (Abidin 2017) to reinvoke imaginaries of their grassroots ethos and humble beginnings.
Second, the aesthetics of Influencers’ content, which is fashioned to pander to the attention economy and factors like ‘Instagrammability’ (Leaver et al. 2020), has shifted towards being more reliant on the sharing of discursive content and away from merely distracting or satisfying followers with visual theatrics. This has involved Influencers who appear to be moving away from ‘picture perfect’ content, and towards more personalized disclosures and storytelling, which highlights their continued role as opinion leaders alongside being amplifiers of sponsored messages.
Third, the accessibility that followers feel towards Influencers, which is shaped by platform-based (e.g. metrics, trendability) and thematic (e.g. quality of content, relevance of content) barriers of entry, has allowed for more diverse presentations of the self and privilege physical embodiment less. As content shared by Influencers during COVID-19 has become more casual and increase in volume, a variety of types and genres of Influencers is sighted, including Influencers whose online fame is not entirely contingent upon their body image, but instead more focused on their content genres and relational styles.
In the climate of ‘call-out cultures’ and ‘cancel cultures’ TikTok has allowed young people to become politically engaged in a format that is entertaining, educational, and palatable among their peers. While social justice pursuits may sometimes involve sophisticated persuasion and the organization of mass rallies and displays (Bogle 2019), in other instances, this takes on the guise of meme-making, trend-setting, virality-seeking, or public-shaming. Taken together, the ‘performativity of YouTube, the scrolling interface of Instagram, and the deeply weird humour usually reserved for platforms like Vine and Tumblr’ (Abidin in Bogle & Edraki 2019) enabled young people to creatively strategise over personalising their message for young peers across various subcultures and interest groups. This has lowered the barriers of entry and expanded the repertoire of what it means to be ‘politically active’, placed value on creative social media skillsets that young people feel confident in, and normalized the idea that being politically involved is not a niche but can be an everyday staple in one’s social media diet. As a result, posts that engage with some degree of social justice on TikTok tend to be favoured, and many TikTokers engage in viral duets and replies that publicly shame other TikTokers as a route to internet celebrity.
While such political engagement may feel contentious, research has shown that social media posts and comments sections can be discursive spaces that serve as ‘a communicative space... in which alternative viewpoints have the potential to circulate and be negotiated’ (Uldam & Askanius 2013: 1190). In times of conflict, users have the potential and ability to convert aggression into ‘teachable moments’ (Johnston 2017: 85), and ‘affectively reiterate their personal politics’ (Abidin 2019: 622) towards various advocacy causes. The value of such ‘discursive tactics’ or ‘tactics focused on communication’ (Clark 2016: 790) is in the capacity to show support for a cause, counter opposing sentiments, and be engaged in ‘the maintenance of social movement networks’ and the ‘political aim of changing discursive perceptions, norms and ways’ (Shaw 2012: 382). These are, afterall, TikTokers employing the lingo and discursive tactics popular to TikTok to speak to their intended audience of fellow TikTokers.
Between 2019 to 2020, many young people have used TikTok to mobilise social action, galvanise social change, and institute new peer cultures of learning, with networks of young TikTok creators who were progressively or suddenly accumulating visibility and public recognition for their advocacy and activist content. For instance, the global climate change rallies aroused by the then-16-year-old environmental activist Greta Thunberg sparked off chains of school strikes and public protests of young people calling for political action to combat climate change –– these were largely organised and mobilised on TikTok through time-lapse videos comparing 2019 to 3019, body art depicting a rapidly depleting landscape, and make up art representing the consequences of polluted oceans on wildlife (Bramwell 2019). Similarly, during the Australian bushfires in January 2020, TikTok became a site for informative updates, peer teaching, and using humour to cope. Many young users turned to TikTok to share live footage from the ground, in the vein of citizen journalists (Wilson 2019), while still others strategically embedded criticism and critique against inactive politicians and government policies in their TikTok videos (Judge 2020). Australian firefighters also took to TikTok with viral dancing videos (Rodd 2020) and a brief uplifting respite while combating the flames on the ground (Mirchandani 2020).
TikTok has also been used by young people to call out racism and race-based violence. In November 2019, then 17-year-old human rights activist Feroza Aziz (@officialferozaaziz 2020) posted a make-up tutorial to raise awareness of Muslim Uyghurs being confined to detention centres in China. The video recording that depicted Aziz curling her eyelashes and applying makeup cleverly disguised her audio commentary describing the oppression. However, the video was deleted and her account suspended by the platform (ABC News 2019). Although there is some contestation from the platform regarding their censorship––including explanations that there was a ‘human error’ (ABC News 2019) and a prior offensive video (Ma 2019)––Aziz’s call out on other platforms (@ferozaazizz 2019) and TikTok’s eventual public apology incited global discussions around the potentials and pitfalls for activism on the app. Aziz subsequently produced a similar video addressing the Citizenship Amendment Act in India that is feared to discriminate against Indian Muslims (Jain 2019).
Recalling that internet celebrity is contingent upon high visibility on the internet, this section reviews some of the key strategies in which TikTokers engage to solicit the attention of followers. Visibility labour is the work that social media users perform to be noticed by their intended audiences, comprising self-posturing and the curation of self-presentations to be ‘noticeable and positively prominent’ among viewers (Abidin 2016). Given that TikTok’s propriety algorithmic is a well-guarded black box (Alexander 2020b), TikTokers have had to rely on repeated attempts, observed patterns, and gut feelings to figure out how the algorithm works, how to please the platform to facilitate their visibility, and how to have their popularity grow.
This sort of ‘guestimation’ and envisioning has been studied by media studies scholar Taina Bucher (2017b) as an ‘algorithmic imaginary’ (“ways of thinking about what algorithms are, what they should be, and how they function”), and by digital humanities scholar Sophie Bishop as ‘algorithmic gossip’ (2019) (“communally and socially informed theories and strategies pertaining to recommender algorithms, shared and implemented to engender financial consistency and visibility on algorithmically structured social media platforms”) and ‘algorithmic lore’ (2020) (“how the subjective decision-making practices of human intermediaries continues to play a significant role in even ostensibly algorithmic symbolic production”).
Ownership practices are users’ engagements in assertive behaviour to stake their authorship and attribution claims, or desires for acknowledgement and credit, when others borrow, reuse, adapt from, or remix a piece of content that they originated.
One of the most common types of ownership claims is the use of
Figure 4
(L–R)

