Friday 26 July 2024
Content suppression techniques against dissent in the Fifth Estate - examples of COVID-19 censorship on social media
Written for researchers and others interested in the many methods available to suppress dissidents' digital voices. These techniques support contemporary censorship online, posing a digital visibility risk for dissidents challenging orthodox narratives in science.
The Fourth Estate emerged in the eighteenth century as the printing press enabled the rise of an independent press that could help check the power of governments, business, and industry. In similar ways, the internet supports a more independent collectivity of networked individuals, who contribute to a Fifth Estate (Dutton, 2023). This concept acknowledges how a network power shift results from individuals who can search, create, network, collaborate, and leak information in strategic ways. Such affordances can enhance individuals' informational and communicative power vis-à-vis other actors and institutions. A network power shift enables greater democratic accountability, whilst empowering networked agents in their everyday life and work. Digital platforms do enable online content creators to generate and share news that digital publics amplify via networked affordances (such as 💌 likes, " quotes " and sharing via # hashtag communities).
In an ideal world, social media platforms would be considered to be a public accommodation, and the Fifth Estate's users would benefit from legal protection of their original content, including strong measures against unjustified suppression and censorship. The latter should recognise the asymmetric challenges that individual dissenters, whistleblowers and their allies must confront in contradicting hegemonic social forces that can silence their opponents' (digital) voices: As recently evidenced in the COVID-19 "pandemic", the Twitter Files and other investigations, suggest how a conspiracy of multinational pharmaceutical companies, unelected global "health" organisations, national governments, social media and traditional broadcast companies all conspired in silencing dissent versus costly COVID-19 interventions. Any senior scientific expert who questioned this narrative in the Fourth or Fifth Estate was forced to negotiate suppression and censorship for the wrong-think of sharing "dangerous" opinions.
Such sanctions reflect powerful authorities' interests in controlling (scientific) language, the window of permissable opinion, and the social discourses that the public might select from, or add. Under the pretext of public "safety", the censorship industrial complex strong arms the broadcast media and social media companies into restricting dissidents' voices as "misinformation" that is "unsafe". Facing no contest, the views of powerful officialdoms earn frequent repetition within a narrow window for narrative control. At the same legitimate reports of mRNA injuries are falsely redefined to be "malinformation", and censored.
Consequently, instead of a pluralist distribution of power in the Fifth Estate that can support vital expression, powerful authorities are enforcing internet policy interventions that increasingly surveil and censor users' digital voices. Infodemic scholars whose work endorses such suppression would seem to be ignorant of how problematic it is to define disinformation, in general. Particularly in contemporary science, where: knowledge monopolies and research cartels may be dominant; dissent should to be welcomed for great science, and a flawed scientific consensus may be very dangerous. Science-, health communication, and media studies scholars may also ignore the inability of medical experts to accurately define what disinformation is, particularly where decision makers are commercially conflicted.
Censorship, plus other forms of silencing, of legitimate COVID-19 dissent is especially dangerous in being heavily asymmetric: Health authorities already benefit from ongoing capital cascades whose exchange largely serve their interests. Such exchanges span financial, social, cultural, symbolic and even other (e.g. embodied) forms of capital (Bourdieu, 1986:2018). By contrast, individual critics can quickly be silenced by attacks on their limited capital, effectively preventing them from exercising rights to free speech and sustained critique. Silencing such dissent has important public health ramifications, particularly where the potential for suggesting, and exploring, better public health interventions is shut down.
Another concern is that the censorial actions of artificial intelligence designers and digital platform moderators are often opaque to a platforms' users. Original content creators may be unaware that they will be de-amplified for sharing unorthodox views, as algorithms penalise the visibility of content on 'banned' lists, and any accounts that amplify "wrongthink". This post seeks to flag the wide variety of techniques that can be used in the Fifth Estate to suppress and censor. Each censorship/suppression technique is ordered within categories- (i) content, then (ii) accounts, (iii) practices against supportive publics, and (iv) for public anger. This post focuses on (i) and lists a category's techniques in order with increasingly severe techniques:
#1 Covering up algorithmic manipulation
Social media users who are not aware about censorship are unlikely to be upset about it (Jansen & Martin, 2015). Social media platforms have not been transparent about how they manipulated their recommender algorithms to provide higher visibility for the official COVID-19 narrative, or in crowding out original contributions from dissenters on social media timelines, and in search results. Such boosting ensured that dissent was seldom seen, or perceived as fringe minority's concern. As Dr Robert Malone tweeted, the computational algorithm-based method now 'supports the objectives of a Large Pharma- captured and politicised global public health enterprise'. Social media algorithms have come to serve a medical propaganda purpose that crafts and guides the 'public perception of scientific truths'. While algorithmic manipulation underpins most of the techniques listed below, it is concealed from social media platform users.
