SCOWTing out surveillance cameras

in Manhattan

I've been giving these SCOWTs -- Surveillance Camera Out-door Walking Tours -- every Sunday afternoon since Thanksgiving 2000. I originally got the idea from a guy named Constant, who for several years has been giving walking tours of surveillance cameras that he's located in Brussels, Belgium. I learned about him through his website, which not only features maps but also close-up pictures of the cameras that he's found.

I started making maps of camera locations in Manhattan and giving walking tours based upon them in part because the group that I've been involved with since November 1996 -- the New York Surveillance Camera Players -- can't really perform in the winter months. It's almost impossible for us to display our "trademark" hand-printed placards for the benefit of the cameras when it's bitterly cold, raining hard or blustery. Unfortunately, the surveillance cameras don't care if its hot or cold, wet or dry: with proper maintenance, they can function 24 hours a day, 365 days a year. By giving the walking tours, I found that I could keep the anti-camera struggle going, even during the winter months.

Keeping the struggle going: this has been very important to me these last few years, during which new and increasingly sophisticated surveillance cameras have been installed steadily in public places. Unfortunately, the public isn't being kept abreast of these developments, which clearly will have a tremendous and possibly irreversible impact on everyone's lives. Here in New York City, the public isn't informed of the location or technological capabilities of the NYPD's cameras. When the cameras are installed, they are installed in secret and rarely bear signs that inform people of their existence. (The only exception is the Lillian Wald Housing Bloc, at which a great many signs were posted in 2001, almost three years after the cameras themselves were installed.) There is no public discussion of their constitutionality or cost effectiveness, and it doesn't appear that anyone in either the Mayor's office or NYPD Headquarters is keeping track of the arrests and criminal convictions that are attributable to them. (Note as well that none of the local newspapers, not even The New York Post, carries stories about the effectiveness of the NYPD's surveillance system. You'd think that such stories would be a daily occurrence in Mayor Giuliani's "zero tolerance," get-tough-on-crime New York.) Unlike the cameras that one finds in England, the majority of those encountered in New York City don't "look like" surveillance cameras. Instead, they look like bulb-like ornaments or street lights, but certainly not swiveling and rotating mini-cameras! As a result, they manage to be "hidden in plain sight." And so, the primary purpose of these walking tours has been to teach people where to look for, how to recognize, and thus how to count the number of surveillance cameras operating in public.

Back in November 1998, a comprehensive count of such cameras was undertaken by a crew of volunteers assembled by the New York City chapter of the American Civil Liberties Union. Counting only those cameras that were installed on poles, posts, the exteriors of public and private buildings, and other public places, the crew counted a total of 2,397 surveillance cameras in Manhattan alone. (The other boroughs of New York City weren't canvassed. The NYCLU also excluded from the count those cameras in operation inside private establishments because of their owners' legal right to monitor their own premises.) Today, the old NYCLU map forms the basis for a constantly updated, computer-generated map of surveillance cameras maintained by Mark Ghuneim, a member of the original NYCLU crew who displays the map on his website. (Were it not for the fact that Ghuneim's map can't be printed out easily, I might never have made my own and would have relied on his, instead.)

Convinced that there were even more cameras in operation in 2000 than in 1998, I set about to estimate the increase by using several neighborhoods in Manhattan as test cases. Between August 2000 and August 2001, I mapped out camera locations in a total of seven neighborhoods: Times Square, the United Nations, City Hall, the Lower East Side, Harlem, Washington Square Park, and Greenwich Village. (Each map took about 6 hours of "leg work" to complete.) An interesting adjustment in my perceptual habits was necessary: instead of looking down at the ground so that I could walk fast without stepping on someone or in something, which is what New Yorkers such as myself are in the habit of doing, I had to slow down and look up -- up at the facades, corners and tops of the buildings, because these are the places in which the cameras are usually located -- which is what tourists do when they walk around New York. And so, making the maps allowed me to see "my" city with new eyes, and so I of course saw places and things (other than surveillance cameras!) that I'd never seen before.

While making the maps, I also discovered or, rather, I re-confirmed the NYCLU's discovery that, contrary to the expectations created by the proponents of public surveillance, there are relatively few cameras in poor neighborhoods and a great many in rich ones. (If one believes that surveillance cameras are installed to prevent "crime" or "criminal activity," then one would expect that they are mostly installed in high crime areas, which are commonly thought to be the poor neighborhoods, not the rich ones.) The maps make clear that surveillance cameras are not intended to and in fact do not protect all citizens from being victims of "crime" or "criminal activity." Rather, the cameras only protect a select few (those who have property worth protecting) from a single and very specific form of crime: crimes against property. Because property can be insured, videotapes of the crime are doubly useful: not only as criminal evidence, but also as the basis for financial compensation from an insurance company.

