My notes on Paolillo, J. C. (2018). The Flat Earth phenomenon on YouTube. First Monday, 23(12).

Even if the resurgent belief in a flat earth remains a marginal phenomenon, it is fascinating for what it reveals about YouTube. In this paper John C. Paolillo documents the emergence of this YouTube community and the issues which preoccupy them. This involved producing a database of flat earth videos:

To identify Flat Earth videos, channel and video metadata was collected in a manner similar to prior studies (Cheng, et al., 2008; Paolillo, 2008). Firefox together with the GreaseMonkey add-in was used to run a user script collecting video and channel IDs from the YouTube public developer API. The script communicated with a PHP/PostgreSQL backend to store the IDs. For each channel three standard “playlists” were retrieved: uploads (videos belonging to the channel), likes and favorites (videos marked as such by the channel owner) [9]. Standard recursive crawling was applied: liked and favorited videos were used to identify new channels, whose playlists were retrieved, etc. [10] Crawling was done in multiple passes from July 2015 to January 2017, each time feeding in additional channels discovered via searching and browsing YouTube.

Two features I find particularly interesting are their hostility towards celebrity entrepreneurs and scientists, as well as public or private institutions like NASA or SpaceX who conduct publicity campaigns and the features they share with wider conspiracy culture, such as the invocation of popular culture dystopias and the notion of ‘red pilling’. These express themselves in a fixation on the epistemic status of their own claims and those of their opponents:

Flat Earth videos have an overwhelming preoccupation with epistemic status: lies, truth, proof, debunking, hoaxes, fakes, revelations, evidence, shilling, etc. all figure heavily in Flat Earth videos. Such an emphasis on knowledge requires that they present a basis from which to cast doubt on a round Earth (the “Globe Model”). The challenge is significant. Flat Earth belief only awkwardly reconciles with modern technologies like rockets (33), communication satellites, the Global Positioning System, the ISS (24), and interplanetary probes.

A whole range of strategies are deployed in the face of these challenges: “citation of religious or secular historical texts, reproduction of video evidence, experimentation and observation, mathematical analysis, speculation, bald contradiction, and ad hominem argument”. These are used to undermine established scientific authorities with the “Flat Earth Model” offered as a viable solution to what is presented as a debunked “Globe Model” (though as Paolillo points out, ‘model’ here is used in a diffuse and non-scientific sense). The material published by agencies like NASA and SpaceX is seized upon in the interest of correcting their claimed distortions. But these are supplemented by counter-experiments, driven by a radical empiricism, in which “viewers are told to not trust anything beyond their direct experience”. The failure of amateur experiments intended to establish the curvature of the earth are taken as proof of the flat-earth phenomenon. These are supplemented by appeals to authorities like engineers, military officers and airline pilots, used strategically to undermine other members of these groups who support the “Globe Model”.

Would we have seen the resurgence of flat earth belief without youtube? Their videos use genres such as vlogs, screencasts, interviews and documentaries, suggesting a deep engagement with the affordances of the platform. These are often accompanied by effective clickbait, competition between video producers, established memes such as ‘red pilling’ and invocation of fictional dystopias which all suggest a community well adapted to the attention ecology of the platform. Paolillo identifies this competition between flat-earthers for attention early on in the paper but doesn’t really develop the point. I wonder if the attentional darwinism of YouTube is as much an explanation of this resurgence as the material itself, which Paolillo explains in terms of a social psychology of stigma as more people are tarred by assocation with flat earth and thus acquire a stake in defending it. It provides an environment in which certain themes are liable to thrive (an overturning of established authority, revelatory esotericism, a radical empiricism perceived to be liberating etc) if packaged together in a way which takes advantage of the affordances fo the platform. What really interests me is the entrepreneurship of the YouTubers within the flat earth community, as well as how techniques spread between them and competition drove innovation.

Outline and Rationale

The Youtube platform has, since its earliest inception, offered the opportunity for topics of interest to be ‘debated’.  Initially these debates were informal (i.e. not following any of the recognised structures of debates) and usually used the ‘video response’ function.  This functionality was removed from the site in 2014, and whilst response videos are still made they do not have the profile they once had (and now tend to appeal more to ‘pwnage’ than honest argumentation).

