On October 7, 2023, I was working as an analyst at a news verification company, which is a fancy way to say that I was a fact-checker. Part of this work involved combing social media for false claims that we could debunk, and since my focus had been Iran and the Middle East, I was quickly assigned to the new Israel-Hamas war. And there I sat, scrolling Twitter and Instagram to see what I could verify (or not).
When violence erupts, social media becomes a tableau. I saw photos showing charred bodies and squinted at videos of bloody corpses and hostages. But the most jarring content was something you might not expect: A video of a baby, gray-faced and swaddled in dusty blankets, shivering as ash fell around him.
I don’t know where that baby is today, or where the video is, or if the footage was even real. I do know that when I saw him for the first time, I lingered on him for a moment before scrolling past him, only to feel some sort of itch enter my head a few minutes later. Another few minutes and I was taking deep breaths in a conference room. Not crying, not exactly—instead of full of feeling, I was empty, taken hold of by a sort of desperate void. And I can still see the baby clearly in my mind’s eye, still feel a whisper of that void, despite the two years that have passed.
I’m thinking of this baby today in the aftermath of two deaths: of the right-wing activist Charlie Kirk, who was gunned down in a single shot in Utah last week, and of Iryna Zarutska, the Ukrainian refugee who was stabbed to death by a schizophrenic man on the Charlotte light rail last month.
Both Kirk’s death and Zarutska’s were captured on camera, and both videos are awful. In CCTV footage, Zarutska is dressed in a dark shirt and light pants, unaware as her killer rises up ghoulishly behind her. Zarustka spasms, and for a moment you wonder why she looks so shocked. Then you see the blood begin to flow. In the footage of Kirk, there is no warning signal; one moment he is talking, seated under a tent, and the next he crumples as if he’s been punched in the neck. People scatter.
The viral spread of these videos has attracted attention. On Twitter, some commentators advised followers to turn off the autoplay feature of their video feeds, and in my own private group chats, people warned others not to watch the Kirk footage. And in our own pages, Emmett Rensin noted how, “By now, millions of Americans have watched Brown rise from his seat and loom like some strange beast over the unsuspecting Iryna Zarutska before stabbing her in the throat.”
For a long time, such things were only visible up close, usually to the victim, the perpetrator, and bystanders. What happens to us, though, when suffering is scattered? What happens when a deathbed is democratized?
We have long wanted to view violence. In 1939, the first filmed guillotining took place in France—and after it became clear that such a public execution was not just a deterrent, but a spectacle, guillotinings became private. “After that, the guillotine was rolled away behind prison walls – not because decapitations were too horrifying to watch, but because people will watch them no matter how horrifying they are,” wrote one historian.
Fast-forward to more modern times. In 2014, photos and videos of beheadings by ISIS began to propagate across the internet, and starting in 2022, footage from the Ukraine war proliferated. But violence does not need to have a geopolitical bent to attract eyes: Last month, the Anti-Defamation League reported on an internet forum called “WatchPeopleDie,” which is what it sounds like—“a forum where users can post and view real images and videos of violence, including murders, torture, rape, executions, beheadings, suicides, dismemberments, accidents and animal killings.”
Two kinds of people form this online bazaar. One is the kind of person who posts these videos. The other is the kind of person who watches them.
The former is more of a wild card. A tiny but influential subset of WatchPeopleDie videos are the personal livestreams of mass shooters as they commit their crimes; the ADL notes that a post with footage of the 2019 Christchurch shootings in New Zealand “has nearly 700,000 views, 5,000 comments and 5,500 upvotes alone.” And while it’s difficult—and probably inadvisable for most people—to tease out the exact motivations of filming such violence, experts have speculated about a drive for attention and a desire to spread violent ideology. With this in mind, suspected motivations of mass shooters and terrorist beheaders actually don’t appear to be that different: According to one historian, the motives for beheadings—and posting them—range from intimidating local populations to attracting the attention of, and concessions from, major world powers. Another motive is recruitment: per Paul Cruickshank, a terrorism analyst speaking in 2014, some viewers “almost have a pornographic attraction to these violent scenes, these violent beheading videos. It really sort of energizes them.”
The idea of posting violence to recruit others blurs the line between poster and viewer—and why anyone would want to view these videos is an open question.
Researchers from the University of California, Irvine, wrote in 2019 that when a person is already anxious about the events at hand, it is likely difficult to turn away from footage of those events. In the case of 9/11, for example, a sample of Americans reported watching more than eight hours of related news coverage—which certainly included videos of the towers coming down—the day after. It makes sense, then, that anyone already afraid of random or sudden violence might scroll on social media, see a video of Kirk or Zarutska, and pause.