Before we go on, let's note that there's a couple of small but important differences between the map made by the NYCLU and the ones that I've made. The NYCLU map (Ghuneim's, too) categorizes the cameras according to their mobility -- are they stationary, rotational or 360-degree mini-cameras enclosed within globes? -- because their mobility is directly related to their invasiveness. Though they can't move, stationary cameras are much more invasive than the eyes of a police officer. Unlike human eyes, camera lenses "see" continuously and never need to blink, rest, cry, wear glasses when they get old or protect themselves from driving winds or bitter cold. Thus, all surveillance cameras (stationary, rotational or global: it doesn't matter) destroy the privacy normally and traditionally afforded to public behavior by the literal humanity of the police, "bad" weather conditions and the inevitable passage of time. Without exception, surveillance cameras also confuse or replace civilian surveillance with military surveillance: instead of gathering small amounts of high-quality information that can be introduced as evidence at a trial, they collect huge amounts of low-quality information that must be sorted through and analyzed before it can be put to any use at all. And so, not only do surveillance cameras make for bad law enforcement, they also jeopardize the separation between the military and the police, which is one of the pillars of American democracy.

Rotational cameras are even more invasive that stationary cameras because they can turn and follow people as they move. If there is a series of rotational cameras in place, all of them connected to a centralized control room, then it's possible for a single police officer to track a person's movements wherever he or she goes, as long as that person remains within sight of at least one of the cameras. In England, where there's a surveillance camera for every 50 people, someone can enter a car in a town outside of London, drive that car all the way into the city, get out of it and go to work, with the police watching and never losing sight the whole time. We shouldn't fail to mention that the very sight of a surveillance camera turning around to focus upon someone in particular is horrifying. A rotational surveillance camera, especially if it is perched high on a building or atop a tall pole, looks less like a human invention than some kind of gigantic insect (perhaps a praying mantis). Quite obviously, this aspect makes rotational cameras even more chilling than their predecessors. One wonders if New Yorkers would be less willing to accept the presence of surveillance cameras if they could see the fucking things moving around. . . .

But by far the most invasive devices are the cameras enclosed within opaque plastic globes, which, as we've already had cause to note, do not even look like surveillance cameras. According to the on-line advertisements for these cameras, as well as a couple of the police officers I've spoken to, the type being used by the NYPD is able to zoom in so close that the watcher can count the change in your hand as you stand at the bus stop or "pull back" and see as far away as a mile, without losing any image-quality. Unlike the glances of a police officer, which are subject to a certain amount of visual accountability, the viewing habits of the person operating a globe camera -- e.g., the people, body parts and behaviors that he or she tends to watch -- cannot be discerned by an on-looker. To make matters worse, the globe-camera operator might be looking anywhere at any time. And so that camera operator -- who may or may not have had training in the appropriate (lawful as well as ethical) use of the very powerful technology with which he or she has been entrusted -- is left free to engage in racial profiling, sexual voyeurism, the settling of personal vendettas, etc. etc., with total impunity.

Though no less concerned with invasiveness than the NYCLU, I've categorized the cameras I've spotted and mapped according to their (presumed) ownership, which I've tried to deduce from their precise locations, as follows: police cameras are mounted on city-owned buildings, poles and posts; and private cameras are mounted upon the sides or tops of privately owned buildings or atop poles and posts on private property. Within the category of "private cameras," I've distinguished three different types: cameras that are installed on commercial properties by the businesses that own or rent them; cameras that are installed on residential buildings by the owner(s), the landlord and/or one or several of the tenants in the buildings; and "web" cameras, which are installed on either public or private buildings for purposes of providing live video "entertainment" via the Internet. (To the security industry, it should be noted, webcams aren't silly toys, but "open circuit" broadcast devices that facilitate long-distance surveillance.) It can be difficult to tell which exact kind of privately owned surveillance camera is in operation: in Times Square, for example, there are several stationary cameras that are used for live television broadcasts by NBC and ABC, but look like surveillance cameras operated by a private security firm.