More recently debating has returned to the site through the use of Google Hangouts, which allow disputants to engage ‘live’ with one another in an online version of more traditional debating formats, c.f. The debate between Dr Kristi Winters and Carl Benjamin (Sargon of Akkad).  This debate had many of the features of a standard debate, including the presence of a moderator, time-limited Opening and Closing Statements, and opportunities for Rebuttals, again time-limited.

(It is worth noting that this online version of a debate had the advantage of allowing participants to cite their sources and have them appear in the description of the video, a feature not present in live offline debates).

There are advantages and disadvantages to both of these systems:

Video Responses:

  • Advantages
    • They are naturally time-limited without the need for moderation
    • Opportunity for good preparation
    • Sources offered by ‘opponent’ can be considered
    • New sources can be found to advance one’s argument
    • Input can be given by viewers which might be taken into consideration
    • Video response format is not limited to ‘talking head’ and might include graphics, quotations, clips from video being responded to etc.
  • Disadvantages
    • Potential difficulty in following chain of debate (particularly since video response feature was removed from Youtube).
    • Different subscriber numbers for participants might mean different viewing numbers for the ‘sides’ of the debate.
    • Response videos have something of a bad reputation.

Google Hangout/Livestream

  • Advantages
    • Liveness
    • All the debate is in one place
    • Familiarity (they are similar to offline live debates)
    • Rebuttals are spontaneous, therefore indicator of live knowledge (maybe)
    • Questions and comments from viewers can be collected from Chat (not necessarily an advantage).
  • Disadvantages
    • As with live debates, they prioritise rhetorical flair and quick-wittedness over capacity to marshal information to construct an argument or rebuttal.
    • Don’t allow for checking of sources.
    • Doesn’t (necessarily) allow for an ‘equilibrating’ of styles.
    • Is constrained to talking head format.
    • Potential technical problems with live streaming.
    • Can be over-long
    • Different experiences for ‘live’ viewers (who have access to the Chat feature) vs. those watching the recording later.

A disadvantage shared by both systems is that the Youtube environment is not always supportive of ordered discourse.  Comment sections particularly can become hostile or partisan, neither of which condition contributes positively to the advancement of intelligent life.


To initiate a system which combines the advantages of Video Responses with those of Hangouts/Livestreaming.  The following structure is proposed:

  1. Debates are between two participants.
  2. A specific proposition is selected in advance, with participants taken the position of ‘Proposer’ (arguing the Affirmative) and ‘Respondant’ (arguing the Negative).
  3. Each participant makes a 10 minute video containing their Opening Statement with regard to the proposition.
  4. These Statements are uploaded to Youtube and set to go live at the same time.
  5. After a fixed period of time, say 24 hours, each participant uploads a Rebuttal video to Youtube, again set to go live at the same time.
  6. A second Rebuttal video is uploaded and made live a fixed period of time after that, say a further 24 hours.
  7. Participants each upload a final video containing their Closing Statement after a final fixed period of time.
  8. (Optional) Participants take part in a joint hangout in which they discuss the debate.
  9. All videos are posted onto a website set up for the purpose.
  10. Comments on the website are moderated according to clear guidelines.

On pg 102 of Jonathan Taplin’s Move Fast and Break Things, he highlights email exchanges between YouTube’s founders, released in a court case, which suggest the invocation of ‘user generated content’ might be a matter of branding rather than a meaningful growth strategy for social media platforms:

In another email exchange from 2005, when full-length movies were being posted on YouTube, Steve Chen, a cofounder of the company, wrote to his colleagues Hurley and Jawed Karim, “Steal it!,” and Chad Hurley responded: “Hmm, steal the movies?” Steve Chen replied: “We have to keep in mind that we need to attract traffic. How much traffic will we get from personal videos? Remember, the only reason why our traffic surged was due to a video of this type…. viral videos will tend to be THOSE type of videos.”

Much critical literature has focused on how social media platforms ossify existing hierarchies and establish new ones. It is too easy to see this as an unexpected consequence of a new social infrastructure, as opposed to an outcome which was knowingly designed in from the start.

In his Uberworked and Underpaid, Trebor Scholz draws out an important parallel between the platform capitalism of YouTube and the near universally praised Wikipedia:

Unsurprisingly, YouTube hires countless consultants to better understand how to trigger the participation of the crowd. They wonder how they can get unpaid producers to create value. But equally, on the not-for-profit site, Wikipedia is asking how they can draw in more female editors, for instance.