There is certainly a subset of people whose violence consumption is driven by a mind so different from my own that any quick efforts to truly understand will fall flat. I am entirely certain that there are people who watched Kirk get shot and laughed—and played the footage again. It’s also true that, even 24 years after 9/11, some people still voraciously consume video of that day, sometimes looking for evidence that will finally confirm their suspicions that it was, indeed, an inside job. There are also those who are attracted to spectacle. One commentator in 2023 confessed that he watches 9/11 footage when he is bored, writing: “You know you ought not to indulge yourself in watching the wretched spectacle, yet you just can’t seem to look away.”
Can’t look away, or won’t? As an anxious person I confess that during moments of national upheaval—Trump getting shot, Kirk getting shot—my instinct is to scroll Twitter. But the great part about having free will is that I don’t have to.
There are many reasons not to lay eyes upon the gory spectacles we are offered. Most simply are the psychological considerations: Studies done in the wake of 9/11 show that, for people who saw comparatively more images and watched comparatively more TV of the event, the risk of post-traumatic stress disorder was higher. This “suggests that exposure to real graphic media coverage carries real psychological risks for viewers,” the University of California, Irvine, researchers wrote—and in 2019, those same researchers found similar results for those who watched terrorist beheading videos. And more generally, children exposed to violence on TV might become desensitized to violence, as well as more fearful of the world around them. Watching graphic violence, then, is simply not good for you.
Then we have the moral considerations. There is something awful about the thought of millions of strangers seeing Kirk’s last moments before many of his loved ones did. Typically when someone is vulnerable we offer them comfort and privacy; what is the protocol when someone is dying, arguably the most vulnerable of all states? Not, I think, to take their vulnerability and milk it for our own entertainment.
So why watch death and dying? If you truly have compulsions you can’t resist, you should probably see a doctor. But I suspect that most people are like me: They scroll, they watch, and—like the video of the shivering baby—perhaps the effects don’t hit them for some bit of time. In the meantime, they scroll and see more violence, more death, more suffering. Anxiety rises. Emptiness builds. And someone bleeding out becomes just another part of the noise.
While discussing the Kirk footage with some friends, I lamely noted that Twitter technically has guidelines against violent content. This was met with laughs, including from myself: Twitter has never been a peaceful paradise, and the Musk acquisition in 2022 seems to have moved it further from that ideal.
Still, let’s look at the guidelines. One section notes: “You may share graphic media if it is properly labeled, not prominently displayed and is not excessively gory or depicting sexual violence, but explicitly threatening, inciting, glorifying, or expressing desire for violence is not allowed.” I’m not really sure what much of this means—how can you “prominently” display something on Twitter?—and the (loose) prohibition on “excessive” content seems more designed to protect Twitter in the courtroom than to protect users. Because what is excessive? If Kirk or Zarutska had bled more, would that have been excessive? Even if that question made it in front of a decision-maker, in the meantime the videos would just keep circulating.
The other major hole is the (again, loose) prohibition on inciting violence. The idea of “contagion” when it comes to violence like mass shootings is common in psychological literature, and the aforementioned ADL report noted that two mass shooters had footprints on the WatchPeopleDie forum before they committed their crimes. “WPD serves as a virtual space where users — young people in particular — can access extremist content alongside graphic violence, potentially desensitizing them and increasing the risk of ideologically motivated violence,” the report states. In other words, watching violent content can cascade into committing violent action.
It does strike me that, even if restrictions against violent content were enforced across social media platforms, they still wouldn’t be fully protective. Even if Twitter had moderators, bot content and engagement farms are powerful, and we know from report after report that if there’s any content the algorithms like, it’s the emotionally charged stuff—and violence certainly counts. And of course there are still forums like WatchPeopleDie. What this means is that, even if social media suddenly became much more policed, there’s no indication that the rot of violence would be rooted out.
But even more so, I imagine that prohibitions against “violence” would still miss a lot of what disturbs us. Think back to the baby I viewed in 2023: No one was stabbing it or otherwise committing a ghastly, personal crime upon it. But still it haunts me, because babies shouldn’t be ashen; babies shouldn’t be shivering. Is this violence? According to a simplistic definition, probably not. But what is violence if not a method of harming a person, of stripping away their wholeness and dignity and sometimes their lives? What is violence if not something that degrades human life itself?
I do not expect software companies or algorithms to care about this, and it doesn’t seem like they will anytime soon. And so it’s up to us, the individuals, to make the choice to actually judge what’s on our screen, and say: No. No more.
