Copyright callouts and the promise of creator-driven platform governance (2024)

This paper is part of Locating and theorising platform power, a special issue of Internet Policy Review guest-edited by David Nieborg, Thomas Poell, Robyn Caplan and José van Dijck.

Introduction

Social media platforms are more popular than ever due, in part, to a growing subset of “commercializing and professionalizing” users known as creators (Cunningham & Craig, 2021, p. 1). Through ad-revenue sharing agreements (Caplan & Gillespie, 2020), brand deals (Hund, 2023), and a combination of merchandise, fan funding, and affiliate programs (Rieder et al., 2023), creators monetise their audiences and contribute to the “creator economy,” currently valued at 250 billion USD (Perelli, 2023). As a career path, content creation offers significant flexibility and autonomy (Cunningham & Craig, 2021), yet comes with precarious working conditions (Duffy et al., 2021). Creators, like all social media users, are subject to opaque and often automated forms of platform governance where content can be removed and accounts terminated with little notice or explanation (Gorwa et al., 2020). These conditions generate a “crisis of legitimacy” (Zuckerman & Rajendra-Nicolucci, 2023), with polls finding, for example, that three-quarters of Americans do not trust social media companies to moderate content fairly (Kemp & Ekins, 2021). Issues with content moderation also generate social, emotional, and economic consequences for creators (Are & Briggs, 2023; Kingsley et al., 2022), chill speech (Myers West, 2018), deter the use of platform reporting tools (Vaccaro et al., 2020), and promote conspiracy theorising about platform operations (Riedl et al., 2023).

In the face of such frustrations, creators have little recourse given the implicitly feudal design of platforms where users are “subject to a power structure that is apparently absolute and unalterable by those who lack such power” (Schneider, 2022, p. 1966). As a consequence, creators employ an assortment of strategies to navigate the fickle creator economy, including optimising content production (Hallinan, 2023), sharing theories about how platforms operate (Kaye & Gray, 2021), and joining intermediary commercial organisations (Siciliano, 2020). However, each strategy concerns opportunities for success within a given platform context. Creators who seek to transform platform operations have two primary strategies for organisational change: exit and voice (Hirschman, 1970; see Frey & Schneider, 2023 for an elaboration of these options in a platform context). In other words, creators can either leave the platform or communicate their displeasure. Neither option is necessarily appealing: the former is disincentivised by the lack of market competition and the latter beset with questions of who can or should address their complaints.

Platform callouts, or public complaints about the conduct of or on platforms, are one of the primary ways that creators voice their frustrations with governance issues. Compared to other expressions of consumer discontent, creators benefit from in-built audiences and their professional specialisation in public communication. Such conditions also differentiate creator complaints from other callouts on social media associated with the collective behaviour of semi-anonymous masses (Kim et al., 2022). While creators have spoken up about censorship, economic opportunity, copyright abuse, algorithmic bias, and predatory behaviour (Kaye & Gray, 2021; Kumar, 2019; Reynolds & Hallinan, 2024; Tarvin & Stanfill, 2022), questions remain over the ability of voice-based strategies to overcome the inherent imbalances of platform power (Nielsen & Ganter, 2022). Platforms are selective about which controversies they respond to (Shapiro et al., 2024) and public responses can leave structural issues unchecked, resulting in what Tarvin and Stanfill call “governance washing” (2022). Furthermore, the reliance on public outrage, from creators or otherwise, may trap platforms in reactionary cycles with unsatisfying solutions (Annany & Gillespie, 2016). Despite these structural constraints, the persistence of public complaints motivates our investigation into the role platform callouts play in the broader governance ecosystem.

To do so, we turn to a particularly entrenched domain of platform governance where we would expect creators to have little influence: copyright policy. Platforms’ copyright policies are heavily shaped by regulations that favour the interests of major corporate rights holders (Dergacheva & Katzenbach, 2023; Gray, 2020) and copyright enforcement is an established point of frustration for creators (Fiesler et al., 2023; Kaye & Gray, 2021). Focusing on YouTube, an industry leader in copyright management (Gray, 2020) and creator monetisation programmes (Caplan & Gillespie, 2021), we analyse callout videos, defined as social media content where creators discuss their experiences with copyright enforcement, challenge claims of infringement, and critique the broader structures of content moderation. In what follows, we present the possibilities for user participation in platform governance and briefly describe YouTube’s copyright management system. We then outline our methodological approach and show how callout videos alternatively address other creators, the platform itself, and external actors like corporate rights holders, spammers, and scammers. Finally, we differentiate between horizontal and vertical callouts, distinguished by the relationship between institutional positioning of the speaker and target, and discuss the implications of both as tools of creator-driven platform governance.