#2 Fact choke versus counter-narratives
A fact choke involves burying unfavourable commentary amongst a myriad of content. This term was coined by Margaret Anna Alice to describe how "fact checking" was abused to suppress legitimate dissent.
An example she tweeted about was the BBC's Trusted New Initiative warning in 2019 about anti-vaxxers gaining traction across the internet, requiring algorithmic intervention to neutralise "anti-vaccine" content. In response, social media platforms were urged to flood users' screens with repetitive pro-(genetic)-vaccine messages normalising these experimental treatments. Simultaneously, messaging attacked alternate treatments that posed a threat to the vaccine agenda. Fact chokes also included 'warning screens' that were displayed before users could click on content flagged by "fact checkers" as "misinformation". With the "unvaccinated" demonised by the mainstream media to create division, susceptible audiences were nudged to become vaccine compliant to confirm their compassionate virtue. At the same time to retain belief in mRNA genetic vaccine "safety", personal accounts, aggregated reports (such as "died suddenly" on markcrispinmiller.substack.com) and statistical reports (see Cause Unknown) for genetic vaccine injuries became suppressed as "malinformation" despite their factual accuracy. Other "controversial content", such as medical professionals' criticism of dangerous COVID-19 treatment protocols (see What the Nurses Saw) or criticism of a social media platform's policies (such as application of lifetime bans and critiques of platform speech codes) have been algorithmically suppressed.
An example she tweeted about was the BBC's Trusted New Initiative warning in 2019 about anti-vaxxers gaining traction across the internet, requiring algorithmic intervention to neutralise "anti-vaccine" content. In response, social media platforms were urged to flood users' screens with repetitive pro-(genetic)-vaccine messages normalising these experimental treatments. Simultaneously, messaging attacked alternate treatments that posed a threat to the vaccine agenda. Fact chokes also included 'warning screens' that were displayed before users could click on content flagged by "fact checkers" as "misinformation". With the "unvaccinated" demonised by the mainstream media to create division, susceptible audiences were nudged to become vaccine compliant to confirm their compassionate virtue. At the same time to retain belief in mRNA genetic vaccine "safety", personal accounts, aggregated reports (such as "died suddenly" on markcrispinmiller.substack.com) and statistical reports (see Cause Unknown) for genetic vaccine injuries became suppressed as "malinformation" despite their factual accuracy. Other "controversial content", such as medical professionals' criticism of dangerous COVID-19 treatment protocols (see What the Nurses Saw) or criticism of a social media platform's policies (such as application of lifetime bans and critiques of platform speech codes) have been algorithmically suppressed.
#3 Title-jacking
For the rare dissenting content that can achieve high viewership, another challenge is that title-jackers will leverage this popularity for very different outputs under exactly the same (or very similar) production titles. This makes it less easy for new viewers to find the original work. For example, Liz Crokin's 'Out of the shadows’ documentary describes how Hollywood and the mainstream media manipulate audiences with propaganda. Since this documentary's release, several videos were published with the same title.
