The Christchurch killer’s livestream video: 2019’s distribution versus 2024’s accessibility

The Southern Poverty Law Centre (SPLC) released on 21 June 2024 a new report titled Digital Threat Report: Mass Shooter Videos, penned by Dr Megan Squire, the Deputy Director for Data Analytics, and OSINT.

The report notes,

Years after three horrific mass shootings, videos created by the different attackers behind them are still present on social media and mainstream streaming services. Some services are facing challenges detecting and getting rid of these videos, and one, Elon Musk’s X (formerly Twitter), doesn’t appear to be trying to remove these videos with any consistency or rigor. After Hatewatch reported the videos, which feature mass murder, X responded that they do not violate their terms of service, then abruptly removed the videos after receiving a request for comment on this story. Perpetrators of massacres in Christchurch, New Zealand; Halle, Germany; and Buffalo, New York, filmed themselves as their crimes took place and simultaneously livestreamed the murders to social media sites for others to discover and view. In each case, viewers have downloaded their own copies of the videos and posted them onto other social media platforms and websites, spreading the videos far beyond their original online footprints.

I spent around 8-10 hours last Tuesday locating the findings of SPLC’s research in Aotearoa New Zealand’s contemporary disinformation ecologies, and discourse. Though a chapter in my PhD studied the March 2019 terrorist attack in Christchurch, it did not study the propagation of the video per se.

One individual thanked in my PhD is David Shanks, who was at the time of the attack New Zealand’s Chief Censor, at the NZ Classification Office. Shanks was at the coalface of responding to the video’s seed, and spread – and a TEDx talk in Christchurch four years ago captures how he responded to it. I believe New Zealanders owe him, and the Classification Office, a deep debt for protecting them, at the time, against the diffusion of the killer’s livestream.

https://www.linkedin.com/embeds/publishingEmbed.html?articleId=7524974196162997135&li_theme=light

The Christchurch killer’s video is one of two others the SPLC report studies the dispersion of. It’s not the first study to do so. The Christchurch Attacks: Livestream Terror in the Viral Video Age by Graham Macklin remains the most comprehensive study of it I’ve read to date. What Squire’s analysis does is to locate March 2019’s livestream video with two others like it, and map a global economy around this content’s appeal, and distribution.

Which brings me to why this is now central to what I study.

In 2021, when I started work with The Disinformation Project project, I thought I’d be looking at anti-vaxx commentary, related disinformation, and what WHO called the infodemic – not at white supremacist terrorism, and its digital footprint. However, through phenomena called network, and community bridging, anti-vaxx networks in 2020 are, in New Zealand, a more amorphous, porous, fluid collection of online networks today, featuring anti-vaxx narratives as much as anti-establishment, anti-government, anti-authority content, including accelerationist, neo-Nazi material. This isn’t an occasional content, and commentary signature. It’s something I now study every day.

Megan Squire’s robust analysis in SPLC’s report flags the very high propagation of the Christchurch killer’s livestream video immediately after the offline violence, which subsequently died down because of platform TVEC policy enforcement, and New Zealand’s classification of the material as objectionable.

My research brings out, counter-intuitively, how this is a completely misleading capture of the livestream video’s persistent seed, engagement, and accessibility within domestic, and trans-national networks I study. This may come as a shock to many who believe that 2024’s global, and domestic information ecologies remain the same as those in March 2019, rendering the livestream video as hard to access as it was five years ago.

And therein lies the rub.

As I wrote in Tuesday’s analysis,

What we find in Aotearoa New Zealand’s information, and media landscapes is far worse than what the SPLC research brings out, and just in relation to the continued dissemination, diffusion, and accessibility of the Christchurch mosque massacre’s livestream video. We find hundreds of videos featuring the complete version of the livestream, and various edited versions, across many web platforms that include, and also go beyond the SPLC’s selection, inter-linked to each other, signalled on domestic Telegram accounts. We find a website dedicated to gore, and graphic harms, which features the full livestream video, and over 2,700 comments celebrating it, and the killer’s ideology. This material is linked to on other platforms. We note that during Christmas/NZ Summer 2023, content featuring the Christchurch livestream exploded on Twitter through what was very likely a bot network.

What follows are lightly edited excerpts from the same analysis. For obvious reasons, I will not disclose specifics of where, and how the March 2019 livestream can be found. Out of respect for those in New Zealand, and following the example of the former Prime Minister, I will also not mention Christchurch terrorist’s name.