Participation structures of platform governance

While content moderation includes advanced automation (Gorwa et al., 2020), most platforms also delegate work to their users via tools to report violations (Crawford & Gillespie, 2016). User reports of copyright infringement or community guidelines violations provide a valuable signal for platforms’ Trust and Safety teams (Pfefferkorn, 2022) and offer a mechanism of user participation in platform governance, although the design of reporting systems minimises public deliberation and collective action (Crawford & Gillespie, 2016). Even in these circ*mstances, users across platforms and the political spectrum have appropriated flagging tools (e.g. Fiore-Silfvast, 2012; Zhao & Chen, 2023), transforming reports into a site of “contested platform governance” where users challenge the “core values, identity, and/or purposes of the platform” (Sybert, 2022, p. 2312). In doing so, users threaten the integrity of user reports as a signal for corporate platform governance objectives (Zhao & Chen, 2023).

Some platforms further delegate responsibility, consulting with civil society organisations (Caplan, 2023) and providing tools for community-driven moderation (Zuckerman & Rajendra-Nicolucci, 2023). The latter setup is particularly evident on decentralised platforms (Struett et al., 2023) and forum platforms like Reddit, where individual interest groups, or subreddits, have specific rules and volunteer moderators who enforce them (Seering, 2020). Seemingly centralised platforms like Facebook provide users tools to moderate different “sub-platforms” like personal profiles, pages, and Facebook groups (Navon & Noy, 2023). Similarly, YouTube provides tools for creators to moderate the comments that appear alongside their videos, posts, and livestreams. Although a community-driven approach to participation is more robust than flagging tools, it remains limited by how platforms define community (e.g. individual accounts on YouTube vs. subreddits on Reddit), the tools they provide, and the supplementary status of community moderation, where users can add new rules but cannot challenge platform-wide policies.

For users frustrated with centralised platform governance, there are few opportunities for recourse. Like any consumer, users can exit the market (Hirshman, 1970) and switch to a new platform or reduce social media use. However, as an industry, social media is dominated by a few big players and lacks robust competition. Furthermore, platforms heavily draw their value from network effects, making platform migration socially and technically difficult (Fiesler & Dym, 2020). These factors are intensified for creators, who draw personal and economic value from the audiences they foster on particular platforms (Cunningham & Craig, 2021). While the rare creator has successfully transitioned their career from one platform to another (Shapiro et al., 2024), this remains an exceptional accomplishment. Finally, given the scale of major platforms, a substantial number of users must exit to send a clear signal to corporate leadership.

Given the constraints on leaving a platform, some users try to change platforms from within through voice-based strategies (Hirschman, 1970, p. 30). Users can express frustrations through official channels such as appealing a content moderation decision. When official channels fail or are unavailable, users may go public with their complaints (Meisner, 2023; Reynolds & Hallinan, 2024). However, platforms are selective in which voices they respond to (Shapiro et al., 2024), favouring users with important roles for platform functionality like moderators on Reddit or high-profile creators on YouTube. Even among these populations, there are concerns about the efficacy of voice as a response, reflected in Frey and Schneider’s (2023) distinction between effective and affective voice where the former refers to direct influence and the latter to indirect persuasion. Public complaints clearly fall in the category of affective voice, joining a broader conversation about callouts, cancellations, and new strategies of public accountability (Lewis & Christin, 2021).

As public complaints that use shaming to promote social norms, callouts heavily depend on social media platforms for circulation. Most research focuses on callouts that target individuals rather than engage in institutional or organisational critique (e.g. Kim et al., 2022; Lee & Abidin, 2024). In this context, researchers have expressed concern that callouts may facilitate harassment or struggle to promote behavioural change (Billingham & Parr, 2020; Kim et al., 2022; Marwick, 2021). Despite these concerns, the act of calling out can provide expressive benefits for speakers (Herbison & Podosky, 2024) and constitute a form of civic engagement (Kligler-Vilenchik, 2017). Furthermore, as a public conversation, the target is not the only relevant audience; indeed, callouts can promote communal norms without changing the attitude or behaviour of the target (Lewis & Christin, 2021). To investigate the role of callouts in platform governance, we turn to a fundamental source of friction between users and platforms: copyright policy.

