The Honorable Analisa Torres, United States District Judge, Southern District of New York
April 21, 2017
The Honorable Analisa Torres
United States District Judge
Southern District of New York
500 Pearl Street
New York, NY 10007
Re: Floyd, et al. v. City of New York, et al., 08-CV-1034 (AT)
Ligon, et al. v. City of New York, et al., 12-CV-2274 (AT)
Davis, et al. v. City of New York, et al., 10-CV-699 (AT)
Amicus Letter of The Leadership Conference on Civil and Human Rights et al. in Support of the Floyd Plaintiffs’ Objections to the Substance of the NYPD’s Body Worn Cameras Operations Order
Dear Judge Torres,
On behalf of The Leadership Conference on Civil and Human Rights, Upturn, and the 10 undersigned organizations, we submit this amicus letter to urge the Court to review the New York City Police Department’s (NYPD) proposed Operations Order on “Pilot Program – Use of Body-Worn Cameras,” dated March 22, 2017. We support the arguments made by counsel for the Floyd plaintiffs objecting to the substance of the NYPD’s Body Worn Cameras (BWC) Operations Order.
We are a broad coalition of civil rights and civil liberties organizations, led by The Leadership Conference and Upturn, which brings a unique national perspective on the use of body-worn cameras by law enforcement. In May 2015, The Leadership Conference published a set of shared Civil Rights Principles on Body Worn Cameras (the “Civil Rights Principles”) that was endorsed by 34 major civil rights, media rights and privacy organizations, including the NAACP, National Council of La Raza, National Urban League, ACLU, and many of the undersigned organizations. Importantly, the Civil Rights Principles recognized that “without carefully crafted safeguards in place, there is a real risk that these new devices could become instruments of injustice, rather than tools for accountability.”
Since August 2015, The Leadership Conference and Upturn have been monitoring the ongoing implementation of body-worn cameras by our nation’s law enforcement agencies. We publish and maintain the BWC Policy Scorecard, which tracks the BWC policies adopted by 50 major police departments across the country, and evaluates these policies based on nationwide standards derived from the Civil Rights Principles. The Scorecard highlights promising approaches that some departments are taking and identifies opportunities for departments to protect civil rights through their policies.
We have a strong interest in the NYPD’s planned implementation and use of BWCs. A primary rationale for BWCs is their potential to provide transparency and accountability over community-police interactions — including unconstitutional stops and frisks — and to help protect civil rights, especially in heavily policed communities of color.
But accountability is not automatic. These cameras could hold police more accountable, or simply intensify police surveillance of our communities. Their impact depends on how the cameras and footage are used, and the specific policies that the NYPD puts into place, including specific, articulated consequences for officers who fail to comply with the policy. It is crucial that the NYPD significantly strengthen its civil rights protection for body-worn cameras. This is important not only to residents of New York, but also for setting an example for other departments across the country.
We have deep concerns about the substance of the NYPD’s recently updated BWC Operations Order, and we do not believe the proposed policies will actually “[address] the constitutional harms at issue in this case.” Because the proposed policies heavily favor the interests of the department and its members, and because the NYPD largely ignored the public’s input as these policies were being developed, we also question whether the proposed policies will in fact “alleviate some of the mistrust that has developed between the police and the black and Hispanic communities . . . .”
This Court should review the Operations Order and at a minimum require the NYPD to amend its order to follow nationwide best practices.
A. NYPD’s Operations Order Should Require Recording in Most Circumstances, with Clear and Limited Exceptions
To ensure that all meaningful interactions between officers and the public will be recorded, the NYPD’s policy on “Mandatory Activation of BWC” (¶ 5) needs to minimize officer discretion as much as possible. The proposed policy, however, appears to give officers significant discretion to decide when to turn on their cameras — specifically, by allowing officers to decide whether an interaction is “with persons suspected of criminal activity.” (¶ 5.d.) But many police interactions start with individuals not suspected of criminal activity, which are among the types of interactions that we are concerned about. If officers are allowed this degree of discretion, it is likely that many questionable interactions will not be recorded, which will make it difficult for the Monitor and this Court to measure, evaluate, and address the constitutionality of NYPD’s policing practices.
In addition, the Operations Order requires officers to record “[p]ublic interactions that escalate and become adversarial.” (¶ 5.h.) But asking officers to activate their cameras only after interactions have become adversarial means that cameras will often not capture the vital and decisive moments that gave rise to an adversarial encounter in the first place. To the extent that BWC footage can help a court better assess an officer’s actions and potential motivations and determine whether or not an officer’s behavior was appropriate, this policy likely forecloses such an opportunity, and if anything, creates a substantial new risk that recordings of such already-escalated situations will lack appropriate context.