2023 vs 2024: What’s changed in a single year?

On 21 March 2023 (i.e., last year), I studied over 250 videos related to the Christchurch livestream, and the Buffalo, New York livestream on a single platform that’s not explicitly noted in the SPLC’s report. This platform is best described as a really niche video hosting platform, with comments that mirror the discourse on 4Chan around content that features explicit gore, violence, and content like livestream videos of terrorists.

To me, this speaks to a possible limitation of SPLC’s study, which relies on software from Pex, “a content identification technology company that helps creators find audio and video content online”. I do my research by looking at community, and network bridging manually, and from a grounded perspective anchored to New Zealand’s unique information, and media ecologies. I’ve also done this research every day, since July 2021, which gives me a country-specific expertise researchers from outside will not have, and even advanced software will not provide. But because SPLC’s research is comprehensive across the platforms it does capture, the insights are a useful foil to what I go on to note below.

Worth noting that the Buffalo livestream is also classified as objectionable in New Zealand. And yet, I studied on 21 March 2023 a greater number of videos around both incidents than what New Zealand’s Department of Internal Affairs (NZ) reported in its annual report it had taken action on over the whole of 2022.

As I noted in that day’s analysis,

1. All the videos celebrate, glorify, and worship (as Saints) the perpetrators of the respective killings.
2. Most of the videos feature hard rock, or heavy metal soundtracks, that to the extent discernible, also feature white power lyrics.
3. The glorification of the terrorism, on the lines of what a 2021 video produced by the Economist magazine noted in related to the Christchurch killer, is through game mods, and game environments – which render the real world terrorism as shoot ‘em ups, erasing the distinction between virtual, and offline, in-game violence, and physical, kinetic consequence.
4. Not a single comment studied was opposed to the depiction of the graphic violence.
5. There are dozens of connected accounts to every single video studied above. It is very likely that a platform-level study of material classified as objectionable in Aotearoa New Zealand will result in hundreds more videos featuring violative material. XXX is thus one of the largest repositories of video material celebrating violent extremism known to TDP outside of the dark web, encrypted, and private file storage.

Last Tuesday, I studied the same XXX platform, which isn’t specifically named in the SPLC’s report. Searching for “Christchurch” surfaced 244 results, and nearly all of them were around the Christchurch killer’s video. Many of the videos studied in March 2023 remain. New ones have been added in the past year. The most recent upload is from nine months ago.

The most viewed videos generate thousands of views each. The single most viewed video was uploaded a year ago and generated 4,186 views at the time of writing. It is a play-through of a Doom game mod/level, set as the Linwood Mosque. The description to the video features a link to the Wayback Machine/Internet Archive that hosts a wad file. wad is the file extension for Doom game level mods. This file has generated 894 views, and was uploaded on 22 January 2022 by an account whose nomenclature features a symbolic and coded reference used by neo-Nazis and white supremacists, embodying their ideology and signalling their identity within the movement.

In March 2023, searching for “mosque shooting” on this platform resulted in 56 entries. On Tuesday, 25 June 2024, the same search surfaced 600 videos. This is a nearly 11 fold increase, year-on-year.

Furthermore,

• The Christchurch killer’s name brought up 76 results in 2023. This had increased to 804. A nearly 11 fold increase.
• “Buffalo shooting” brought up 40 results in 2023. This had increased to 428. Another nearly 11 fold increase.
• “Payton Gendron” surfaced 24 results in 2023. This had increased to 294. Over a 12 fold increase.

I also searched for,

• A deified variation of the Christchurch killer’s name, used in far-right discourse. This resulted in 19 videos .
• The killer’s surname. 903 videos.
• “Linwood”. 9 videos.
• “Christchurch mosque”. 48 videos.
• “Buffalo shooting”. 428 videos.

The volume of content was too high to individually assess the violative nature of. To get around this, I looked at the first three pages of the search results for all the terms above. For every term, the still video image against videos that surfaces were of the Christchurch killer, the mosque, victims, and in some cases, extreme gore (stills of victims being shot or soon after they were killed).

Exactly like last year, I studied in a single day content that was by order of magnitude more than what the 2023 annual report of the Department of Internal Affairs (NZ) records as material related to the Christchurch massacre it took action on over whole of last year.

Prima facie, it suggests that the apex agency of the government of New Zealand looking at online harms is unaware of or can’t do anything about the explicit presentations, and various violative permutations of the Christchurch video hosted on just a single video platform, leave aside the others that SPLC’s report surfaces, and my own research regularly discovers.