Copyright on YouTube as a test case for platform power

Copyright has long been a source of contention on YouTube as an area where “balancing the interests of the platform’s various ‘markets’ (of content suppliers, audiences, advertisers, and media partners) is most complex and questionable in the fairness of its application” (Burgess & Green, 2018, p. 48). States legislate copyright agendas to protect intellectual property rights, foster innovation, and, occasionally, serve the public interest (Gray, 2020). Platforms enforce policies and develop technological solutions to maintain legal compliance, pre-empt potentially burdensome legislation, appease copyright holders, and foster an appealing environment for users (Gray, 2020; Suzor, 2019). Where copyright was “once the exclusive domain of corporate lawyers and policymakers”, its centrality on digital platforms has increased public awareness and interest (Gillespie, 2007, p. 5), reflected in the prevalence of “copyright gossip” on YouTube (Kaye & Gray, 2021) and discussions among transformative fandom communities (Fiesler et al., 2023). Yet, as Gray argues, when it comes to legislation, the public interest is poorly represented and large corporate stakeholders dominate (Gray, 2020, p. 132).

The external governance of copyright on YouTube includes a diverse and international patchwork of legislation. Central among these is the Digital Millennium Copyright Act from the United States, passed in 1998, which instituted the “notice and takedown” approach to copyright violations (Decherney, 2014). Even beyond the context of copyright, the notice and takedown approach “has become the go-to model for those attempting to solve any number of online disputes over intellectual property, online speech, and other issues”, including trademark and the right to be forgotten claims (Decherney, p. 19). Google has also played a major role in exporting U.S. copyright norms, including the “notice and takedown” approach and “fair use” principles, to other countries through policies, lobbying, and user education initiatives (Decherney, 2014). Yet the United States is not the only relevant regulatory actor, as the 2019 passage of Article 17 of the European Union Copyright Directive demonstrates. This legislation addressed long-standing complaints of music rights holders in Europe that digital platforms, especially YouTube, used safe harbour principles to undervalue copyrighted works (Bridy, 2019). While the implications of the Copyright Directive continue to unfold, research suggests that its enactment has led to increased copyright takedowns on YouTube (Dergacheva & Katzenbach, 2023).

The internal governance of copyright by YouTube revolves around the platform’s Content Management Suite which includes three primary tools: the Webform, the Copyright Match Tool, and the Content ID system. The Webform is an online reporting tool that allows anyone to manually claim infringing content. The Copyright Match tool automatically identifies videos reuploaded by other channels and is available to members of the YouTube Partner Program or creators with a history of content takedowns. Finally, Content ID is a “scaled tool” that fully automates the copyright claim process and is available to “those with the most complex rights management needs, such as movie studios, record labels, and collecting societies” (Google, 2022, p. 3). In the first half of 2022 alone, ContentID processed more than 750 million claims, accounting for 98% of all copyright claims on the platform (Google, 2022, p. 4). Automated copyright enforcement is both an expression of Google’s platform power and a reflection of the economic power of media conglomerates. Indeed, part of the market appeal of Content ID is that it enables rights holders to earn significant revenue, upwards of US$30 billion over the past three years (Google, 2022, p. 1).

YouTube’s Content Management Suite offers unequal resources to corporate stakeholders and independent content creators, leading some creators to conclude that copyright governance on the platform does not serve their interests (Fiesler et al., 2023; Hui, 2021). As ethnographer Michael Siciliano explains, many creators he spoke with “felt powerless, describing a silence and inscrutability similar to other users of Google’s infrastructures and recounting lengthy appeal processes that often failed” (2020, p. 149). One way that creators aim to make their work less precarious is through the circulation of “gossip” in videos where they share “their experiences with copyright enforcement on YouTube” and strategies for avoiding or addressing copyright enforcement (Kaye & Gray, 2021, p. 1). Creators have also critically engaged with copyright policy in the case of the #WTFU hashtag campaign1 started by The Nostalgia Critic in 2016 over frustrations with Content ID, which led the YouTube policy team to issue a statement acknowledging community concerns (Edwards, 2018). Together, this work provides evidence that users play a role in the governance of copyright enforcement on the platform, fitting within a broader pattern of “communal solidarity” emerging among creators (Kumar, 2019), although the particular norms, community boundaries, and mechanisms of governance remain opaque.