In
In
Another type of ownership claim is when the originators of audio memes, dialogue memes, or other TikTok trends call out other TikTokers for the
Figure 5
(L–R)

Following this, many TikTokers immediately responded by creating various videos to ‘take back’ the audio from those who have misused it. Two days later, @absofacto posted an update using the same audio meme, and thanks several groups of TikTok users––such as ‘alt tiktok’, ‘lgbtq tiktok’, ‘kpop stans’, and ‘everyone who doesn’t fit in any group’ (
At times, ownership claims on TikTok take on a more macro stance. In July 2020, one of the milestone viral trends on TikTok was the popularity of
Figure 6
(L–R)

However, within a week there was swift backlash over the alleged ‘colonization’ of Douyin by TikTok users, due to the influx of English-language comment spam on Douyin videos demanding subtitles for non-Chinese speakers, or complaining about how difficult it was for them to navigate the Douyin interface. Many TikTokers, especially those who were Chinese diaspora like @k.hyli (
Algorithmic practices are users’ engagements in patterned and routine behaviour in the belief that their repeated actions will persuade and trigger the platform’s algorithm to work in their favour, and is informed by a collective ‘algorithmic imaginary’ (Bucher 2017).
A popular algorithmic practice is the
Figure 7
(L–R)

Users are then invited to ‘stay while’ or ‘hang out’ if they like, which in ‘TikTok algorithmic speak’ translates to an invitation to rewatch and loop the video for a longer period of time, or to discover more of such similar content by engaging with the TikToker’s other posts or hashtags. The underlying intention is for the viewer’s engagements with the post to teach the TikTok algorithmic recommendation system to serve them more of such similar posts. However, these rabbit roles and silos may be actual subcultures on TikTok, like ‘Black Australian TikTok’ (
Figure 8
(L–R)

Another ‘
Another
Figure 9
(L–R)

In some instances, TikTokers may call upon others to engage with their posts for affirmation. This was the case with @carli.mochi calling for viewers to ‘press random buttons’ so that she will ‘wake up’ to some activity on her account (
Figure 10
(L–R)

Interactive practices are users’ engagements in parasocial behaviour to appeal to others, manifest displays of support or disavowal towards specific users and issues, foster and maintain allegiances within trends and silos, and maintain feelings of connection and enjoyment among each other.
In August 2020, @bellapoarch rapidly ascended into virality when her 18 August
Figure 11
(L–R)

Another interactive strategy is
Figure 12
(L–R)

As mentioned earlier in the paper, ‘call-out cultures’ are a mainstay on TikTok, with many users gaining visibility and virality for bringing TikToker’s attentions to specific social justice issues or social causes. This celebration of social justice pursuits on TikTok is exemplified in @hannahmchutchison’s video, which depicts ‘Tiktok influencers’ receiving pile-ons of compliments simply for stating obvious facts or rhetoric, like ‘racism is bad’ (
Figure 13
(L–R)

However, TikTokers have observed that some types of posts tend to attract more attention from other users, and have attempted to adopt these new formats to relay the same social justice message. For example, @sammyysambo opens her COVID-19 post with a text overlay that reads ‘Dances get further reach but please pay attention to what I’m going to say’ (
Legacy practices are users’ engagements in established microcelebrity and Influencer behaviours that originated and flourished outside of TikTok, by extrapolating, contextualizing, and updating these visibility practices and Influencer logics, in order that they may maintain some brand coherence and sustain some attention outflow to their existing digital media estates.
There are dozens of subcultures that have been imported onto TikTok and have attempted to adapt their content norms through
Figure 14
(L–R)

Mainstream media outlets have also taken to TikTok to promote their stories and attract outflow traffic to their sites. To do so on TikTok requires the mainstream media outlets to play with various elements of social media pop cultures in order to appeal to the young and trendy prospective audiences on the app. For instance, @washingtonpost has socialized and popularized newsworthy issues by posting a TikTok featuring a man acting out a POV of ‘TikTok’s lawyers walking [in to] file a lawsuit against President Trump’ in light of the US’s potential TikTok ban in August–September 2020 (
Influencers who have been established on other platforms may also engage in
Figure 15
(L–R)