As the SPLC report also noted,

Some of the videos were taken down in the direct aftermath of the incident, but as we compiled this report, we found dozens of cases of all three videos still available for viewing on multiple platforms, including X/Twitter, Facebook, Vimeo, Soundcloud, Reddit, VK, Streamable, Dailymotion and others.

Highlights from the research

Since March 2023, I’ve penned around 35,000 words of analysis pegged just to the spread of the Christchurch livestream video in domestic (i.e., New Zealand specific) online ecologies. A dozen salient points of this analysis follows.

1. There was unprecedented spread of the Christchurch killer’s livestream content online since August 2023, triggered by a takedown notice from New Zealand’s Department of Internal Affairs (DIA) to Gab, regarding a documentary – also classified as objectionable within New Zealand (when Shanks held the office of Chief Censor) – which features the entire livestream embedded in it.
2. The DIA takedown notice created a Streisand effect, leading to the widespread sharing of links of this film (first released early 2022). This resulted in an explosive diffusion of the Christchurch killer’s livestream content.
3. Gab’s CEO Andrew Torba publicly refused to comply with the takedown notice and shared direct links to the content, leading to massive exposure across Gab’s entire global user base at the time.
4. The news of the takedown, and links to the content spread rapidly across multiple platforms including Gab, Telegram, Twitter, Facebook, and Reddit, reaching potentially hundreds of thousands of users.
5. Several domestic far-right and conspiracy theory channels on Telegram shared links to the film, exposing their followers to the content.
6. Local downloads and private sharing of the film likely occurred, creating permanent access points that are difficult to track or remove. This includes versions of the film that remain on specific apps I study, known to harbour far-right, neo-Nazi communities as well as anti-vaxx, anti-government, anti-establishment networks.
7. Custom game level mods based on the Christchurch attack for games like Doom were discovered, further propagating imagery and themes from the attack.
8. Numerous videos on platforms not included in the SPLC’s study continue to host edited versions of the livestream footage, often framed as “exposing the truth” about the attack. This includes the full version of the livestream, and various edited versions of it.
9. A marked resurgence of false flag conspiracy theories around the Christchurch attack, with many users claiming the footage proves it was staged.
10. There are cumulatively thousands of comments against these videos.
11. Content from an Australian conspiracy theorist featuring extensive Christchurch footage has gained significant attention after another DIA takedown notice. Like the takedown notice issued to Gab, the owner – who is not in New Zealand – refused to comply, resulting in another Streisand Effect moment leading to the rapid diffusion of the original video. Dozens of copies now exist under various accounts, in addition to the original’s continued accessibility.
12. My research consistently suggests current content moderation and takedown strategies, some of which were conceived of after March 2019’s terrorism, are inadequate for dealing with the evolving nature of extremist content sharing online. This includes significant issues with the Christchurch Call Foundation’s approach to TVEC in contemporary media, and information ecologies (something I will return to later, because it’s flagged in SPLC’s report as well).

SPLC and TDP research: Points of convergence

The research I lead at The Disinformation Project into the propagation of the Christchurch killer’s livestream mirrors several key aspects of the Southern Poverty Law Centre’s (SPLC) research on mass shooter videos. The quotes below are from the SPLC’s report or from analysis I’ve penned.

1. Persistent availability: SPLC/TDP research highlight the ongoing availability of these videos long after the initial incidents. TDP flagged, and found an “unprecedented spread” of the Christchurch livestream content in 2023, while the SPLC report notes finding “dozens of cases of all three videos still available for viewing on multiple platforms” years after the attacks.
2. Cross-platform spread: SPLC/TDP research emphasise how the videos spread across multiple social media platforms. I observed the content spreading “across Gab, Telegram, Twitter, Facebook, and Reddit,” while the SPLC tracked the videos “across more than 20 platforms.”
3. Challenges in content moderation: SPLC/TDP research highlight the difficulties platforms face in detecting and removing these videos. We noted that “Meta’s own hashing seems to have failed,” while the SPLC found that some platforms, particularly X (formerly Twitter), were inconsistent or ineffective in removing reported content.
4. Role of alternative platforms: SPLC/TDP research point out how less regulated or “alternative” platforms play a significant role in hosting and spreading this content.
5. Modifications and remixes: SPLC/TDP research discuss how the original videos are often edited, remixed, or incorporated into other content. We observed custom game mods based on the attacks, while the SPLC found “audio remixes” and attempts to “remix the originals into video games.”
6. Inspiration for copycats: SPLC/TDP research highlight how these videos can inspire copycat behaviour. My analysis noted the “unprecedented global spread of the Christchurch livestream,” potentially inspiring others, while the SPLC explicitly discusses the “copycat behaviour that carried over into multiple other incidents.”
7. Conspiracy theories: SPLC/TDP research mention the proliferation of conspiracy theories around these events. We observed a “resurgence of false flag conspiracy theories,” while the SPLC noted videos centred around “false flag” conspiracy theories.
8. Inadequacy of current moderation strategies: SPLC/TDP research suggest that current content moderation and takedown strategies are insufficient. Our analysis notes that “enforcement protocols, and regulatory frameworks around stymieing the spread of harms that are very close to expiration, and irrelevance,” while Megan Squire’s research recommends more proactive methods of detection and removal.