Both involve an orientation to their users which sees them as objects of management, even if we might see the ends to which they are being managed in very different terms. This makes a lie of what Nick Couldry describes as the ‘myth of us’: the imaginary of platform capitalism which sees it as facilitating the free expression of natural sociability which older socio-technical systems had constrained

In the last few months, I’ve begun to seriously plan a much more sophisticated follow-up to Social Media for Academics, investigating the implications of social media for academic labour. A crucial aspect of this, which seems likely to become much more so with each passing year, concerns the toxicity of many of the online environments in which academics are participating. If academics increasingly find themselves expected to use social media as a means of demonstrating engagement or at least signalling engagement-willingness then the toxicity of these environments will become an increasingly central labour issue.

My fear is that we will have the worst of both worlds. Academics will be coerced outwards into these online environments under the sign of ‘impact’, while finding themselves blamed if anything they do online attracts disapprobation for their employer. It’s easy to imagine how the moralism we see lurking beneath the impact agenda (those who claim not to ‘get it’ should be ‘ashamed’ as I recently heard an extremely senior person say) could find similar expression in managerial expectation of social media use. On our present trajectory, the likely outcome will be an individualised one: take responsibility for your own engagement and take the blame if you bring about any perceived damage to the corporate brand. This problem is compounded because, as Tressie McMillan Cottom puts it “the risks and rewards of presenting oneself “to others over the Web using tools typically associated with celebrity promotion” (Barone 2009) are not the same for all academics in the neo-liberal “public” square of private media.” Far from counteracting exclusion in higher education, social media for academics is amplifying the risks for those already marginalised.

As an example of how this is developing, consider this dispiriting reflection on being an academic video blogger on YouTube which Philip Moriarty passed on to me:

One of the main reasons why I think the promise of YT as a place where intelligent life might flourish is failing is the well-documented level of trolling and hatred that permeates the site, and which threatens to silence any but the most obnoxious or innocuous voices. I stopped making regular videos a couple of years ago when the vitriol I was receiving for having the temerity to make unpopular content spilled over into my personal life. In addition to receiving the usual grammatically-challenged insults and thinly-veiled threats the university I was working at was also contacted several times by folk demanding my removal. Eventually these ‘downsides’ to being an academic on Youtube outweighed the benefits and I gave up making public videos entirely.

And it isn’t just me. Over the past three years I have known four other academics leave Youtube for reasons very similar to my own. These were folk who were similarly motivated to bridge the gap between ‘town and gown’, between universities (which are often seen as elitist) and the wider world represented on social media. These people wanted to contribute their knowledge and also to learn from the contributions of others. They wanted to find ways to speak and to listen in ways which were more inclusive, and which the diverse communities on Youtube seemed to be able to offer. These fine people, like myself, became disheartened by the inability of YT to foster anything but the lowest common denominator, the most clickbaity, the most provocative, the most crudely entertaining, and the failure of the platform to support those who wanted to raise the bar.

Some might say (and indeed have said) that this toxicity is just a natural part of the online ecology and we should grow a thicker skin, or not feed the trolls, or any of the other platitudes that are trotted out to excuse bad behaviour, but I don’t think that’s good enough. When the comment section under a video is two thirds insult or threat then the value of that comment section drops to zero. No one with anything to contribute wants to be part of it. When you have to wonder if your latest video will prompt some faceless anti-intellectual gonk to contact your employer then the chilling effect takes hold and you censor yourself, (God forbid you should talk positively about feminism, or BLM, or the representation of women in video games). The number of eyeballs on the site might increase but the I.Q. of the site goes down.

The architecture of these platforms militates against their sustained pedagogical use. It might be that, as Pausé and Russell put it, “Social media enables scholarship to be publicised more widely within the  academy,  and  in addition to that, it enables  scholarship to become part of broader  social conversations”. The problem is that the incentives of these platforms have over time proved to be generative of a dialogical toxicity which tends to be obscured by the high-minded rhetoric of public engagement. The promise that social media might “bridge the gap between ‘town and gown’” is proving to be rather misleading. A large part of my new project will be exploring the implications of this at the level of the institutional politics of the university, with a particular focus on what it means for academic labour.

