The Trump administration and some Democrats have drawn divergent conclusions from bystander video of the fatal shooting of a woman in Minnesota by an Immigration and Customs Enforcement agent. How can one side say the agent was “recklessly using power” and the other determine he “fired defensive shots”? Experts told us it’s common for people to view the same video differently, and that the early evidence isn’t enough to reach definitive conclusions.
Shortly after the Jan. 7 incident, in which 37-year-old Renee Nicole Good was killed by the agent in Minneapolis, President Donald Trump posted on Truth Social a video clip of the shooting, captured from a distance, and said that the woman “violently, willfully, and viciously ran over the ICE Officer, who seems to have shot her in self defense.” Additional bystander video, captured closer to the shooting, showed the agent wasn’t run over but left unclear whether the vehicle struck him. Republicans and Democrats have still disagreed on what the early video evidence depicted.
Department of Homeland Security Secretary Kristi Noem said in a Jan. 7 press conference, “This appears as an attempt to kill or to cause bodily harm to agents, an act of domestic terrorism,” determining that the ICE officer “fired defensive shots” because he was “fearing for his life.” The next day, Vice President JD Vance echoed that assessment, saying, “She was trying to ram this guy with his — with her car.”
Democratic leaders in Minnesota have disputed that. Minneapolis Mayor Jacob Frey said on Jan. 7, “So they are already trying to spin this as an action of self-defense. Having seen the video … myself, I want to tell everybody directly. That is bullshit. This was an agent recklessly using power that resulted in somebody dying, getting killed.” In a press conference that day, Gov. Tim Walz referred to “a very difficult video to watch,” saying it was “beyond me” that Noem “has already determined who this person [Good] was, what their motive was.” He criticized the large deployment of federal agents in Minneapolis and said that Good was killed “for no reason whatsoever.”
New York City Mayor Zohran Mamdani told CNN on Jan. 9 that Good had been “murdered.” He said, “That was the conclusion I came to just in watching that video. And I think that many Americans came to that same conclusion.”
“It is not only very common for different people to watch the same video and come to different conclusions, it is almost inevitable,” Seth W. Stoughton, a law professor and faculty director of the Excellence in Policing & Public Safety Program at the University of South Carolina, told us in an email. “First, people can disagree about the underlying facts of what happened. For example, viewers may disagree about whether the agent was standing in front of a car or next to the car at a specific time. Second, even when people agree about the facts, they can disagree about the conclusions that can be drawn from those facts. For example, even if everyone agrees that the agent was next to the car, they might disagree about whether he was in danger of being struck by the car or whether the use of deadly force was appropriate.”
Ed Obayashi, an expert on use-of-force cases, and deputy sheriff and policy adviser for the Modoc County Sheriff’s Office in California, told us that “it is basic fundamental human nature” for people to take sides when viewing these videos. “It is too premature” to make conclusions, he said in a phone interview. The investigation “is at its baby stages. It’s going to take months and months, if not a year or more, to come to a conclusion.”
John R. Black, a former law enforcement officer who has been an expert witness in police practices cases and specializes in analyzing decision-making, told us, “All the video does is demonstrate the need for questions,” adding that the video “by itself, can never be conclusionary.”
Yet, viewers jump to conclusions. “While video is objective evidence, interpretation is always subjective,” he said in a phone interview. The viewer attaches meaning to the video, “but that’s the viewer’s meaning.”
Readers can watch these videos and other bystander footage for themselves, including frame-by-frame breakdowns by the New York Times and Washington Post.
Stoughton said disagreements on what video in these types of cases shows are largely due to “cognitive biases,” or “unconscious processes that our brains use to make sense of the world without being overwhelmed” by filtering our perceptions “through the lenses of previous experience, identity, and expectation.” He cited two cognitive biases: motivated reasoning and confirmation bias, both tendencies that lead observers to interpret information to reinforce or confirm their identities and worldviews.
“A progressive Democrat who identifies, in part, as being anti-immigration enforcement and a MAGA Republican who identifies, in part, as staunchly supportive of the administration’s immigration crackdown are motivated to see the same incident very differently,” he said, as an example of motivated reasoning.