Established singer-songwriter-actor @troyesivan first kicked off his music career on YouTube, and in his TikToks, occasionally talks about how his past experience as a YouTuber has been valuable for his music career. In one TikTok (
There are dozens more legacy practices established in prior attention economies and internet celebrity cultures that have since been carried over to TikTok. As future research pivots to these diverse cultural practices on TikTok, I would like to offer a few that are pertinent for further scholarship. Firstly,
Secondly, ‘
Thirdly, legacies of ‘
Lastly, there is an emergent culture of TikTok being used as a dating app, with some accounts aggregating and recommending TikTok, Instagram, or Tinder profiles of eligible men and women, participating in hashtag challenges to find a date (Iovine 2020), and following and interacting with selected accounts through comments and direct messages with the prospect of seeking dating partners (Crelin 2020). Future research may like to adapt from dating app scholars’ notion of ‘off-label uses’ (Albury & Byron 2016; Duguay 2020) to explore creative circumventions and applications of TikTok for various agendas.
The considerations of how internet celebrity is situated on TikTok, the key elements of TikTok’s attention economies, and the emergent visibility labour practices on the app have finally led us to understanding how TikTok-native Influencers are cultivated, which is the goal of subsequent papers based on this project. A continuation of similar attempts to map out internet celebrity and Influencers cultures on TikTok should ideally also consider the role of intermediary institutions like Influencer agencies, TikTok creator networks, digital media and advertising firms who have taken to TikTok, and early adopter brands and clients––the latter of which were among the first to monetize TikTokers and TikTok Influencers. While TikTok’s ‘For Business’ page does not list the full range of options until a brand signs up for an account (TikTok n.d. c), industry insiders have observed various advertising options including ‘In-Feed Video’, ‘Brand Takeover’, ‘Hashtag Challenge’, ‘Branded AR Content’, and ‘Custom Influencer Package’ (Hines 2020).
Most pertinently, the nature and landscape of internet celebrity and prospective Influencer cultures on TikTok will soon undergo a rapid restructuring. At the time of writing in August-September 2020, TikTok has activated its USD$1 billion and GPB$54 million TikTok Creator Fund, which is an in-house partnership programme for users to monetize their content (TikTok 2020). First announced in July 2020 (Pappas 2020a), signing up for the TikTok Creator Fund allows TikTokers to accumulate and cash out earnings based on the engagement of their posts. The first batch of US-based recipients was announced in August (Pappas 2020b), with subsequent recipients based in parts of Europe including the UK, Germany, Italy, France, and Spain (TikTok 2020). At the time of writing, TikTokers must be ‘18 years or older’, have a ‘baseline of 10,000 followers’, have ‘over 10,000 video videos in the past 30 days’, and ‘post original content in line with [TikTok’s] Community Guidelines’ to be eligible (Pappas 2020a; Pappas 2020b).
In essence, the TikTok Creator Fund corporatizes the process of grooming TikTok Influencers, filtering out who can qualify to be one and be promoted as such on the app. That is, TikTok Influencers may not always necessarily emerge ‘organically’ any more, but have their visibility and internet celebrity designed and manipulated by TikTok’s concerted manipulation. This move is likely to cause the middle-men and third-parties––like Influencer agencies and TikTok creator networks (Tan 2020)––to lose some leverage and importance, despite them being the first to experiment with grooming, managing, and monetizing TikTok Influencers thus far. More pertinently, it is also speculated that TikTok may eventually decide to allow only ‘registered’ and ‘authorized’ users to profit off their posts, as ByteDance has had a record of banning and deleting accounts of previously ‘organically viral’ TikTokers for delving into contentious topics (ABC News 2019; Ma 2019).