#4 Blacklisting trending dissent
Social media search engines typically allow their users to see what is currently the most popular content. In Twitter, dissenting hashtags and keywords that proved popular enough to feature amongst trending content, were quickly added to a 'trend blacklist' that hid unorthodox viewpoints. Tweets posted by accounts on this blacklist are prevented from trending regardless of how many likes or retweets they receive. On Twitter, Stanford Health Policy professor Jay Bhattacharya argues he was added to this blacklist for tweeting on a focused alternative to the indiscriminate COVID-19 lockdowns that many governments followed. In particular, The Great Barrington Declaration he wrote with Dr. Sunetra Gupta and Dr. Martin Kulldorff, which attracted over 940,000 supporting signatures. After its publication, all three authors experienced censorship on search engines (Google deboosted results for the declaration), social media platforms (Facebook temporarily removed the declaration's page, while Reddit removed links to its discussion) and on video (Youtube removed a roundtable discussion with Florida's Governor Ron DeSantis whose participants questioned the efficacy and appropriateness of requiring children to wear face masks).#5 Blacklisting content for interactions from dodgy accounts
Limited visibility filtering also occurs when comments are automatically commented on by pornbots, or feature engagement by other undesirable accounts. For example, posts mentioning the keywords/subjects such as 'vaccine, Pfizer' may receive automated forms of engagement, which then sees posts receiving such "controversial" engagement becoming added to a list ensuring these posts censorship (see 32 mins into Alex Kriel's talk on the 'The Role of Fake Bot Traffic on Twitter/X'.
#6 Making content unlikeable and unsharable
This newsletter from Dr Steven Kirsch's (29.05.2024) described how a Rasmussen Reports video on YouTube had its 'like' button removed. As Figure 1 shows, users could only select a 'dislike' option. This button was restored for www.youtube.com/watch?v=NS_CapegoBA.
Social media platforms may also prevent resharing such content, or prohibit links to external websites that are not supported by these platforms' backends, or have been flagged for featuring inappropriate content.
#7 Making content unsearchable within, and across, digital platforms
Social media companies applied search blacklists to prevent their users from finding blacklisted content. Content contravening COVID-19 "misinformation" policies was hidden from search users. For example, Twitter applied a COVID-19 misleading information policy that ended in November, 2022. In June 20023, META began to end its policy for curbing the spread of "misinformation" related to COVID-19 on Facebook and Instagram.
#8 Content takedowns
Social media companies could ask users to take down content that was in breach of COVID-19 "misinformation" policies, or automatically remove such content without its creators' consent. In 2021, META reported that it had removed more than 12 million pieces of content on COVID-19 and vaccines that global health experts had flagged as misinformation. YouTube has a medical misinformation policy that follows the World Health Organisation (WHO) and local health authorities guidance. In June 2021, YouTube removed a podcast in which the evidence of a reproductive hazard of mRNA shots was discussed between Dr Robert Malone and Steve Kirsch on Prof Bret Weinstein's DarkHorse channel. Teaching material that critiqued genetic vaccine efficacy data was automatically removed within seconds for going against its guidelines (see Shir Raz, Elisha, Martin, Ronnel, Guetzkow, 2022). The WHO reports that its guidance contributed to 850,000 videos related to harmful or misleading COVID-19 misinformation being removed from YouTube between February 2020 and January 2021.#9 Deception conceals the motives behind censorship, and who its real promoters may be
Social media platforms not only hide algorithmic suppression from users, but may also be misused to hide from users the full rationale for censorship, or who is ultimately behind it. Professor David Hughes prepared a glossary of deceptive terms and their true meanings (2024, pp 194-195) to highlight how the meaning of words is damaged by propaganda. A term resonating with technique #9 is “Critical” - pretending to speak truth to power whilst turning a blind eye to deep state power structures. The official narrative positioned COVID-19 as a (i) pandemic that had zoonotic (animal-to-human) origins, and alternate explanations were strongly suppressed. However, rival hypotheses do merit investigation, particularly if they can be proven to be more plausible. SARS-COV-2 might have stemmed from (ii) an outbreak at the Wuhan Lab's "gain of function" research, or a (iii) deliberate release in several countries from a biological weapons research project? Some critics even dispute the existence of SARS-COV-2, alleging that (iv) viral transmission is unproven, and that the entire COVID-19 "pandemic" is a psychological propaganda operation. By silencing such views, social media platforms limit their users' access to important debates, whilst weaponising deception in the interests of powerful groups.
N.B. This post is a work-in-progress and the list above is not exhaustive- kindly comment to recommend techniques that should be added, and suggestions for salient examples are most welcome.
Subscribe to:
Posts
(
Atom
)
Posts
orcid.org/0000-0001-9566-8983
