A woman, stuck in her car in rush hour traffic in Baltimore, saw some police officers beating up a young man. She took out her smartphone, got out of her car, and began to film the event. The police officers noticed her, and came running at her, yelling “You want to film something, bitch? Film this!” The officer threw her phone to the ground, threw the woman to the ground and beat her, and then arrested her for assaulting a police officer and “resisting arrest”.
The charges were dismissed in court. She’s now suing Baltimore police for $1.5 million.
This isn’t an unusual story. Man is bloodied trying to document his own arrest with his smartphone in San Diego. Man is arrested for videotaping police in Hawthorne, California; they handcuffed him and then shot his dog dead. All over the United States, there are stories of people being beaten, harassed, and arrested by police for video recording police activity.
There was a time when people fantasized that it would be enough to simply give technology to the common people, and that would give them the power to expose corruption and fight oppression. The belief was that giving the power of surveillance to the masses would provide a counterbalance to the official authoritative power held by police and military forces. The dream was that these powers would therefore keep each other in check.
It’s not working out that way. Apparently, the power of surveillance is not itself enough to go up against the power of institutional authority and violence. We see it on a small scale in these isolated attacks by police, just as we saw it on a large scale with the failures of the so-called “Arab Spring”.
This is not to say that surveillance is not a powerful tool; but the lesson we’ve been learning is that the power of the camera depends entirely on who is holding the camera. A powerless person training a camera on a powerful person doesn’t necessarily do any good, and can just bring on anger and retaliation. A powerful person training a camera on a less powerful person, on the other hand…
The Geometry of Power
Jeremy Bentham wrote about the Panopticon in 1791. George Orwell wrote about Big Brother and two-way telescreens in 1949. And then there was that 1985 episode of The Twilight Zone with those floating cameras that watched everyone to make sure they didn’t talk to people they weren’t supposed to talk to. The idea that constant surveillance is a means for controlling the population has been a part of our public consciousness for hundreds of years.
Moreover, we know that it works. A 2012 study by the Campbell Collaboration has shown that police presence in high-crime areas will significantly reduce crime incidents. Other studies have shown that speed cameras and CCTV’s both reduce crime. Psychological experiments in academic settings have shown that even subtler manipulations to make people “feel watched”, such as the use of mirrors, windows, or even pictures of eyes, all make people more compliant.
But all of these cases examine situations where the powerful are observing the powerless. Even in the cases where there is not an actual observer present, the driver assumes that a policeman is on the other side of the speed camera, a security guard is on the other side of the CCTV, or the authority figure running the academic experiment is on the other side of the window.
The recent cases of police officers abusing citizens with smartphones show clearly that good behaviour doesn’t materialise merely as a result of being observed. The driver of good behaviour is knowledge that one is being observed by someone with power.
This was a key element of Jeremy Bentham’s original description of the “Panopticon” surveillance structure over 200 years ago. The structure was a circular building with rooms that all had windows facing inward to a courtyard. In the center of the courtyard was a tower for those with authority: from that tower, they could look outward and see what was going on in every single room in the building.
When constructed properly, every person in every single room in the outer ring would always know that there was the possibility that he was being observed, without ever knowing for certain whether he was being observed.cOf course, this building can be interpreted metaphorically: many people today have used the Panoptican as a metaphor for the constant watchful surveillance that people are under as a result of security cameras and the constant monitoring of the internet.
Even without the physical geometry of a centralized authority looking outward to a ring of people, the conceptual geometry is the same: some centralized power is able to observe everyone at once, and nobody knows for sure whether he is being observed or not.
The people who believed that individuals with smart phones could “democratise” this process, and allow every citizen to be a kind of “watcher”, have forgotten a basic element of the Panoptican geometry: the line of vision in the Panoptican is always the powerful observing the powerless, and never the other way around.
Watching the Watchers
Some people have more power than others. Although it might not be possible to keep police officers in line by giving cameras to the rabble, it is possible to watch the police officers.
A recent year-long study has shown that when police officers are required to wear cameras, they are less likely to behave badly. Specifically, over the course of the year, members of the Rialto Police Department were asked to sometimes wear cameras during their shifts, and sometimes not wear the cameras.
Shifts without cameras experienced twice as many incidents of use of force as shifts with cameras. During the entire year when the experiment was being conducted, the total number of citizens’ complaints against the department was nearly one-tenth the number of complaints that were filed the year before.
These cameras, where the eyes on the other end of the surveillance were figures of authority, had a dramatically different effect on police behavior than the camera-phones wielded by mere citizens.
Naturally, there will be follow-up studies and further questions surrounding the impact of body-worn cameras and police behavior. But the results of this initial study are suggestive. Would it make sense to require police officers to wear cameras while they are on duty?
Jay Stanley, a senior policy analyst at the American Civil Liberties Union, has said: ”We don’t like the networks of police-run video cameras that are being set up in an increasing number of cities. We don’t think the Government should be watching over the population en masse.” But requiring police officers to wear video cameras is different, he said: ”When it comes to the citizenry watching the Government, we like that.”
Of course, his quote is not quite a correct description of what is going on with these body-worn cameras: the citizenry isn’t watching the police officers, the police supervisors are. This is a very important distinction, because who is behind the cameras makes all the difference.
Stepping up the privacy debate
The current debate on the issue of surveillance and privacy is stuck in the shallow end of the pool. Conversations about Google Glass are stuck in an endless loop over whether data collection is universally “good” or “bad”, and whether facial recognition is an invasion of privacy, and what will be done with the data once it is collected.
But these conversations almost never acknowledge the complex relationship between surveillance and power. Privacy is almost always treated as something that is inherently “good”, even though we universally reject the idea that a police officer has a “right to privacy” while on duty. It’s time to ask more detailed questions about the boundaries of “privacy” in different power dynamics.
Why does the ACLU think that police officers using cameras to surveil citizens is oppressive, but police supervisors using cameras to surveil police officers is not? The police can certainly abuse the information on citizens out of prejudice or malice, but couldn’t a corrupt police supervisor with a prejudice against an officer do the same thing? Why is the ACLU able to visualize citizens as victims of corruption but not police officers as victims of the same?
If Google Glass becomes more common, perhaps we need to be less concerned with regulating “face recognition algorithms” and the details of where and how the data is stored, and more concerned with social dynamics. Should police officers be required to wear Google Glass, or should they be forbidden from wearing Google Glass?
How does the answer change depending on whether the data is owned by the officers themselves, their supervisors, or a third party private corporation like Google? How do these dynamics shift and change when considering power relationships among different groups of citizens? What about employers and employees? What about teachers and students?
This will be the next generation of privacy dialogue, not trivial questions about whether or not data mining “should be allowed” in Google’s databases.