Video “can be highly informative,” Stoughton said. “However, the existence of video does not assure complete agreement about what happened or how the facts should be characterized.”
Black also cited “outcome bias.” The people shown in a video are making decisions without knowing how it will turn out, but viewers watch these actions in the video while already knowing the outcome. “We as a viewer of the video, we are weighing in on the outcome, not the decision,” he said.
Videos Only Part of Investigation
The two early videos of the incident — the one Trump posted and a closer-up video widely circulated later on Jan. 7 — together show ICE officers approaching an SUV that appears to be partially blocking traffic on a residential street. One officer approaches the driver’s window and then reaches for the door handle. Another officer walks around the passenger side of the car and then in front of the vehicle. The shooting incident happens quickly. In less than five seconds, several actions occur: The car backs up and then moves forward, and the officer at the front of the car draws his weapon and fires, while moving to the side of the vehicle as it speeds away.
The experts we interviewed said there’s a lot more information for investigators to gather. Initially, the FBI and Minnesota Bureau of Criminal Apprehension were jointly conducting an investigation, but now the FBI is solely leading it, the BCA said in a Jan. 8 statement. Walz said that same day that “Minnesota must be part of this investigation.” In a press conference, the governor said of the FBI being solely in charge, “It feels very, very difficult that we will get a fair outcome, and I say that only because people in positions of power have already passed judgment.”
Obayashi told us having a moving vehicle “makes it much more complicated in terms of evaluation.” He said investigators will recreate what happened, including the speed, direction, where everyone was positioned. “You’re going to have to correlate all that against witness statements,” and analyze all of the available video, including any body-worn police cameras.
On Jan. 9, video that appeared to be from the shooting agent’s cell phone was obtained by several news organizations.
“It can be extremely helpful for investigators to synchronize multiple videos, which can create a way of looking at the same scene from multiple perspectives simultaneously,” Stoughton said. The investigators will also interview the agent who fired the shots, other agents and bystanders. “Witness descriptions almost inevitably disagree to some extent, but they can help shed light on what happened off camera or on how to interpret what is visible on camera.” Forensic evidence on the trajectory of the bullets “can also help shed light on the facts of what happened.”
Black said that “no video can tell you what occurs in the mind of the actor,” meaning we don’t know about the agent’s “subjective inference” and his decision-making. For example, a New York Times analysis of video from different angles said that “the vehicle appears to be turning away from a federal officer as he opened fire.” But, Black asked, could the agent have perceived that the car was turning away in time? “It has to be examined.”
We also can’t know the viewpoint of Good.
The agent’s thought process is critical to a determination of whether the use of deadly force was justified. “Police officers can generally use deadly force in two circumstances,” University of Virginia School of Law professor Rachel Harmon, an expert on policing and the law, explained in a 2021 video on the topic. “The most common is when they feel they are threatened with a use of force or a threat of force that would cause serious bodily harm or death. … And the other is when someone is fleeing a dangerous crime – when someone has engaged in a crime of violence, and are fleeing and the only way to subdue them is to use deadly force against them.”
The administration has cited the first reason. DHS also has a policy on use of force that said, as of 2023, that its officers and agents “may use force only when no reasonably effective, safe, and feasible alternative appears to exist and may use only the level of force that is objectively reasonable in light of the facts and circumstances.”
Stoughton told us there are three likely questions for the investigation: “1) Did the agent put himself into the vehicle’s path of travel unnecessarily? 2) Would a reasonable officer in the agent’s position have perceived that the vehicle’s movement presented a threat of serious bodily harm or death? And 3) if so, could the agent have reasonably dealt with the threat by stepping out of the way rather than shooting the driver? For all three questions, we will need detailed information about the timeline and sequence of events that can only come from obtaining and comparing multiple videos and other sources of evidence.”
Editor’s note: FactCheck.org does not accept advertising. We rely on grants and individual donations from people like you. Please consider a donation. Credit card donations may be made through our “Donate” page. If you prefer to give by check, send to: FactCheck.org, Annenberg Public Policy Center, P.O. Box 58100, Philadelphia, PA 19102.