The need to abide by TikTok’s Community Guidelines and the grey thresholds of acceptable topics may also result in the de-platformizing of prominent TikTokers who have attained their internet celebrity from speaking out against issues of social (in)justice, given that TikTok has had a history of moderating and censoring pro-LGBT content in countries where ‘homosexuality has never been illegal’ (Hern 2019), shadow-banning sexuality hashtags like ‘#gay’ and ‘#transgender’ (Ryan et al. 2020) and political hashtags like ‘#Putin is a thief’ in Russian and ‘#why do we need a king’ in Thai (Li 2020; Ryan et al. 2020). It is speculated that TikTok’s moderation policies may grow more hard-handed and traipsing the fine lines that police politics and morality, as its sister-app Douyin was earlier found to be banning users for speaking Cantonese instead of Mandarin in livestreams (@davidpaulk 2020; 羊城网 2020).
Already in the first month since its launch, media reports indicate that users who have signed up for the TikTok Creator Fund are disappointed with the lack of transparency over how payouts are calculated, citing that their high engagements into the ‘hundreds of thousands of views’ were only earning them ‘a few dollars’ (Matsakis 2020). To this, a spokesperson for TikTok responded that payouts were calculated through a combination of views, ‘video engagement’, and ‘the region where the video was seen’, of which the latter two are arbitrary and do not provide registered TikTokers with much more information on the performance in the programme (Matsakis 2020). On TikTok, dozens of videos sighted in this project’s study have been circulating, with users claiming to have been ‘shadow banned’, ‘de-ranked’, ‘blocked from the FYP’, or ‘hidden’ after having registered for the TikTok Creator Fund. Disgruntle users speculate in comments sections that their viewership has been ‘halved’, that TikTok was ‘avoiding payouts’, that the partnership was ‘too good to be true’, and that they have ‘regretted’ their decisions.
On the other hand, dozens of other users sighted in this project’s study seem to hold good faith in TikTok’s promises, and have been pleading with fellow users for ‘interacts’ in order that they may ‘cash out’ with TikTok to support urgent financial needs, ranging from medical support to escaping instances of domestic violence. While the rigor and transparency of the TikTok Creator Fund is still in flux, TikTokers have already been altering their posting behaviours, pushing more aggressively in their ownership practices, experimenting more with their algorithmic practices, employing more parasocial strategies to win over engagements in their interactive practices, and borrowing from the established strategies of legacy practices to find their footing. As scholars delve into the next stage of TikTok’s maturity and development––including recent controversies like the optics and politics of the TikTok ban in the US and elsewhere (Savic & Abidin 2020), the change in TikTok’s ownership in the US (O’Brien & Arbel 2020), and issues of content moderation given the circulation of suicide content among very young users (Matamoros-Fernández & Kaye 2020)––it is hoped that this piece can provide at least one early account of how vernacular internet celebrity cultures had developed on TikTok.
Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8
![(L–R) 8.1: ‘the non-killable kind, don’t…’ (@supah_jp 2020); 8.2: ‘COntestants: Do not take…’ (@theaugust27thshow 2020); 8.3: ‘[untitled]…’ (@the.hannahjensen 2020); 8.4: ‘[untitled]…’ (@coopij 2020); All screenshots by the author.](https://sciendo-parsed-data-feed.s3.eu-central-1.amazonaws.com/60a479462dbd400ff9693b4a/j_csci.140_fig_008.jpg?X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Date=20230610T202349Z&X-Amz-SignedHeaders=host&X-Amz-Expires=17999&X-Amz-Credential=AKIA6AP2G7AKP25APDM2%2F20230610%2Feu-central-1%2Fs3%2Faws4_request&X-Amz-Signature=b3dd729ae0947da8baafae7fb85c7e8b552aea250cd7e783598cfc8580499564)
Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Figure 15

