Another Killing on ICE

The right wing propaganda machine can’t spin the killing of Alex Pretti

I woke up Saturday to the news that ICE agents had killed another person in Minneapolis. I scrolled through my Bluesky feed for a few minutes, trying to find my orientation in an unsettled and rapidly evolving information space. Again, there was a video circulating. And again, the video had both a chilling clarity and just enough ambiguity to spark an intense period of collective sensemaking and provide fuel for the inevitable framing contests — and political spin — to come.

For more background on sensemaking, frames, and framing contests, please see this quick primer.

Several posters in my feed encouraged people to avoid the video. I support that recommendation. But I watched it anyway. Please skip the rest of this paragraph if you do not want to read about what is in that video. At the beginning of the clip, several agents have already surrounded and are violently engaging with a man who is struggling against them from the ground. At one point an agent begins to strike the man repeatedly. The man’s hands appear to be empty, on the ground in front of him, but it’s hard to see exactly what is happening. There is a flurry of movement. Then shots ring out — and continue — as agents move quickly away from the man’s still body.

The events captured in the video are horrifying. In the posts I saw in my feed on Saturday morning, the framing of those events was clear: another unjustified murder by ICE agents who continue to terrorize the people of Minnesota. (Renee Nicole Good was killed by ICE agents less than three weeks ago.)

But I know my Bluesky network skews left and isn’t a good reflection of conversations happening elsewhere. I was curious how things looked over on X, a platform that has in recent years developed a marked bias in the other direction, towards the right. In particular I wanted to see if, how, and how quickly influencers and others on the right were framing the events captured by this video in very different ways.

So I started collecting data, launching a series of queries of Twitter/X data using Brandwatch, a platform that provides access (for a price) to social media data, primarily for companies monitoring their brands. I focused my searches on posts that mentioned both the shooting, the agents, and the video.

What I found over on X was surprising, and the implications might suggest a changing information environment. I expected to see the right wing propaganda machine in action, spinning a story that blamed the victim, the community, the left, and anybody other than ICE agents or the Trump administration for the killing. And yes, those efforts were clearly there. But the frames they were pushing were not prevailing. The right was having trouble controlling the narrative, even on their home turf on X.

Let me show some of my work and invite you all into some of my data to see for yourself. Warning: I’m about to nerd out here, and you’re welcome to skip past the data section to the conclusions in the “Big Insights” section below.

---

Diving into the Data: Analyzing the X Discourse about Pretti’s Killing

My team has developed a visualization called a “Spotlight Graph” that shows how influential posts and accounts shape conversations on social media. I’ve created an interactive tool that allows me to load a few thousand X posts, plot them on a Spotlight Graph, and then poke around to see who is talking about what.

Below is a screenshot of data that I pulled from X from the morning of the shooting (a time window of ~8am to 1:30pm CST). This is a cumulative graph, which has time on the x axis and shows the total number of X posts (including reposts) on the y axis. For orientation, by 1pm CST there had been about 170,000 posts that met my search criteria for being related to the shooting. I plot individual X posts as circles along the graph at the time they were posted and these posts reveal the shape of the graph. The posts can be sized by different properties. In the screenshot below, they are sized relative to their follower counts, so accounts with large audiences appear bigger. In the interactive view, you can zoom in and select specific posts to see their author and content.

The image reveals that the early posts about the shooting were shared by accounts with relatively smaller followings. We see a first burst of activity from higher following accounts around 10:10am CST (which I will explain below). Overall engagement in the conversation surges (higher slope) between 11am and 12pm, as higher following accounts begin to participate. Large media accounts (like @FoxNews and @ABC) join shortly afternoon, but engagement begins to diminish after that time.

I’ve selected the first three X posts (that met my search criteria) about the shooting (with the red circle). The first, posted at 9:39am CST, stated that “ICE had shot a man multiple times” and noted that a video was circulating. The second, posted two minutes later, narrated the video, stating that “ICE wrestled [an] observer to the ground and then shot them.” At the time I collected this data (~1:30pm CST), neither of these posts had garnered much engagement.

The next post — from @intelFromBrian, a blue-check account “specializing in international breaking news” — was the first to go viral. It read, “#BREAKING ICE just shot and killed someone again. In Minneapolis. 5 shots. In front of the Glam Doll Donuts.” Though the language is somewhat neutral, the inclusion of the word “again” put the incident in conversation with the killing, less than three weeks prior, of Renee Nicole Good. Over the next four hours, @intelFromBrian’s post was reposted nearly 19,000 times.