To organize all this information, and to add in a little history, I refer to three "generations" of surveillance cameras on my walking tours. "First generation" surveillance cameras were developed in the late 1930s by both the American and German governments. Unlike live television cameras, which use an open circuit to broadcast the images they capture, first generation cameras are "closed-circuit" devices, which means only those who have physical access to the circuit can see what is being broadcast on it. (Open circuit broadcasting means that anyone with certain information -- the frequency and time of the broadcast -- can "tune in" to it, even if they don't have physical access to the system.) You'll hear Brits speak of CCTV (closed circuit television) instead of surveillance cameras. Except when it comes to wireless surveillance cameras, which are "OCTV" (open circuit television), the two terms are virtually interchangeable, though I find that the acronym "CCTV" doesn't refer to the reasons that the cameras are installed and so encourages people to ignore or forget them.

It was in the late 1950s that such devices were first used in public places for the purposes of law enforcement (typically in remote locations and at night and in conjunction with flood lights). In the early 1960s, federal law mandated the use of video surveillance in banks. At the end of the 1960s and well into the 1970s, there was a vogue for installing cameras in "dangerous neighborhoods," where valuable properties (mostly automobiles) were at risk of being damaged or stolen. Cameras went up in Times Square, Mt. Vernon, New York, and Newark, New Jersey. But it quickly became apparent that these cameras -- all of which were stationary devices -- were very costly to keep in working order, were broken or inactive on too many occasions, required a whole team of round-the-clock viewers to make the effort worth the trouble and expense, and netted too few arrests and convictions to be deemed "cost effective." And so they were taken down. By 1985, when the City of Detroit removed the camera system that it had installed in 1980 on the occasion of the Republican National Convention, the vogue for public surveillance cameras among American police forces appeared to be over. (Video surveillance continued to be used by private companies, which needed to surveill relatively small areas of space -- their own premises -- and could benefit from reduced insurance rates if they installed camera systems.)

"Second generation" cameras were developed in the early 1980s. Rotational (and later, in the 1990s, globe-shaped) devices, they were much more efficient than their predecessors because they could track people's movements as well as make it increasingly difficult for people to locate, hide in and commit criminal acts because of "blind spots" in the surveillant's field(s) of vision. Because one rotational camera could do the work of several stationary ones, and because one globe-shaped camera could do the work of several rotational ones, police departments were able to cut down on the numbers of cameras they were using, which in turn allowed them to cut down on the numbers of video cables and camera-watchers they were using. Second generation cameras really began to proliferate in the early 1990s, at first as a result of federal funds made available as part of the "war on drugs," which targeted youths who congregated in urban centers, and later as a result of federal measures taken in the late 1990s to prevent terrorist attacks, such as those that took place at the World Trade Center in 1993, at the entrance to CIA headquarters, where in 1994 several employees were shot and killed as they waited in their cars to get into the compound, and at the Murrah Building in Oklahoma City in 1995. (The recently enacted PATRIOT Anti-Terrorist Act of 2001 builds upon these precedents.)

In order to make sure that this second wave of surveillance cameras was deemed "cost effective," police departments and Mayor's offices all over the country in the 1990s simply stopped keeping accurate records of the numbers of arrests and convictions that could be attributed to the presence of a surveillance camera. And so, there's no way to either prove or disprove the contention that surveillance cameras are effective crime-fighting tools. (In 1998, when a Californian legal scholar named Nieto canvassed the police departments that employed public surveillance cameras and asked them to share with him their statistics, he found that none of them were keeping records and that they were all relying instead on anecdotal evidence.)

"Third generation" cameras are first or second generation cameras that have been attached to and enhanced by a computer that is housed on the watcher's side of the technology, and so is completely invisible to the person being watched. Unlike their predecessors, third generation cameras are active devices. Their attention to the faces and behaviors of particular people doesn't wait for and isn't triggered by the suspicions, "hunches," prejudices or predilections of the person(s) who are watching. Instead of arbitrarily deciding that some people are "OK," while others are "suspicious," third generation cameras take pictures of each and every person they see and then, assisted by an automatically triggered computer program, compares these pictures to images that are already stored in a database. Two questions are asked and answered in a very short period of time: 1) is this person in the database? and 2) is this person in the database because he or she is a privileged customer or a dangerous criminal? Then, depending on how it was programmed, the computer can decide to ignore you, welcome you or call the police on you.

Third generation cameras can also be "enhanced" by the use of gesture recognition software, which allows surveillance cameras to recognize inappropriate movements, suspicious gestures and other "strange behaviors." That is to say, third generation cameras can be connected to systems that have been programmed to think like a police officer (make judgments based upon interpretations). And so the replacement of human beings by machines continues: first, human eyes were replaced by camera lenses; now, human brains are being replaced by computers. What's next, a society without people? There would certainly be no more crime . . . .