Eisenstein’s ‘Magic of Art’ Sharing in Action: The Systemic Concept of the Environment in Aleksandr Bogdanov Eisenstein’s ‘Cinema of the Masses’ Somatic Montage for Immersive Cinema Aleksandr Bogdanov and Sergei Eisenstein on Emotions: The Affectional, the Theory of Expressiveness, and the Emotional Script Seiwert’s ‘Open Letter’ to Bogdanov Aleksandr Bogdanov’s Podbor and Proletkult: An Adaptive Systems PerspectiveCultural Science Meets Cultural Data Analytics Aleksandr Bogdanov’s Tektology: A Proletarian Science of Construction Aleksandr Bogdanov, ‘Science and the Working Class’ Aleksandr Bogdanov and Lenin on “Things-In-Themselves” Interview as Archive: Moving in Disciplinary Space from Cultural Studies to Cultural Science. An Interview with John Hartley AM Towards a Tektology of Tektology Biosemiotic Foundations of a Darwinian Approach to Cultural Evolution Tangential Points: Aleksandr Bogdanov and Sergei Eisenstein Revisited The Culture as System, the System of Culture: Aleksandr Bogdanov on Proletarian Culture and Proletarian Art Aleksandr Bogdanov’s Concept of Culture: From Workers’ Circles to the Proletkult Movement Special Thematic Section “Eisenstein, Bogdanov, and the Organization of Culture”: Guest Editorial Introduction Aleksandr Bogdanov’s Sociology of the Arts Sergei Eisenstein’s System Thinking: Influences and Inspirations Knowledge as Film vs. Knowledge as Photo: Alternative Models in Early Soviet Thought Sergei Eisenstein in the Proletkult