Christmas 2023, and Twitter’s role in spread of the Christchurch video

I also (sadly) studied the unprecedented seed, and spread of the Christchurch killer’s livestream video on Twitter during the Christmas break of 2023.

From 25 to 27 December 2023, there was a significant surge in the propagation of the Christchurch livestream video, Buffalo mass-shooting livestream, and related content on Twitter. Content included the entire Christchurch killer’s livestream, and various permutations of it, including excerpts, edited content, memes, stills (i.e., captures of still frames from the livestream), and other WSVE content including the Buffalo killer’s livestream, and the cover page of the Christchurch killer’s screed.

At the time, direct reports of this material to New Zealand’s Department of Internal Affairs (NZ) went unheeded, and received a standard out-of-office/on-holiday response. It was only because of some tweets of mine, which caught the attention of a senior Twitter representative, leading to the application of relevant TVEC policies, that this content is no longer present on the platform to the degree it was for that 72 hour period.

And this is also why this content doesn’t find expression in SPLC’s report.

I wrote on Christmas Day 2023 that “The presence of these accounts represents a multifaceted threat, extending beyond the potential for terrorism to broader societal impacts, including the erosion of democratic norms and the promotion of hate and division. It also speaks to the inability of the Christchurch Call, and GIFCT to effectively, and proactively capture, contain, and control the spread of TVEC on Twitter.”

This gels with what SPLC’s report noted,

X/Twitter failed a similar “stress test” of its systems conducted by the EU in 2023 to determine if the platform was prepared to be governed by this new law. One potential solution is to include the signatures of these shooter videos inside a detection technology, such as GIFCT, that platforms use to proactively find and remove other types of terrorist content. GIFCT announced in 2022 that it had done this with these shooting incidents, and reported their findings in an annual transparency report. Nonetheless, it is clear from our study that some platforms, like X, either have not adopted the GIFCT technology or have not implemented it correctly, since these terror videos are still available on the platform.

Twitter remains a significant reputational threat to the Christchurch Call Foundation.

Implications

The fact that I do not focus my research on far-right ecologies is consequential. It means that when I do increasingly find far-right content, and commentary, it’s circulating in communities, and networks that are largely NOT defined by subscription to violent extremism, and will vehemently deny they are part of the extremist networks. The presentation of material like the Christchurch killer’s livestream, and screed to these communities, and networks is deeply problematic, because of the immediate, and enduring radicalisation pathways created.

Anti-authority, anti-government, and anti-establishment conversation is near exclusively defined, and undergirded by the vehement conspiratorial condemnation of things that are believed to be deliberately hidden from the public by an all powerful ‘cabal’. This belief is driven by a militant distrust in official sources, institutions, and narratives. When violative material, which in New Zealand is illegal to engage with, is presented as ‘hidden truths’ that people need to do their own research around, the initial interest in engagement is very high, and may lead to a subset of those who see violative material becoming more interest in similar content. The networks within New Zealand I study feature hundreds of thousands of accounts, which even allowing for duplication, is a lot of people in a country of just around five million.

Relatedly, what the SPLC doesn’t get into is the discoverability of the Christchurch killer’s screed. This is also something I’ve studied, alongside the diffusion of the livestream video. And this too is in 2024 easily accessible, and discoverable including through alternative search engines to Google, cloud hosting platforms, and leading torrent archives/repositories.

Why does any of this matter?

Megan Squire reiterates in SPLC’s report what earlier research has also established: “Plentiful evidence from his own writings indicates that Payton Gendron, the perpetrator in the Buffalo attack, was inspired by both the beliefs and the methods employed by the Christchurch attacker.”