When I began giving these walking tours, third generation cameras weren't in use in public places in the United States, though the major manufacturers of the software (Visionics and Viisage) are based in New Jersey and Massachusetts, respectively. But over the course of the last year, face recognition software has been installed on public surveillance cameras in several American cities, including the Ybor City area in Tampa Bay, Florida, and Virgina Beach, Virginia, and in a growing number of airports, including those in Oakland, Boston and Providence. I consider these developments to be the most alarming I've seen in the five years I've been active in the anti-surveillance movement, because racial profiling -- unequal protection under the law -- is so clearly involved in each and every case: Ybor City is a Cubano ghetto; Virginia Beach is visited by thousands of black sorority and fraternity members every summer; all of the airports are on the alert for Middle Eastern or Central Asian men, and no one else.

Unlike the NYCLU's map, the maps I made in 2000 and 2001 record the position of microwave relay transmitters (MRTs) installed in secrecy by the NYPD. Strictly speaking, MRTs aren't surveillance cameras; but since they allow mobile police units (on foot-patrol, in squad cars or on-board helicopters) to use microwaves to transmit and receive live video images, MRTs form part of the surveillance system. (Wireless video surveillance devices such as MRTs are very important in England, where the authorities don't want to expose their cables to damage caused by the elements, wild birds and animals, or human saboteurs, and don't want to tear up beautiful old streets simply to install their video cables in protected underground tubes.) In New York, MRTs can be found atop hundreds of public and private buildings, despite the fact that many of these devices have been installed in violation of Federal Communication Commission regulations concerning the placement of microwave antennae on rooftops. (The NYPD have gotten around federal law thanks to Mayor Guiliani, whose appointee at the Department of Buildings issued a 1997 directive that -- in the name of stimulating investment in New York City by the big cell phone companies -- effectively nullified the provisions of the City's housing code that mandated the approval of all antennae by the Board of Standards and Appeals.)

Significantly, excluding the city's motorways, highways and bridges, the highest concentrations of MRTs are found in Harlem and the Lower East Side. In addition to being poor neighborhoods that have comparatively few surveillance cameras, these areas are often the scenes of various kinds of mass gatherings (parties, celebrations, festivals, pickets, rallies and marches), all of which are in the eyes of the police "unpredictable" situations that require live monitoring for "signs of trouble." And so, instead of maintaining and protecting their vulnerable video cables (unless they are protected, the cameras are vulnerable, too!), the NYPD relies upon mobile cameras, microwave transmitters and MRTs. It appears that the police are doing this without regard for the potential risks that these devices pose over the long-term to the health of both the police officers who operate them and the private citizens who live or work within the NYPD's concentrated beams of microwave radiation. (These health risks include cataracts, sterility and leukemia.) And so, in the name of protecting public safety, the NYPD is jeopardizing the public's health.

Using the same criteria the NYCLU used (i.e., only count the visible cameras that are installed in public places), I found that, in Times Square, for example, there were 129 cameras, while the NYCLU had counted 75 of them. If this increase was representative of the increase in cameras throughout Manhattan as a whole -- the other maps I made suggested that it was -- then, two years after the NYCLU made its count, there were more than 3,800 public surveillance cameras. There are certainly even more in use today. I haven't re-mapped Times Square or any of the other areas in which I looked for and marked down the location of cameras, but, in the course of giving these walking tours, I've spotted lots of new ones. As a result of the federal legislation passed in the aftermath of the disasters at the Pentagon and the World Trade Center on 11 September 2001, it is certain that there will be even more cameras installed in public places, despite the fact that cameras did absolutely nothing to deter, prevent or lead to the capture of the perpetrators of either one of the attacks on the WTC. There were in fact so many surveillance cameras in use at the WTC prior to the disaster that I wasn't able to map their locations.

At this writing (November 2001), a liberal estimate would be that there are 10,000 surveillance cameras in operation in Manhattan. Such an estimate would be large enough to take into account those cameras that are either too small or installed too high to be seen by the "naked eye" of an amateur such as myself. While 10,000 cameras might seem like an awful lot to a New Yorker (that's an average of six cameras per square block), it's nothing compared to London, where 10,000 cameras are in operation in a single neighborhood ("The City," London's financial district). There are more than 1.5 million cameras in operation in England as a whole. If New York and the rest of America becomes like London and the rest of England, which unfortunately seems likely in the aftermath of the 11 September disaster, one can expect to see hundreds of thousands of surveillance cameras in operation in Manhattan in the next few years. If ex-NYPD Chief and current member of New York State's Office of Homeland Security Howard Safir gets his way, these new cameras will be equipped with face recognition software.