Stepping back to look at the broader graph, we can see that the early conversation featured posts from relatively low follower accounts. But perhaps more interesting than follower count is the amount of engagement each post received. Here’s another version of the graph where individual posts are sized by how many reposts they received (in the first four hours).

Note the cluster of posts between 10am and 10:40am CST. (I’ve added a red box there.) These posts received, by far, the most engagement in our dataset. These are the posts that shape the broader discourse. And the vast majority of them were critical of ICE, sometimes implicitly, and other times explicitly calling out and blaming them for the “murder” or “execution” of “another person.” Below are a selection of the most highly reposted posts from that time:

During this same time period, a counter frame began to emerge — with the help of a Fox News journalist. Shortly after 10am CST, Bill Melugin reported via X that the victim, which he referred to as the “suspect,” had been armed. His post was sourced to DHS (the Department of Homeland Security) and contained an image of gun. This new information, which could be easily fit into a counter frame, set off a flurry of activity on the right.

We can see this in our data, as several right wing influencers posted content highlighting the “evidence” that DHS produced and using that to place blame on the victim. Here’s a selection of some of those posts, sized by number of reposts:

These five posts, shared between 10am and 11am CST, reveal the prominent frames on the right, suggesting that the victim was responsible for his killing, that he was armed and resisting arrest, and that Democratic leadership contributed. Some of these posts seemingly extend beyond claims from official sources to make false allegations that the victim was an “illegal alien” and contested claims that he brandished and/or fired the weapon. Others simply spin the new evidence — of the victim’s gun — into alternative interpretations about the causes of the event.

But perhaps the most striking thing about this graph is that these posts from influencers on the right framing the event as self defense by ICE agents do not get anywhere near the same amount of engagement as the posts by influencers on the left framing the event as another “murder” by ICE.

---

Big Insights from the First Few Hours after the Shooting

Let’s pause here and reflect a bit on the major insights from this analysis:

(1) On Saturday morning, ICE agents killed another person in Minneapolis. The killing was captured on video, and there were many witnesses.

(2) In the first few hours after the event, most social media users seemed to view the event in ways that were sympathetic to the victim and his community. This was even true on the X platform, where the majority of posts referencing both the shooting and the video were critical of ICE, expressing outrage at what was viewed as another unjust and unprovoked murder.

(3) The Trump administration (via DHS) activated quickly to try to shift the narrative. Approximately one hour after the shooting, they strategically released information about the victim — i.e. that he was carrying a firearm — that could be assembled to fit a counter frame. Right wing influencers activated quickly to spin this evidence into suggestions that the shooting was justified and made unsupported allegations that the victim had brandished his weapon.

(4) The speed at which the right wing propaganda machine activated and their dedication to selectively amplifying evidence to push their strategic frames was remarkable. But perhaps more remarkable was how relatively ineffective this effort was, at least in the short term, even on X.

(5) Caveat: It is possible that there are other conversations on X that better reflect the frames and politics of the political right — but that I missed them because my query focused on posts that mentioned something related to the shooting, ICE or federal agents, AND the video. There are other prominent posts, like the one from Bill Melugin introducing the “evidence” of the victim’s gun, that my collection did not pick up. I will note that Bill’s post (3400 reposts) would have been large in our graph, but certainly not the largest.

(6) To get a sense of what I might have missed, I did some follow up analysis of data from the next 24 hours, querying X data on the last name of the victim, Alex Pretti. Posts were predominantly sympathetic to Pretti, highlighting his occupation (ICU nurse), his background (boy scout, choir member), and stories from friends and others who were positively impacted by his many acts of kindness. In discussions about the gun he was carrying, the prevailing frame was that he was merely exercising his First and Second Amendment rights, with posters highlighting (and reposters amplifying) assertions that he had not threatened anyone with the weapon, and that the gun had already been confiscated when ICE agents shot and killed him.

---

A Final Note

My heart goes out to the families of Alex Pretti and Renee Nicole Good, as well as the people of Minneapolis, and all of Minnesota, especially the immigrants who have been terrorized by the actions of ICE. I can hear your pain in the videos of these shootings and in the testimonials you all are sharing on social media and beyond. I am awed by your courage, your efforts to try to protect each other from everything happening there. I hope that you can all stay safe. I pray that this horrible ICE shooting is the last.