Speeches

Watching You: Privacy Rights and Video Surveillance

McMaster University, Communications Studies Programme

February 13, 2002
Hamilton, Ontario

George Radwanski
Privacy Commissioner of Canada

(Check Against Delivery)


I'm here to talk to you about what I regard as the most urgently important privacy issue facing Canadian society today: video surveillance of public places.

I'm here because it's reported that the Hamilton police department wants to install six video surveillance cameras in downtown public areas from James Street to Walnut Street.

As Privacy Commissioner of Canada I don't have jurisdiction over the Hamilton police force. But as the Officer of Parliament mandated to oversee and defend the privacy rights of all Canadians, I not only have the jurisdiction but the duty to warn if something is about to destroy those privacy rights.

And that, I hope to persuade you today, is what installing police surveillance cameras in our public streets will do.

I want to persuade the people of Hamilton to think very carefully about this proposal by your police force-and to consider that trading away the fundamental human right of privacy for an illusion of greater safety is not the way to go.

The choices you make will quite literally help determine what kind of society-how free a society-we have in Canada, not only for ourselves, but also for our children and grandchildren.

The cameras you're contemplating here in Hamilton-and those that will follow in all the other cities that will imitate you if you go ahead with this-are the thin edge of the wedge that will irrevocably change our whole notion of our rights and freedoms.

What I'm talking about is what Justice LaForest of the Supreme Court of Canada said in a 1990 decision:

"To permit unrestricted video surveillance by agents of the state would seriously diminish the degree of privacy we can reasonably expect to enjoy in a free society . . . We must always be alert to the fact that modern methods of electronic surveillance have the potential, if uncontrolled, to annihilate privacy."

After the tragic events of September 11, I stated publicly that privacy is not an absolute right. There clearly can be circumstances where it is legitimate and necessary to sacrifice some elements of privacy in the interests of vital security precautions. But I suggested that the burden of proof must always be on those who say such a sacrifice is necessary.

I recommended that any proposed measure to limit or infringe privacy must meet four very specific criteria. First, it has to be demonstrably necessary to address a specific problem.

Second, it must be demonstrably likely to be effective in addressing that problem. Third, it must be proportional to the security benefit to be derived. In other words, you don't use a sledgehammer to kill a fly. And, finally it must be demonstrable that no less privacy invasive measure would suffice to achieve the same result.

I believe that these tests must be applied to any proposal to install the surveillance cameras of the state on any of our city streets. And I believe that the proposal-certainly here in Hamilton-would fail every one of those tests.

I'll come back to this in a moment, but first let me talk a little about privacy itself. Privacy-your right to control access to yourself and to information about yourself-is a fundamental human right, recognized as such by the United Nations.

It is, as our Supreme Court has said, at the heart of liberty in a modern state. In fact, many believe it's the right from which our other freedoms flow-freedom of speech, freedom of thought, freedom of association, just about any freedom you can name.

But privacy is more than a human right-it's also an innate human need.

When you go home at night, you probably close the blinds. It's not because you're trying to hide something. You just instinctively need your privacy, your freedom from being observed. If you're on a bus or a plane, and someone starts reading over your shoulder, you probably feel uncomfortable. What you're reading isn't a secret. It's just that your privacy is being invaded.

If you've ever had your home or even your car broken into, you'll know that the sense of personal violation can be even worse than the actual loss of whatever was stolen. This essential human need, and fundamental right, is in danger of slipping away from us. And once it's lost, it will be very, very hard to get it back.

We're living in a time when technological, social, and political developments threaten our privacy at every turn. That's why I believe that privacy will be the defining issue of this decade. We are at a crossroads: If the choices we make allow privacy to be destroyed, freedom will be destroyed with it.

And if privacy is the defining issue of the decade, I believe that video surveillance is the defining issue within that broader issue.

I know that some people say, "What's the harm? So there's going to be a few cameras - big deal." In fact, in Kelowna, B.C., one of the first cities where I confronted this issue, a business leader was quoted as saying that having cameras everywhere would be no different that having a police officer on every corner and nobody could object to that.

Well, there are places in the world where there's a police officer on every corner. They're called police states. That's not the way we do things in Canada.

We have a right to go about our lives-our lawful, peaceable business-without feeling that we're under the constant eye of agents of the state. So when you ask, "What's wrong with cameras?" My answer is that, for one thing, being watched changes the way we behave.

That's well known. In the late 1920s, Werner Heisenberg showed that you couldn't observe subatomic particles without changing their behaviour.

Ever since, physical and behavioural scientists have used the term "Heisenberg effect" to describe what they're all aware of-that being observed has real effects. The psychological impact of feeling under constant observation-the kind of thing we'll have if we allow surveillance cameras to proliferate-is enormous, incalculable.