Each walking tour I give meets and begins directly beneath a surveillance camera operated by the NYPD. This is done for several reasons, the most important of which is the fact that, unlike the cameras in England or in other American cities such as Baltimore, very few of the police surveillance cameras in New York are properly labeled. None of them bear signs that warn the passers-by that they are under constant video surveillance by the police. This is very peculiar. If it's true that the cameras have been installed to prevent criminal activity before it takes place, rather than simply document it as it takes place, then you'd think that big bright warning signs would be the norm, because a criminal is more likely to see and understand the meaning of such a sign than he or she is to see and understand the meaning of a small, globe-shaped object on top of a pole somewhere. The fact that the NYPD's cameras aren't labeled suggests that they weren't in fact installed to prevent criminal activity, but as part of an experiment in the social control of law-abiding citizens, which to be successful requires that the test subjects be unaware of the fact that they are part of an experiment.

Another reason each tour begins under an NYPD surveillance camera is the fact that, unlike private camera operators, the police are required to obey the Fourth Amendment to the United States Constitution and all the federal and state laws that have been built upon it. For both dramatic and informational purposes, each tour features a public reading of the entirety of the Fourth Amendment, with the text held so that the surveillance camera under which the reading takes place can zoom in and see it.

"The right of the people to be secure in their persons, houses, papers, and effects, against unreasonable searches and seizures, shall not be violated, and no warrants shall issue but upon probable cause, supported by oath or affirmation, and particularly describing the place to be searched, and the persons or things to be seized."

The unregulated use of police cameras in public places systematically violates every single provision contained in this elegantly worded but severely over-burdened sentence. None of the operators of the NYPD's untold numbers of surveillance cameras are getting warrants that are based upon clearly established probable cause that a crime is being committed or is going to be committed at the precise locations in which the police cameras are installed. Nor are the police particularly describing the type of visual information to be searched (by the camera) and seized (by the videotape recorder and/or the eyes of the camera operator). In other words, each and every person observed or photographed by each and every one of the NYPD's cameras has been the victim of an unreasonable search and seizure.

Before concluding, I'd like to note that the Fourth Amendment protects the "right of the people to be secure," not the right of the people to be anonymous or to have their privacy respected. But privacy is cleared implied by the word "security." One doesn't feel secure if one has no privacy. And so there can be no opposition or choice between "privacy" and "security." The Fourth Amendment was appended so that, unlike the colonialists, the people of the United States would never be the victims of "writs of assistance," which the British had used to justify searches and seizures of people going about their business in public whenever such invasions of privacy suited the authorities. In any event, several state constitutions -- including that of Hawaii, which updated its equivalent of the Fourth Amendment in 1978 -- explicitly refer to "the right to privacy." We should note as well that the Fourth Amendment mandates that citizens have the right to privacy in both public and private places, not just in the latter. Note well the phrasing: "The right of the people to be secure in their persons, houses, papers, and effects." The word "persons" precedes "houses." That is to say, one doesn't simply have a reasonable expectation of privacy when one is inside one's house; this expectation is valid when one leaves one's house and takes one's person out into public. The expectation of privacy continues until one enters someone else's privately owned establishment, in which case the reasonable expectation of privacy is temporarily suspended.

This interpretation of the Fourth Amendment was affirmed by the Supreme Court in its landmark 1967 decision, Katz v. the United States. In this case, a man suspected of dealing illegal drugs was audiotaped as he made a telephone call in a public phone booth and was convicted on the basis of the tape recording. The Supreme Court threw out the conviction because, under the Fourth Amendment, the man had a reasonable expectation of privacy when he used the telephone, even though it was installed in a public place. Unfortunately for us, the Supreme Court has never ruled on an appeal of a conviction based upon a videotape that was recorded in a public place. Even more unfortunately, the federal laws that cover audio surveillance (wiretapping and eavesdropping) haven't been updated to include video surveillance as well, even though there is a strong analogy between the two forms. As so, there's a giantic hole in the law. This omission or oversight must be corrected immediately, because it is the only reason that the systematic video surveillance of public places by the police and private security firms has spread and been tolerated in a putatively democratic society such as ours.

Surveillance cameras are never installed permanently. In New York City, they were taken down once. They can be taken down again.

Alright, people -- let's get walking.

Bill, the tour guide

27 November 2001

Contact the NY Surveillance Camera Players

By e-mail

By snail mail: SCP c/o NOT BORED! POB 1115, Stuyvesant Station, New York City 10009-9998

NY Surveillance Camera Players