The Black Lives Matter movement was galvanized in 2014 by the death of Michael Brown, a killing we learned of secondhand. But the movement has been largely sustained by further deaths and misconduct caught on film. Without footage from a camera on the dashboard of a police car, for instance, it's unlikely that the Chicago Police Department would be seeking to fire seven officers for trying to cover up the police murder of Laquan McDonald. It's even less likely that the officer who shot him would be now facing first-degree murder charges.

Given how important the existence of video footage has been to the movement, it is perhaps not surprising that among the many demands from anti-police violence activists, requiring police departments to outfit their officers with body-worn cameras has gotten the most legislative traction.

The importance of video footage to Black Lives Matter, of course, is not the only reason that body-worn cameras are such a broadly popular proposal. These cameras play to the liberal instinct to adopt technocratic solutions to problems of injustice. If the problem is racism that emerges from the criminal justice system, then all we need to do is slightly tweak the parameters of that system and the system will fix itself. Although there are no markets and no prices, turning to body-worn cameras is also the neoliberal solution — put in place the right system of incentives and the police will police themselves. Bulk camera purchases don't require a fundamental restructuring of policing, or deep structural reforms. In a very real way, they are a solution that takes the path of least resistance, and the readiness with which police chiefs seem willing to adopt them should give activists pause.

The problem is that body-worn cameras don't film the police. They film from the perspective of the police. Indeed, it is the fact that these cameras capture the police's side of the story that makes them not just tolerable for police officers, but in many cases, desirable. Not only do these cameras face away from officers, but they are under the control of the police. They can be disabled, or blocked when inconvenient. Police are often allowed to view their footage before writing reports, allowing them to shape a narrative that can't be debunked by the evidence. Captured footage is kept by the police, and in many communities, it is up to the police or institutions adjacent to them whether that footage can be seen at all. And if they think that the video doesn't exonerate them, then they often do their best to make sure it isn't.

Even when the camera is operating and footage is recorded, and even when that footage is useable and released to the public, if we've learned anything from the past few years of police brutality caught on camera, it's that visibility is not the same as accountability. The mere fact that a video of a police officer killing or mistreating a civilian exists does not mean that justice will be served, whether the video is filmed by the police or by a civilian bystander. Without district attorneys willing to pursue indictments against officers in departments with which they have to have working relationships, or to file wrongful death civil suits that actually affect department budgets and priorities — in the absence, in other words, of real consequences — cameras won't ensure justice. All they can offer is the illusion of oversight.

There is a small amount of research evidence on body-worn cameras, and while it shows that they may have some real promise in curbing police misconduct, there are also some concerning findings. A 2012 study of a small police force in Rialto, California, randomly assigned body-worn cameras to police officers across shifts over a period of about a year. Researchers found that incidents of use of force were twice as high among those without cameras. They also found that the number of complaints against police officers dropped by 88 percent during the time of the study compared to the year before. Obviously, since complaints are filed to the police department itself, they are likely an underestimate of police misconduct. Even so, the drop is evidence that actual misconduct or allegations of misconduct fall when police wear cameras.

A more extensive study in Phoenix, Arizona, in 2014, found similar drops in the number of use of force incidents and number of complaints against police officers, but it also uncovered two patterns that should be disquieting to activists. While both police officers with and without body-worn cameras increased their number of arrests during the time of the experiment, the increase among those with cameras was triple that among those without. As in Rialto, complaints against officers wearing cameras fell, but in addition, complaints that were filed against those officers were less likely to be sustained. To those skeptical of the idea that there is widespread police misconduct, the interpretation is simple: The police are often falsely accused, and an objective record of the facts is likely to vindicate them. For anti–police violence activists, this exposes police cameras as tools in service of the wider carceral state — instruments to enforce civilian compliance, not to enact police accountability.

It's not clear that whatever positive effects the cameras have on use of force or misconduct more broadly will continue after the police get used to the cameras, either. If changes in behavior are being driven by the fact that police officers are newly conscious that they are being recorded, then if it becomes obvious that there won't be accountability for what happens on camera, cops will slowly revert back to their old patterns. For instance, in San Diego, use of force incidents and civilian complaints had both fallen six months after implementing body-worn cameras, but after a year of implementation, while complaints remained down, use of force incidents increased.

Even more disturbing are the findings of a recent working paper, based on a nationwide sample of police departments, which showed that departments that were using body-worn cameras in 2013 had a slightly higher (and statistically significant) probability of police killings two years later than departments that weren't. To be fair, the study has some difficult obstacles to navigate. For one thing, unlike the studies in Rialto and Phoenix, this study was not a randomized experiment. Since departments didn't randomly decide whether or not to outfit their officers with cameras, this higher probability could merely be a result of the fact that departments that believe they have a problem with police brutality are more likely to adopt cameras in the first place. The study's authors do their best to correct for this possibility statistically, but there are some (dry and technical) reasons to be skeptical that they succeeded. (Another question is how well 2013 data on body-worn cameras captures their implementation in 2015, which is when the data on police killings comes from; some departments have adopted them between then and now, and others could even have dropped them.)

If the result holds, though, then this is something of an inkblot test for views on policing. It could be that the police act more confidently when using deadly force when they have body-worn cameras because they believe that the objective evidence will vindicate their actions. Or it could be another case of the ineffectiveness of body-worn cameras in curbing police violence in an environment in which real, harsh consequences for police misconduct are rare.

For New York City organizer and educator Mariame Kaba, body-worn cameras are a time-sink for activists and a waste of money for communities, and at worst they do damage. “Inserting body-worn cameras in an already corrupt system of policing is harmful. Body-worn cameras mean reinforcing the surveillance apparatus of the state,” Kaba told MTV News. “The resources that are invested — in buying more cameras, paying for systems to retain [footage], etc. ... takes away from other social goods.” In the end, civilians filming the police might be more important to opposing police violence than the police filming civilians. Our cameras aren't under police control, and they can't be turned off when the account they give becomes uncomfortable.