Methods

To collect YouTube copyright callout videos, we conducted targeted keyword searches using YouTube Data Tools (Rieder, 2015).2 We restricted the results to videos published after 2019 to update existing research,3 and removed videos that were off-topic, uploaded by non-creators (i.e. from an official YouTube channel), and not primarily in English. We also screened out exclusively descriptive videos for lacking the element of “complaint” central to our definition of callouts. We then manually added 20 videos that appeared in platform recommendations or news coverage of copyright issues, resulting in a dataset of 230 videos. Two authors watched 10 videos from the dataset to develop the codebook, adapting categories from previous research to fit with the data (Kaye & Gray, 2021; Reynolds & Hallinan, 2024). Our codebook addresses how callout videos define the problem of copyright enforcement, attribute responsibility, and propose solutions (see Appendix 1 for the full codebook). Given prior research showing strong gender segregation in YouTube communities (Wegener et al., 2020) and greater representation of men in platform callout videos (Reynolds & Hallinan, 2024), we also coded the perceived gender of creators, prioritising self-identification when available, using conventional visual and verbal markers otherwise,4 and treating VTubers and furries as a separate category.5

Two authors began independently coding the videos, one working from the top and the list and the other from the bottom, meeting to discuss findings throughout. We stopped coding at 135 videos when we noticed significant repetition in the results (see Appendix 2 for video details). Most videos received more than 50,000 views, although there was nearly an even split (see Table 1). The overwhelming majority of speakers were men (n=101), along with a small minority of women (n=13) and VTubers/furries (n=7), aligning with prior work on YouTube callouts (Hallinan & Reynolds, 2024). Channels represented various genres including gaming, music reviews, film reviews, react videos, and cultural commentary.

Calling out copyright on YouTube

In our dataset, when YouTubers called out copyright enforcement, they were primarily concerned with so-called “false” claims that ignore fair use or serve ulterior motives like harassment, censorship, and financial extortion (see Table 2). Creators criticised the abuse of manual copyright reporting tools at nearly twice the rate of automated claims, even though Content ID accounts for 98% of all copyright claims on the platform (Google, 2022). The pronounced concern with the abuse of copyright reporting tools echoes findings from interview and survey research with creators (Fiesler et al., 2023; Kingsley et al., 2022), as well as previous research on copyright gossip (Kaye & Gray, 2021), lending credence to the claim that creators understand the problems of copyright enforcement differently than the platform or large corporate rightsholders (Gray & Suzor, 2022). Whether automated or manual, callout videos framed false copyright claims as a threat to financial and personal well-being given the risk of account termination from repeated copyright violations. Creators also expressed frustration about the lengthy process of appealing copyright claims, its associated risks to privacy,6 and the platform’s opaque communications with users, in line with previous research (Fiesler et al., 2023; Reynolds & Hallinan, 2024).

Following their definition of the problem, the videos we analysed primarily discussed solutions for navigating false copyright claims. Almost no creator proposed leaving YouTube.7 Instead, creators described an assortment of “little hacks” for navigating content moderation (Gillet et al., 2023), including self-censorship to avoid being copyright-claimed, editing or removing videos that have already been copyright-claimed, and strategies for subverting automated copyright detection on the platform. A majority of videos also discussed voice-related strategies for expressing dissatisfaction (Hirschman, 1970). Creators described their often ineffective experiences appealing a decision or contacting creator support. In response to the perceived limitations of YouTube’s official communication channels, they also discussed alternative strategies like going public and enlisting audiences and fellow creators to “signal boost” the message by sharing it on social media. Even when voice was not explicitly discussed, creators implicitly endorsed the strategy by publishing a callout video to the platform. Yet we found the role of voice depended on how creators assigned responsibility for the problem of false copyright claims. In what follows, we present three main targets of accountability: other creators, the platform, and external actors.