The role of social media for academics discourse in obscuring these issues, mystifying the complex politics of social media in the university through breathless reiteration of the individual benefits to be accrued through engagement, means it will be a central object of critique for the project. But I want to avoid slipping into utopian/dystopian, pro/anti framings of social media for academics. I still believe in its scholarly importance and it’s capacity to inculcate solidarity and (in limited ways) flatten hierarchies. There’s a great example of the latter in this paper by Pausé and Russell which I’m otherwise pretty critical of:

Accessibility means individuals who are not academically trained are able to  learn  about  a  field  of  research  and  contribute  to  it,  bringing  their  own  ideas  and  experiences  to  the  table.†    And  accountability  has  enabled  greater  criticism  of  the  process  of  scholarship  and  research.    Through  connecting  on  social  media,  marginalised  people  have  been  able  to  gather  sufficient  force  to  challenge  the  conventions  of  research;  to  insist  on  an  intersectional  perspective.    The  lived  experience  of  a  Māori  woman  living  in  Aotearoa  New  Zealand  can  challenge  the  theorised understanding of an academic.‡ People have objected to being studied, and  have demanded the right to participate in framing the discussion.  For example, the  Health  at  Every  Size®  (HAES)  movement  has  largely  been  led  by  advocates  from  within  what  is  known as  the  Fatosphere  (Harding,  2007),  prompting  research  that  questions the basic assumptions made about the relationship between body size and  health by health scholars and those working in the health field. This both challenges  and enriches scholars’ research.  There is now a rich empirical literature on the efficacy  of HAES (Burgard, 2014).


YouTube Conference: Call for Papers

23/4 September 2016, Middlesex University, The Burroughs, Hendon, London.

Keynote speaker: Professor Jean Burgess

Please send an abstract of 350 words plus a short bio of 100 words for single papers or 500 words and individual bios for group panels by email attachment to<>.

Deadline for receipt of abstracts is 4 April 2016.

YouTube has just passed its tenth birthday and it is timely to review not only how it has changed in that time, but also its wider influence.  By focusing on YouTube as a platform we want to draw together research that is distributed across disciplines to help cross-fertilise knowledge about YouTube and its users, and to identity the research questions and methods that best capture its ever-expanding reach, impact and significance.  We plan to include a panel of industry insiders to offer insights into possible futures in the light of current developments alongside the academic papers which we now invite you to propose.

Keynote Address

Professor Burgess will consider how YouTube and the broader online video environment have changed in the past decade, and what its competing futures look like. She will also discuss how we might learn to recognise such patterns of change empirically, and the key methodological approaches to studying the co-evolution of proprietary digital media platforms and their cultures of use over time.

Possible questions to address (but not limited to these):

*   How has the institutionalisation of YouTube changed its nature?
*   Has YouTube accelerated processes of media convergence and transformation?
*   What is the changing relationship of television to YouTube?
*   How have production techniques and practices developed as the platform matured?
*   What communities of practice have been influential in the development of YouTube norms?
*   How has the development of new aesthetic forms been enabled by YouTube?
*   What innovations in performance and modes of address can be detected on YouTube?
*   To what degree do YouTube’s affordances operate as a social medium?
*   What new forms of celebrity and fandom have emerged on YouTube and why?
*   What wider social, cultural and political changes can be attributed to YouTube’s influence?
*   Why do we need to regulate the corporate power of YouTube’s owners Google?
*   Is YouTube a positive space for self realisation and expression of marginalised identities?
*   How do concerns over data harvesting and privacy apply to YouTube?
*   How have conflicts over rights affected the monetization of YouTube activities?
*   What potential does YouTube have as a repository of curated archives?
*   What are the genres that have thrived on YouTube and what wider significance does this have?  (e.g. education, journalism, advertising and marketing,  campaigning and propaganda,  entertainment, documentary,  drama, comedy and parody, how to …)
*   Is the development of specific apps for Music, Kids and Gaming a significant new trend?
*   What research methods are used to study YouTube? What are their strengths and weaknesses?

Conference organisation team:

*   Professor Jane Arthurs
*   Dr Alessandro Gandini
*   Dr Paul Kerr
*   Nicola Skinner