This is a gaping oversight in the Operations Order. We agree with Floyd and Davis Plaintiffs that “forc[ing] officers to turn on their cameras in the middle of quickly escalating encounters . . . could pose serious safety risks to those officers.” By asking officers to attend to recordkeeping at the very moment when they or others may face greatest danger, this policy not only puts officers at greater risk but also, and importantly, opens a door for officers who were not in fact at risk to claim that they lacked a reasonable opportunity to activate the camera. This will likely mean that many critical situations will not be recorded.
Recent research has also indicated that giving officers discretion regarding when to record may actually increase use-of-force incidents. In a randomized controlled study in eight UK and U.S. departments, RAND researchers found that “[i]f officers turned cameras on and off during their shift then use-of-force increased, whereas if they kept the cameras rolling for their whole shift, use-of-force decreased.” These findings underline the need for this Court and the Monitor to fully assess the implications of NYPD’s proposed policy on mandatory activation.
As it currently stands, the NYPD’s proposed policy leaves officers with far too much discretion to not record both routine and critical incidents. While there are certain types of law enforcement interactions that should happen off-camera (e.g., when attending to victims of sexual assault), the vast majority of interactions with the public should be captured on video. Narrow and clearly defined exceptions should circumscribe this general rule.
Other major departments around the country already recognize the need for simple, clear and wide-reaching rules. For example, the Philadelphia Police Department requires that cameras “shall be activated when responding to all calls for service and during all law enforcement related encounters and activities involving the general public.” Similarly, the Los Angeles Police Department requires officers to “activate their [cameras] prior to initiating any investigative or enforcement activity involving a member of the public,” including officer-initiated, consensual pedestrian encounters. The Court should require the NYPD to adopt a similar policy that both broadens and simplifies its mandatory activation requirements.
B. NYPD’s Operations Order Should Require Officers to Write Their Initial Reports Prior to Viewing Footage.
The Court ordered the NYPD BWC pilot as a remedy partly in the hope that “the recordings will diminish the sense on the part of those who file complaints that it is their word against the police, and that the authorities are more likely to believe the police.”
But the NYPD’s proposed policy will likely do the exact opposite, further undermining complainants and exacerbating structural advantages that officers already have. The policy creates an unfair and uneven playing field by always allowing officers to view footage before writing their initial reports (¶ 17), and never providing complainants with a guarantee of footage access. A policy that affords officers an automatic privilege to review video evidence before providing their own statements, while withholding that same right from New York City residents, is at odds with the purpose of this Court’s prior orders, which sought to level the playing field between officers and complainants.
If the Court allows this policy to stand, officer reports will counterproductively and systematically seem more accurate than they really are, because officers will always have the opportunity to comport their reports to a version of events that videos appear to show, rather than what they actually remember. The result will be that officer reports will often appear fully consistent with the footage. In contrast, those filing misconduct complaints (and other witnesses) won’t have this advantage. Their statements will likely contain natural inaccuracies because they — like everyone else, including officers — don’t have perfect memories. But these natural inaccuracies will make civilian statements appear less consistent, and thus less credible, when compared to officer statements.
The proposed policy will not only fail to accomplish the goal of putting complainants on a more even footing with officers. It will make the situation decidedly worse. And this will be a problem in both administrative and court settings.
The Court should require that the NYPD instead establish a more even-handed policy, like the “two-step” viewing process adopted by the Oakland Police Department in some situations. In their policy, before watching any footage, officers are first required to submit a preliminary statement. Only then can they watch relevant footage and have the opportunity to add supplemental information to their preliminary statement.
While our concerns about when officers can view footage are particularly acute for officer-involved shootings and other high-level uses of force where an officer may be the subject of an investigation, the two-step viewing process should be required after all incidents. It’s not always evident at the time of an incident that an officer’s behavior will be subject to an internal investigation. For example, if an officer pushes somebody to the ground or uses racial slurs, that behavior might not come to the attention of the department until that person files a complaint, which may not happen for days or weeks. The only way to preserve the independent evidentiary value of officer reports in those situations is to always enforce the two-step viewing process.
Indeed, the New York Office of the Inspector General for the NYPD (OIG-NYPD) also recently recommended that the department adopt the two-step process for all incidents:
. . . in all instances, officers should be permitted to review video after an initial report and an official statement have been made, and file supplemental reports in order to document details of an incident recalled through footage review and maintain a complete and accurate record of events surrounding an incident. In this regard, consideration of any mitigating factors leading to differences between BWC footage and officer testimony should be considered before officers are disciplined for these discrepancies, and officers should not be sanctioned without clear evidence they intentionally and materially obscured the truth.