A couple of points spring to mind.

• In 2024, the Christchurch video livestream is distributed less on the platforms that are part of the Christchurch Call Foundation (including Meta, Google etc), but counter-intuitively is far easier to discover, access, download, and engage with on other platforms. While the mass spread of this material is indubitably good, and beneficial, what many don’t realise is the degree to which the content, and endless permutations of it are downloadable, and discoverable online – without ever going to the dark web. Those in New Zealand’s disinformation networks, for example, can find this material easily, including through unsolicited presentations of links by accounts clearly connected to the far-right.
• The platforms, and apps this content is now on defy domestic regulatory, policy, and policing oversight. Telegram is one of them, and in New Zealand, is one of the worst repositories of gore, graphic harms, and violent extremism. Even with now around a billion users globally, the platform has no demonstrable interest in or commitment to addressing the spread of TVEC. As Eva Galperin, the Director of Cybersecurity at the Electronic Frontier Foundation (EFF) noted recently, “What makes Telegram different (and much worse!) is that Telegram is not just a messaging app, it is also a social media platform… [Just] Thirty engineers’ means that there is no one to fight legal requests, there is no infrastructure for dealing with abuse and content moderation issues.”
• The extreme gore, violence, and harms associated with the Christchurch killer’s livestream, and the violent extremism communicated in his screed are foundations for radicalisation, and stochastic terrorism, even within New Zealand.
• On Tuesday, one of the video platforms I studied, which I hadn’t encountered previously, was dedicated to gore, and videos of individuals being killed or dying in horrible accidents. The entire Christchurch livestream is archived there. There are nearly 3,000 comments against the video. Some of the comments feature edited clips of the livestream, with additional graphics of the killer superimposed to music, or edits that focus on the moments victims were gunned down. All of the comments studied celebrated the killer, and the murders.

Instead of redoubling efforts to take an all-of-government, and whole-of-society approach to violent extremism’s enduring threat to democracy, institutions, a rules-based-order, and social cohesion, the current government in New Zealand is defunding grounded research into countering violence extremism, and jettisoning critical architectures, safety nets, and official capabilities to address online harms, in addition to more distressing measures to cut, and curtail March 2019 victim support.

New Zealand’s Department of Internal Affairs (NZ), which hasn’t been spared from job cuts, and very possibly because of this, now takes an incredibly pedantic, and irascibly myopic approach to online harms.

The country’s disinformation ecologies, that at the height of the Covid-19 pandemic were almost exclusively coloured by anti-vaxx discourse, are now defined by mixed, unclear, and unstable violent extremism (MUUVE), in addition to network, and community bridging with outright militant accelerationism, white supremacism, and neo-Nazism. And yet, the country’s foreign minister, in addition to regularly appearing in productions, now publicly participates in functions with individuals behind a very well financed, highly motivated, far-right disinformation network that openly promotes ‘The Great Replacement’ conspiracy theory – which ironically is exactly the same violent ideology that inspired the Christchurch killer.

I genuinely can’t fathom whether the likes of NZ Police, NZ SIS, and GCSB aren’t aware of what I study, or if the classified advisories they give government officials are going unheeded or under-appreciated.

My research accounts for the world as it is, not as it should be. Things being the way they are in New Zealand, it is doubly concerning that policymakers, and official institutions do not grasp how 2024’s information, and media ecologies are so far removed from what David Shanks competently, and conscientiously responded to in the aftermath of the horrific March 2019 terrorism. This is something the SPLC’s report also highlights. What was done, and worked five years ago aren’t remotely viable solutions to the challenge of this content’s continued diffusion, easy accessibility, and persistence online despite existing TVEC oversight led by the likes of GIFCT, and various domestic entities. Megan Squire has some good recommendations in this regard.

What this all simply means is that what inspired, and radicalised the Christchurch killer, and the video content he produced are widely available today on platforms that the SPLC report flags as well as a number of others it doesn’t, but are easily discoverable through often explicit link sharing. The presentation of this material in New Zealand’s disinformation ecologies is well established, and in ways that defy regulatory oversight, blocks, bans, and takedown orders.

This should be a wake-up call for New Zealand’s policymakers, given what SPLC’s, and other’s research very clearly establishes as what results from the consumption of this material.

Cover image courtesy Time magazine, and photographed by Virginia Woods-Jack.