Calling out creators

In a video calling out another creator for employing a rights management company to claim copyright (and thus ad revenue) on videos that feature clips of their content, political streamer Hasan Piker appealed to shared community norms:

There are certain rules that every content creator knows not to f*cking break… there's a reason why so many people lose their minds, so many random YouTubers will get incredibly f*cking mad about this because it would ruin the entire space.8

The rule that Piker invoked is a prohibition against using copyright management tools to claim clips of content featured in reaction or commentary videos.9 While platforms like YouTube regulate copyright enforcement through policy documents and design, unspoken rules also determine community membership. Such norms are reinforced through callout videos like Piker’s, which does not exist in isolation. Other videos in our dataset described creators who “falsely” claimed copyright as coming “under the commentary crosshairs”10 for committing “one of the biggest crimes a YouTuber can do”11 and undermining “a very homey system, a very handshake system” that creators depend on to negotiate copyright on the platform.12 Entire genres of content like reaction videos and established practices like collaborating depend on this “handshake system” where creators permit forms of copying and sharing without remuneration, just as the commercial viability of YouTube depends on copyright enforcement tools that appease major corporate intellectual property rights holders (Gray, 2020).

Creators in our dataset shamed others for claiming videos to primarily extract revenue, taking down videos to suppress speech, or filing multiple strikes to punish an antagonist on the platform — especially if the perpetrator was a successful channel targeting smaller accounts. While some of the videos in our dataset acknowledged legitimate reasons for disagreement and grievance among creators, they maintained that creators should not use the copyright enforcement system as a tool to harm an opponent or settle a dispute. Responding to a situation where another creator used copyright claims to silence critique, The Act Man explained

YouTube is a platform where we should all be able to hate each other and co-exist… Hopefully, the YouTube community can come together and demand YouTube implement better systems so that content creators who abuse this system are much more heavily reprimanded and discouraged from abusing it.13

Creators like The Act Man framed copyright reporting tools as more powerful than other content moderation systems for good reason.14 Reporting a community guidelines violation sends a signal to the platform that it may or may not act upon while filing a copyright claim produces an immediate effect: the user receives a notice about the claim and must either accept or dispute it. The copyright appeals process is also structured differently, requiring the targeted account to provide their name and mailing address to the claimant and ultimately leaving the assessment of the appeal in the claimant's hands. As movie channel Heavy Spoilers put it, when it comes to copyright enforcement, creators are “guilty until proven innocent”.15 Given the public criticism and comparative power of copyright reporting tools, some creators expressed significant trepidation or unwillingness to copyright strike other channels,16 even in cases of blatant abuse like ripping and re-uploading unedited videos.

Creators occasionally shamed those who used their (purported) experience with copyright enforcement to exploit audiences. For example, our dataset included two videos accusing Lady Decade, a gaming creator, of lying about being “extorted” over a copyright strike and “exposing” her behaviour as a grift.17 Finally, although minimally represented in our dataset, some videos shamed other creators for taking copyrighted material, upholding the importance of copyright, identifying valid reasons to use copyright management tools, and giving guidance about how to “fairly” use copyrighted material.18 By drawing boundaries around legitimate uses, creators try to protect popular genres and practices that fall in a legal grey area from additional enforcement.

Calling out the platform

In a video describing how Onision, an infamous creator accused of an array of personal misconduct, has repeatedly abused the copyright reporting tools to take down any videos that talk about him, YouTuber Repzion directly appealed to the platform for redress:

Hi Google, people at Team YouTube. This is a plea. I know some of you guys are watching this because this has to go through the approval process and I'm gonna tweet you guys this video but this is a real plea to anyone who works at Team YouTube or is involved with YouTube as a whole. This is a video simply begging you guys to take this seriously.19

Videos that directly addressed the platform, including its employees and corporate leadership, typically sought to rectify an enforcement problem, punishing an account for as-of-yet unrecognised misconduct or overturning punishments wrongly issued. Most wrongful decision claims were concerned with the inability of the Content ID tool to recognise fair use.20 These arguments reflect the uptake of fair use standards from the United States, even among creators from other countries with other policies for using copyrighted material — a finding that aligns with an interview-based investigation of creatives from Australia (Pappalardo et al., 2017). Occasionally creators critiqued structural aspects of copyright enforcement, focusing on specific policies or the appeals process. For example, Australian arm wrestler and sports podcaster Ryan Blue Bowen called on YouTube to “stand up and listen” after his experience of having his videos claimed by Argentinian hackers and to change the system so that other creators would not be subjected to this kind of abuse in the future.21

Callouts that targeted the conduct of YouTube often invoked the importance of a cross-platform messaging strategy in response to the challenge of getting a corporation to listen. Reflecting on the limitations of official channels for appealing a claim, music educator Rick Beato concluded, “the only recourse is to go to Twitter”.22 Other creators emphasised the importance of collective action and asked their audiences to signal boost the message by tagging official YouTube accounts, retweeting and replying to messages from the creator, and using a designated hashtag. The importance of Twitter, or X as it is now known, is echoed in the findings of other platform callout research (Berge, 2023; Reynolds & Hallinan, 2024) and interview studies with creators (Kingsley et al., 2022). Although the platform has since been purchased by Elon Musk and undergone a corporate name change operation, these developments were not reflected in our dataset. It thus remains an open question whether and to what extent X provides a viable means of “making some noise” (Reynolds & Hallinan, 2024), even as YouTube’s corporate accounts like @TeamYouTube remain active on the platform.

