Jay Stanley, Senior Policy Analyst, ACLU Speech, Privacy, and Technology Project

If there were ever a moment for police body cameras to prove their worth, this month’s protests against police violence should be it. Instead, the protests are revealing the technology’s limitations even as numerous voices call for expanding cameras, including local political leaders from both parties, civil rights groups, public defenders, protesters, and Democratic congressional leaders. Polls, meanwhile, find very high public support for body cameras.
 
But reform advocates should not simply demand hardware. If they want police in their community to wear body cameras, they should demand effective body camera programs, which include not just cameras but also strong policies and institutional practices to make those cameras effective. They should ask themselves whether their department will — or can effectively be compelled to — actually comply with and enforce good policies (preferably enshrined in law) and practices that will ensure that the cameras provide transparency and oversight and don’t become simply surveillance devices.
 
After the shooting of Michael Brown in Ferguson, Missouri and the protests that erupted in its wake, many reformers looked to police body cameras as a critical component of holding police accountable for excessive force abuses. It seemed plausible that the cameras would deter or at least capture evidence of abuses. At a time when the proliferation of bystander videos was increasing awareness of police abuse among many Americans, forcing police to record their own actions seemed like it would improve police behavior and help bring justice when it didn’t.
 
But in the years since, it has become clear that the same impunity from oversight that has given so many officers the feeling that they can break the law — even when being filmed by bystanders — has often rendered body cameras ineffective. While there have been many problems with how body cameras have been implemented (which we attempted to forestall through the body camera policies recommended in our model legislation), the two biggest problems are officers failing to record, and departments refusing to release footage that is recorded.
 
Officers turning off cameras

Body cameras do nothing if they’re not turned on when they should be. There have been too many reports — during the George Floyd protests and before — of police officers failing to record their interactions with the public. The mayor of Chicago cited such failures during the protests, for example, threatening punishment against officers for not recording. There have been many similar reports in New York and other cities. In Louisville, the police chief was fired because officers who killed a man didn’t have their cameras on.
 
This has been a persistent problem since the advent of body cameras. As we have said following prior incidents, police departments need strong, non-discretionary policies that make crystal clear to line officers that they are expected to start filming at the initiation of any law enforcement encounter. And departments need to effectively enforce those policies — something that has clearly been lacking in many departments.
 
We have also called for a presumption against officers in litigation in cases where police should possess body camera footage but don’t (whether because it was never recorded or because it has mysteriously disappeared). Criminal defendants who reasonably claim that exculpatory evidence was not captured when it should have been, or civil plaintiffs suing the government, should have a rebuttable evidentiary presumption in their favor in such cases — in other words, a starting assumption that their claims are correct. Courts should also provide jury instructions that empower juries to discount or disregard police officer testimony when body camera footage should be available but isn’t.

Body cameras raise significant privacy concerns, and there are sensitive questions around when police should turn on their body cameras during protests. We don’t want police to use the cameras to collect video of peaceful protesters, which could chill the freedom people should feel to exercise their First Amendment rights. Indeed, we recommend — and many local policies include — provisions barring the police from recording events such as peaceful protest marches. This has created confusion during events such as the George Floyd protests, which consist of mostly peaceful marches that sometimes include violent police-citizen interactions.
 
But while there may be some gray areas, it really shouldn’t be that hard. If the police are observing peaceful marchers, they don’t need to record. If they decide they need to assert their authority or engage in a law enforcement action of any kind, their cameras should be turned on. Certainly there is zero excuse for police officers failing to record when they are wielding batons or poisonous chemicals against protesters.
 
Departments refusing to make footage public

Another crucial element of an effective body camera program is taking away police department control over the video that body cameras capture. If police departments have discretion over what video footage to release or not release, they (like all bureaucracies) will have a powerful urge to use that power to protect themselves instead of serving the public interest. When police have discretion to release the videos they want, departments often end up releasing the ones that make them look good while withholding the ones that don’t — transforming body cameras from an accountability tool into a propaganda tool.
                            
In Boston, for example, police are refusing to release bodycam footage of recent protests, citing the excuse of “ongoing investigations” — which, as we have explained, in no way justifies withholding footage. In Minnesota, incredibly, footage from the body cameras worn by the officer who killed George Floyd and the officers who were with him has still not been made public. An exception was footage from a Minneapolis Park Police officer who was not directly at the scene, which was only released by that police branch “in an apparent effort to clear the department of being directly involved in the incident.” In a shooting in Louisiana, the authorities showed body camera footage to reporters, but didn’t release it to the public.
 
These patterns of arbitrary, inconsistent, and self-interested releases of footage highlight the problem. Not only should police be releasing body camera footage of critical incidents and those that gave rise to complaints, but it should not be up to them to decide. Where there is a public interest in seeing how police officers are enforcing the law, law enforcement should be required to turn video footage over to the public (with all other footage protected for privacy purposes). That certainly includes video of police enforcement actions during racial justice marches.
 
Some cities have reacted to the pressure from protests by improving their policies, but the public should assume that most police departments will relinquish control over video only very reluctantly. In Washington, D.C., where the police have long stubbornly refused to release bodycam footage even in cases that spark significant public outcry, the D.C. Council passed temporary emergency legislation requiring release of footage, but restricted to cases of “death or serious use of force.” If an officer walks up and clubs a peaceful protester, but doesn’t cause an injury the police define as “serious,” they might refuse to release the bodycam video. Similarly, in New York the mayor recently announced a new policy under which the NYPD will release within 30 days video of officers using force that results in “death or serious physical injury.” That policy, however, also leaves the NYPD with far too much discretion.
 
Beyond the hardware

Some reformers seem too optimistic about the degree to which body camera footage can resolve questions around police use of force. It’s important to remember that like all cameras, bodycams can provide distorted views of incidents and are not an “objective” record of what takes place. Even when all policies are followed, recordings can sometimes be manipulated by officers and provide distorted views of events. They are, however, an independent record that, once captured, does not morph and meld the way human stories and memory can, and is certainly better than solely depending on the word of officers.
 
We’re living in a moment of new possibility for police reform. Those who see body cameras as part of the solution should make sure to push not just for the hardware, but for policies that will make them work for the public.