Appeals Court Reverses Earlier Decision, Says Baltimore's Aerial Surveillance Program Is Unconstitutional
from the on-the-other-hand,-this-does-look-constitutionally-sketchy dept
After two court decisions declaring the Baltimore Police Department’s Persistent Surveillance Systems-provided aerial surveillance system wasn’t actually persistent surveillance, the en banc Fourth Circuit Appeals Court has reversed itself, finding that the system capable of capturing the movements of people and vehicles over an area of 32-square-miles violates the Fourth Amendment.
The 192-million megapixel camera system flew over the city for up to 12 hours a day, only shutting down when darkness or weather rendered the cameras useless. This footage was retained by the PD, which could review the footage to trace movements to and from crime scenes. The resolution meant people and vehicles were little more than unidentifiable pixels, but this information — combined with a plethora of ground-based surveillance equipment — made it possible to identify suspects and other persons of interest.
The AIR [Aerial Investigation Research] program was challenged by the ACLU, which cited the Supreme Court’s Carpenter decision in its request for an injunction. The federal court said the surveillance wasn’t persistent and its built-in limitations prevented it from tracking people’s movements for days at a time. As such, AIR’s surveillance was unlike the location tracking in the Carpenter decision, which involved the use of cell site location information obtained from cell service providers.
The Fourth Circuit Appeals Court reached pretty much the same conclusion seven months later. Despite covering a vast area for up to 12 hours a day, the court felt the surveillance was far from persistent. It also seemed to feel the surveillance was a means justified by the ends, citing the Baltimore PD’s inability to solve more than a third of the city’s murders over the past several years.
The court agreed to perform an en banc review of its November decision and has, somewhat surprisingly, declared this form of surveillance a violation of the Fourth Amendment. And it makes no difference that the PD has dumped the program. The ruling [PDF] serves to prevent the PD from trying it again the next time it runs out of better ideas.
Even though the program has ended, the PD has retained some of what was collected during its run. It’s still quite a bit of footage and data.
Rather than store entire days’ worth of data, they elected to retain images from 15 minutes before and after the first and last “track point” for a case, and only within a quarter mile of any track point. They believed this data was the “minimum amount” necessary to support PSS’s reports and “to support the prosecution and the defense teams” in the 200 cases aided by the AIR program, including 150 open investigations. The deletion “result[ed] in a total retained imagery data of 14.2% of the captured imagery data.” Id. In raw numbers, that is 264.82 hours of coverage, comprised of 953,337 cropped images. Id. In addition, “[t]he 200 investigation briefings and other ground-based videos” generated by the AIR program “have already been uploaded to BPD’s Evidence.com.”
The city thought this deletion made everything ok. It was no longer untargeted surveillance of a 32 square mile area but rather targeted surveillance of criminal suspects. That’s the wrong way of looking at this, the court says. The PD wouldn’t have this stockpile of targeted surveillance if it hadn’t engaged in broad surveillance first.
Defendants also stress that BPD has deleted “all but 14.2 percent” of AIR images. But 14.2 percent of all the data collected—millions of photographs documenting thousands of hours of public movements over six months—is a significant quantity of information. Indeed, the preserved 14.2 percent is the needle in the proverbial haystack that the AIR program was designed to discover. Only after recording movements across Baltimore for twelve hours per day could BPD zero in on specific dates and locations related to its investigations and then delete the excess. And BPD still has the briefings and reports, which feature AIR images and tracked movements, information and images from other BPD systems, and insights from PSS’s analysis. Even after the bulk deletion, Plaintiffs have a concrete interest in an injunction barring BPD from accessing what remains, “however small” the interest may be.
Then the court says the lower court failed to comprehend the program’s impact on residents’ privacy.
Plaintiffs argued the AIR program violates Carpenter. The district court rejected the analogy, relying on precedents that approved warrantless pole cameras and flyover photography, and distinguishing CSLI as “a far more intrusive, efficient, and reliable method of tracking a person’s whereabouts.” The district court’s conclusion arose from its read of the facts: “the AIR pilot program has limited location-tracking abilities” because it “will only depict individuals as miniscule dots moving about a city landscape”; the planes “will not fly at night and cannot capture images in inclement weather”; and “gaps in the data will prohibit the tracking of individuals over the course of multiple days.” From that premise, it believed the AIR program could not expose the “privacies of life.”