Calling out external actors

Interventions directed towards external actors were less frequent, perhaps reflecting a perception that callout videos were less likely to influence these actors. As gaming creator Matt Lowne lamented,

To be honest, I'm not expecting much major progress. YouTube is infamously apathetic when it comes to this sort of thing and Sony has no real incentive to try and straighten things out, so this is probably just to the situation and there's not much I can do about it.23

Although large media companies are active on the platform, their activity is not subject to the same community ideals professed by creators (Burgess & Green, 2018; Lewis & Christin, 2021). Indeed, the antagonism between creators and media corporations has a long history on the platform (Burgess & Green, 2018). If major corporations like Sony have little incentive to acknowledge appeals from independent creators, other external actors like hackers and spammers, which operate outside community and legal norms, are even less invested, making persuasive appeals to modify their behaviour unlikely to succeed. Additionally, it was not always possible to distinguish between the work of bad actors and corporations when it came to creators’ complaints about copyright trolls.24 As a rhetorical move of delegitimation, so-called copyright trolls could refer to entities employed by corporations to protect their IP or independent operators serving their own interests.

Notably, while creators occasionally described copyright law as part of the problem, they almost never attributed responsibility to the law.25 Indeed, none of the videos in our dataset addressed conventional state policy actors, even as earlier copyright callout videos have served this purpose.26 Like other external actors, the law exists beyond the boundaries of YouTube. Its remoteness may contribute to the perception that the law is not a part of their daily lives, in contrast with the proximity of copyright enforcement on YouTube. A final explanation comes from the positive reception of fair use principles, which were frequently invoked, including several creators going through the four factors that courts in the US consider when making a fair use evaluation.27 The popularity of fair use discourse may be due to the concept's prominence in YouTube's "Copyright School", a set of educational videos that creators must watch after receiving a copyright strike (Fiesler et al., 2023). Regardless of its origins, the repeated invocation of fair use as a friend to creators that is overlooked by the platform and ignored by corporations fosters an affinity between creators and the law.

Vertical and horizontal callouts as tactics of platform governance

Our dataset of copyright callout videos indicates that frustration with the prevalence of “false” copyright claims is a primary motivator for creators to speak out. As a category, the strategic ambiguity of false claims makes the accusation persuasively appealing: it is broad enough to encompass a wide variety of behaviour, it is prohibited by both YouTube’s policies and copyright law (Google, 2022; Mazzone, 2011), and the label’s negative associations make few likely to defend it wholesale. Despite general agreement over the problem of false copyright claims, creators offered diverse accounts of what distinguishes fraudulent uses of copyright management tools. While a few videos expressed extreme views such as the position that claiming copyright “is the worst thing you can do”28 or that copyright enforcement is a “no man’s land” where valid strategies are only determined by what someone can get away with,29 most professed relatively moderate positions somewhere between total freedom and total control. For example, H3 Podcast host Ethan Klein described the competing interests of different stakeholders shaping copyright policy to his audience:

You guys got to understand there are laws that YouTube has to abide by. There are legal precedents that if they don't abide by — the DMCA — then the system which we enjoy is just simply not possible. YouTube cannot exist. Because they have this treaty, it's like a diplomatic agreement in a war. 30

In approaching copyright management as a balancing act, most creators adopted the general disposition of copyright law (Burgess & Green, 2018; Gray, 2020), even if they disagreed on what constitutes balance in practice.