The NYPD makes the blanket argument that the two-step process is always infeasible because there are a limited set of use-of-force circumstances where there may be both a criminal and administrative proceeding, which would complicate the footage access protocol. But there are many other important situations where a confrontation or problematic incident leads to only an administrative review, or no internal investigation at all. Under the NYPD’s proposed policy, officers will be in a position to tailor their reports and statements to the video footage in these cases. This major loophole in NYPD’s policy risks creating a situation where widespread patterns of officer misconduct could go unremedied.
C. NYPD’s Operations Order Should Create a Simple, Streamlined Process for Recorded Individuals to View Relevant Footage.
If cameras are to provide any transparency value at all, the NYPD should guarantee — at a minimum — that subjects of recordings, like those looking to file police misconduct complaints, will have access to footage of their own incidents. During the NYPD’s public comment process last year, the vast majority of the public (78 percent) either “strongly agreed” or “agreed” that the department should be required to show footage to subjects of recordings upon request. This is clearly a widely held public expectation.
Currently, however, NYPD provides access to footage only through the Freedom of Information Law (FOIL) public records process. Yet NYPD has been known to obstruct FOIL requests for BWC footage, by claiming broad confidentiality exemptions (“criminal investigation” and “non-discloseable intra- or inter-agency communication”) as well as by charging requestors prohibitively expensive fees.
In its recent public report, the NYPD claimed that the FOIL process is “far superior to the live-streaming of NYPD policing online,” but this statement is nothing more than a straw-man argument — streaming is far from the only alternative to provide meaningful footage access to individuals outside the department.
Leading major departments, like those in Las Vegas and in Washington, DC, have implemented simple, streamlined processes that provide special access to recorded subjects. These departments’ processes differ slightly, but in general, recorded subjects can fill out a simple form to schedule a time to view (without necessarily obtaining copies of) relevant footage, potentially together with a legal representative.
Given that the purpose of NYPD’s BWC pilot is to help complainants and reduce mistrust, especially after potentially unconstitutional practices by officers, the Court should demand that the NYPD implement a similar, streamlined process for recorded subjects to access footage. Providing special access to footage subjects is critical to leveling the playing field between officers and affected individuals.
D. NYPD’s Operations Order Should Sharply Limit the Use of Face Recognition Technologies with BWCs.
Leading body-worn camera vendors — like Axon and VIEVU (which was recently awarded NYPD’s camera contract) — are rapidly working to incorporate face recognition and other artificial intelligence (AI) capabilities into their camera systems. Earlier this year, Axon acquired an AI company called Dextro. And just this month, VIEVU entered into a partnership with Veritone, an AI company that could soon bring face recognition capabilities to NYPD’s cameras.
When body cameras gain these powerful capabilities, it will fundamentally change the nature and purpose of BWCs and create enormous new risks to the residents of New York. In combination, cameras would become invasive, constitutionally questionable tools for mass surveillance, scanning the faces of every (innocent) person officers encounter, and looking those faces up in real-time in police and other intelligence databases. A recent study from Georgetown University Law Center found that half of all American adults are in law enforcement face recognition databases, with little to no regulatory constraints on their use.
When used together with body cameras, face recognition would give officers powerful tools for enhanced police surveillance — especially in communities of color where cameras will be pervasive — rather than tools for accountability. This dangerous combination could also amplify existing disparities in law enforcement practices and undermine progress toward building public trust.
We are concerned that NYPD’s proposed policy does not place any limits on the use of such biometric technologies, which could open up a whole new realm of constitutional challenges to NYPD’s policing practices. During the public comment process, some public commenters urged the NYPD to place limits on the use of biometrics, but the department brushed off these concerns by saying that “the cameras that the NYPD will use in our upcoming pilot will not have capabilities that significant differ from the human eye . . . .” But these capabilities are coming sooner than many may think.
Several major departments, including those in Baltimore and Boston, have proactively made public commitments in their BWC policies that strictly limit their own use of face recognition and other biometric enhancements, together with their BWC systems. If the NYPD indeed does not plan to use these technologies together with their BWCs in the future, then it should be simple for them to make a similar public commitment. The Court should urge the NYPD to do so in its written policy.
In conclusion, the undersigned organizations respectfully urge this Court to review the substance of NYPD’s proposed Operations Order and direct the NYPD to amend the order in line with the national best practices and critical civil rights protections described in this letter.
The Leadership Conference on Civil and Human Rights
18 Million Rising
Center for Democracy and Technology
Center for Media Justice
Color Of Change
The Constitution Project
Electronic Frontier Foundation
National Hispanic Media Coalition
New America’s Open Technology Institute