The district court misapprehended the AIR program’s capabilities.
This is long-term tracking of people’s movements, even with periodic interruptions due to nightfall or inclement weather.
Because the data is retained for 45 days—at least—it is a “detailed, encyclopedic,” record of where everyone came and went within the city during daylight hours over the prior month-and-a-half. Law enforcement can “travel back in time” to observe a target’s movements, forwards and backwards. Without technology, police can attempt to tail suspects, but AIR data is more like “attach[ing] an ankle monitor” to every person in the city. “Whoever the suspect turns out to be,” they have “effectively been tailed” for the prior six weeks. (“[P]olice need not even know in advance whether they want to follow a particular individual, or when.”). Thus, the “retrospective quality of the data” enables police to “retrace a person’s whereabouts,” granting access to otherwise “unknowable” information.
[T]he program enables photographic, retrospective location tracking in multi-hour blocks, often over consecutive days, with a month and a half of daytimes for analysts to work with. That is enough to yield “a wealth of detail,” greater than the sum of the individual trips.
The fact that the surveillance ended at night doesn’t matter. Most surveillance happened during the day and most activities people engage in happened during the time periods the surveillance was occurring
AIR data is a photographic record of movements, surpassing the precision even of GPS data and CSLI, which record variable location points from which movements can be reconstructed. And while the coverage is not 24/7, most people do most of their moving during the daytime, not overnight. Likewise, many people start and end most days at home, following a relatively habitual pattern in between. These habits, analyzed with other available information, will often be enough for law enforcement to deduce the people behind the pixels.
As for the claims the PD never engaged in persistent surveillance with this system, the record says otherwise. The system may deliver unidentifiable pixels in its raw form, but the PD still found it useful enough to perform aerial stakeouts.
The Policing Project reported that one AIR investigation “monitor[ed] the home of a suspect’s mother over the course of two days and track[ed] the individuals who came and went.” Policing Project Br. App. at 16. And an AIR report “detail[ed] a vehicle’s movements over the course of three days, listing eleven locations at which the vehicle stopped, and noting the interactions the driver had with other individuals.”
The court says the surveillance planes can still fly if the city wants to try that approach again. But they’ll need warrants to do it.
The AIR program records the movements of a city. With analysis, it can reveal where individuals come and go over an extended period. Because the AIR program enables police to deduce from the whole of individuals’ movements, we hold that accessing its data is a search, and its warrantless operation violates the Fourth Amendment.
There are several concurrences and a dissent that is criticized heavily by the concurring judges. The most stinging take on the “ends justifies the means” posturing of the dissent is delivered by Chief Judge Roger Gregory:
The dissent faults the majority for making “[n]o mention whatsoever” of Baltimore’s high murder rate. Because the dissent would not enjoin a police surveillance system, it purports to champion “our dispossessed communities” and “the most vulnerable among us.” It suggests the majority, by contrast, “contribute[s] to the continuation” of violence and “leaves only hopelessness” for “the good people of Baltimore.”
This critique depends upon a certain premise: Policing ameliorates violence, and restraining police authority exacerbates it. As surely as water is wet, as where there is smoke there is fire, the dissent takes for granted that policing is the antidote to killing. Thus, the dissent repeatedly evokes the grief and trauma of gun deaths only in the name of a familiar cause: police and prisons. Of course, “it is tempting, if the only tool you have is a hammer, to treat everything as if it were a nail.” Abraham Maslow, The Psychology of Science: A Reconnaissance, 15–16 (1966). But many Baltimoreans know these institutions all too well as the only response to violence.
Just because there’s criminal activity doesn’t mean citizens want fewer rights, especially those in neighborhoods where violent crime is most prevalent. And the Fourth Circuit Appeals Court shouldn’t assume fighting crime is a higher calling than protecting the rights of Americans.
Wherever they call home—from East Baltimore to West Baltimore, from Sandtown to Roland Park, from Cherry Hill to Locust Point— Baltimoreans need not sacrifice their constitutional rights to obtain equal governmental protection.
Even though Baltimore’s aerial surveillance program is dead (possibly just for the moment), its legacy lives on: more precedent saying persistent tracking of people’s movements is unconstitutional.