Despite their shared concern with “false” copyright claims, we found a distinction in how the videos leveraged shame that broadly aligns with who they deem responsible for the problem. Creators who called out other creators addressed their peers, while creators who called out the platform or other corporations addressed entities with greater institutional status. While all callouts present a violation of social norms and appeal to (presumably) shared values (Herbison & Podosky, 2024), we argue that callouts perform different functions for platform governance depending on the institutional relationship between the speaker and the target of the callout. The distinction between horizontal and vertical communication in organisational scholarship helps clarify our argument. Horizontal communication “refers to that between colleagues on an equal hierarchical level” while vertical communication “is that which travels up and down the hierarchy” (Bartels et al., 2010, p. 212). Building on this distinction, we differentiate between horizontal and vertical callouts.

Horizontal callouts refer to public criticism where the speaker and the target share the same institutional status, which, in the context of our study, involved social media creators addressing other creators. Horizontal callouts primarily appealed to community as a shared value. Although it would be a mistake to speak of a singular culture on YouTube given its scale, geographic reach, and linguistic diversity, as well as the platform’s commercialisation (Burgess & Green, 2018), creators and audiences alike have long adopted the language of community, especially when engaging in normative debates about platform values (Burgess et al., 2016; Lewis & Christin, 2021). Some callout videos specified particular groups on the platform such as the “gacha reaction community”,31 while others appealed to a broader platform identity through the use of terms like “creators” or “YouTubers”. While community is an expansive category in copyright callout videos, it is not limitless. We did not, for example, find any callouts trying to enrol spammers or scammers into the shared norms of a community — instead, creators typically appealed to the platform to intervene or encouraged other creators to protect themselves. In drawing boundaries around communal identity and acceptable behaviour, horizontal callouts perform a similar function to metadiscussion on forums, with both offering a “mechanism through which groups can interrogate the boundaries of what is acceptable, can construct norms… and can enforce a certain degree of compliance to those norms” (Burnett & Bonnici, 2003, p. 342).

Vertical callouts refer to public criticism where the speaker and the target have different institutional statuses: in the context of our study, this typically involved creators addressing social media platforms. Vertical callouts primarily appealed to fairness as a shared value, highlighting both individual content moderation decisions and the structural conditions of copyright enforcement on YouTube as unfair. However, the entrenched power disparity between creators and corporations requires a collective response to increase the chance of being heard (Reynolds & Hallinan, 2024). Thus, calling out platforms and external actors is largely about marshalling a community, composed of audiences and fellow creators, to draw attention to an area of concern and appeal to broadly endorsed values. Where horizontal callouts direct community practices, vertical callouts seek corporate intercession to address problems that cannot be resolved through community agreement, including specific content moderation actions like removing a copyright strike, as well as more structural factors involving the process of adjudicating copyright claims. Although the vertical callouts in our dataset primarily addressed YouTube, a few creators requested media conglomerates like Nintendo and Universal Music Group to treat users more fairly and acknowledge reviewing a video game or teaching audiences how to play music as fair use. The infrequency of addressing external corporations may reflect a lack of confidence that callout videos are an effective mechanism for change. While creators also expressed doubt that “YouTube” was listening, they invoked precedents from their own experience or the experience of other creators where going public with a problem successfully led to its resolution.

Conclusion

Responding to the call to analyse how “power is distributed amongst various stakeholders whose platforms exert distinctive yet interdependent functions in the ecosystem” (van Dijck et al., 2019, p. 12), our account foregrounds the existing and potential role of creators in platform governance without romanticising the agency of individuals or overlooking structural factors. While platforms configure highly asymmetrical relationships among different stakeholders (Nielsen & Ganter, 2022; Shapiro, 2024), scholarly focus on the exercise of power by a platform risks treating the power of platforms as an overly deterministic, top-down affair, something done to and through users. In so doing, these theorisations struggle to account for a distinctive feature of social media as a venue where messages circulate and publics form, and thus downplay the role of users as “vital agents of platform politics” (Reynolds & Hallinan, 2021, p. 3268). Although YouTube lacks any “formalised process of stakeholder participation” (Kumar, 2019, p. 15), contributing to the precarity of creators and giving the company significant discretion in the concerns it chooses to address, creators engage in platform governance through novel strategies of “voice” (Hirschman, 1970). We identified two strategies of participation: horizontal callouts, which direct public criticism towards peers, and vertical callouts, which direct public criticism towards institutional superiors including the leadership of commercial platforms.

Horizontal and vertical callouts serve different roles in the platform governance ecosystem: the former offers a mechanism for community self-policing (Seering, 2020) while the latter provides a mechanism for influencing centralised content moderation policies and practices (Shapiro et al., 2024). Assessing the efficacy of callouts, or the “conditions under which voice-based mechanisms” can “ensure accountability” (Schneider, 2022, p. 1980), would thus benefit from understanding the different functions that callouts serve. For example, the effectiveness of horizontal callouts could be reflected in shared ideas of acceptable conduct, along with behavioural or attitudinal changes from the target of the callout. While the creators in our dataset roughly agreed that false copyright claims are a problem, to what extent other creators or audiences share particular ideas of platform community or “fair” use remains an open question. Similarly, the effectiveness of vertical callouts could be reflected in whether, and in what ways, the platform responds to publicised complaints, only some of which will be reflected in public statements. Although our dataset contained stories of YouTube overturning particular decisions, the prevalence of this practice, as well as the inequalities involved in who is able to successfully leverage public appeal (Reynolds & Hallinan, 2024; Shapiro et al., 2024), merit further investigation. Finally, the assessment of both horizontal and vertical callouts would benefit from greater attention to the expressive benefits of calling out as a practice (Herbinson & Podosky, 2024; Kligler-Vilenchik, 2017). What factors lead creators to go public with their problems? How do they evaluate the experience of doing so? And is there variation between the motivations and experiences of horizontal versus vertical callouts?

Horizontal callouts, like any form of community governance, raise important considerations around the boundaries of the community (Marwick, 2021; Seering, 2022), especially on a platform the scale of YouTube. While we found significant commonalities in the articulation of community in our dataset, our study also highlights a few prominent limitations. First, we only analysed English-language videos on a profoundly multi-lingual, multi-national platform. Second, the speakers in our sample are overwhelmingly men, even though women actively participate in copyright discussions in other contexts (Fiesler et al., 2023) and engage in other genres of YouTube callouts (Lewis and Christin, 2022). More work is needed to understand the relatively low levels of participation from women in copyright disputes and its implications for community governance. Third, our dataset highlights the meaningful differences in participation among big and small creators. YouTube provides big creators with better tools for managing copyright and official channels for voicing frustrations (Caplan & Gillespie, 2021). While small creators may be able to leverage community-directed power within particular niches on the platform, their overall ability to participate is contingent on catching the attention of a larger creator or going viral.

Although vertical callouts can be successfully mobilised to affect the platform’s decision-making, they are a blunt and unwieldy tool that reactively responds to issues with existing policies and enforcement rather than proactively participating in the development of new policies. Vertical callouts may also be a strategy of diminishing returns. Social media platforms are no strangers to controversy, resulting in a situation where “individual controversies — small shocks that make platform governance look less legitimate — can be weathered by technology companies without real lasting change” as tech companies develop and adopt crisis communication strategies (Suzor, 2019, p. 121; see also Ananny & Gillespie, 2017). Furthermore, the platform ecosystem itself is changing. While YouTube seems relatively stable in social media terms, X, the other platform that YouTubers relied on to mobilise public attention, is in a more liminal state following its purchase by Elon Musk. Whether X will continue to help users attract the right type of attention to publicly pressure YouTube and other corporations remains an open question.

Together, practical considerations about the continued efficacy of vertical callouts combined with the exclusions inherent to any conception of community invoked through horizontal callouts, especially in large and diverse platform environments, offer reasons to resist any simple celebration of user agency. At the same time, our analysis of copyright callouts complicates accounts of platform power that focus on the technological and economic influence of major corporations. Theorisations of platform power should consider not only the power of platforms or the regulatory power over platforms but also the power afforded to creators by virtue of the audiences they garner and the communities they cultivate on and through digital media platforms. Here we invoke a different connotation of platform as “a place from which to speak and be heard” (Gillespie, 2010, p. 352). Yet creator-driven approaches to platform governance need not stop at callouts. Platforms like YouTube can move beyond the minimal tools for user involvement like flagging systems (Crawford & Gillespie, 2016), taking inspiration from the community moderation tools developed on other platforms (Seering, 202) to harness community expertise and bring more humans into the moderation process — a feature consistently requested by users (Vaccaro et al., 2020). For regulators, creators represent an opportunity to bring more public participation into legislation. As the early example of the Nostalgia Critic’s #WTFU campaign attests, creators can effectively mobilise their audiences and networks of other creators to participate in public hearings (Edwards, 2018). Such opportunities promise ways of expanding an understanding of the unspoken rules governing digital platforms, the roles of creators in shaping and enforcing them, and pathways towards a more balanced configuration of platform power.