Let's think about those police states I was talking about a moment ago, where people always feel watched by a visible or undercover police presence.

The thing that has always struck me when I was travelling in countries that were not free, in police states, is the grimness and drabness of life that results from this utter lack of privacy. There's a kind of sullenness in the air.

People know they're being watched-or worse, they're never quite sure whether they're being watched. They censor their speech and their behaviour. They hurry along the streets with their heads down. They're reluctant to talk to, or even look at, any strangers. They even hesitate to make contact in public with people they do know. There is very little street life or spontaneity.

You've probably felt this same phenomenon yourself, on a much smaller scale. If a police car pulls alongside you while you're driving or stays right behind you, don't you automatically feel just a little paranoid and wonder if you've done anything wrong?

You become self-conscious and maybe a little nervous. You might even drive over-cautiously, to the point of being dangerous. In short, you change your behaviour, just because you feel that you are under scrutiny by an agent of the state.

Is a police camera really any different?

Limiting the way the police collect information about us is fundamental to our freedoms in Canada. Some people are going to argue that it makes the job of the police harder. Well, we all want crime prevented, and criminals caught. Our police do a vital job, usually excellently. But in this country, in a free society, we don't give the police unlimited power to violate the rights of the citizenry.

We don't allow them to build dossiers on citizens "just in case."

We don't allow them to force people at random to identify themselves on the street.

We don't allow them to enter and search homes, or open mail, or wiretap phones, without a warrant.

Instead, we say that there always has to be a balance-a carefully reasoned balance-between the legitimate needs of law enforcement and security on one hand, and the need to maintain our rights and freedoms and values on the other. I suggest to you that video surveillance cameras on public streets needlessly and dangerously tip that balance.

You'll notice that I'm making a distinction here between cameras on public streets and those that are in private places open to the public, for instance in banks or convenience stores.

First of all, in places such as stores there is an element of consent. If you don't want to appear before a camera, you have the choice of refusing to enter a given store. But if we end up with cameras all over our public streets, short of levitating above them, you have no way of withholding consent and still getting from place to place.

Second, those cameras in private places are for a very limited kind of anti-crime purpose. There's no likelihood of anyone being interested more broadly in you as a citizen-who you meet, where you go, what you do.

I'm sure some people will say, "Come on: police cameras on the street will make us feel safer." Look, we all want to feel safer.

But we'd be safer if the police could come into our houses whenever they wanted, just to look around. They could catch more criminals if they were free to listen to all our telephone conversations, just to make sure we weren't committing a crime. We could allow them to intercept and read all our mail and e-mail, just in case. These would be pretty effective means of making us safer-but no one in their right mind would argue for them.

We don't violate our most fundamental values, the very things that make our society worth living in, just because it makes us feel safer.

Now, some people may object that the video cameras will be in public places, not in private ones, so privacy isn't an issue. How can it be reasonable to expect privacy in a city street? My answer is that, obviously, there are degrees, gradations of privacy.

You're entitled to more privacy in your own home than when you step out into the street. But you still have the right to expect some privacy when you step out your door. Sure, you have to expect to be noticed, in passing. And people who know you may recognize you. But that's not the same as being systematically monitored and observed by anyone, let alone by the police.

If you're on a street chatting with a friend or on a cell phone, you can expect that strangers passing by may overhear parts of your conversation. But if a stranger stops and stands close to you and is obviously listening, you'll probably feel-quite rightly-that your privacy is being invaded.

And if someone across the street or on top of a building were systematically listening to your conversation through a directional microphone, you would certainly believe that your right to privacy was being violated. If you're sitting on a park bench reading a letter, you should expect that people can see you, and see what you're doing.

But you wouldn't expect someone to sit beside you and read over your shoulder, or to zoom in on your letter with a video camera.

Suppose a police officer decided to walk directly behind you on the street all day, quite obviously and deliberately following you everywhere you go. I'm sure you'd find that unacceptable, even if he wasn't saying a word to you or bothering you in any direct way. You'd probably consider it harassment-because he was invading your privacy.

How different is it, then, if your movements are simply tracked by a police offer watching you on screen as you pass from the range of one camera to the next?

So, I believe you do have a reasonable expectation of privacy when you're out in public. And more than that, you have a fundamental right to privacy. That's the problem with that term the courts have been using, "reasonable expectation of privacy."

A notice telling you you're going to be monitored or recorded by video cameras in a public place can tell you that it's not "reasonable" to expect any privacy. But it can't magically erase your fundamental right to privacy. This fundamental right can't be eliminated simply by putting up signs telling you that it's going to be ignored or trampled on.

Some people in Hamilton may say, "the cameras won't bother me because I don't go into those neighbourhoods". But if those cameras have the same effect-or lack thereof-that surveillance cameras elsewhere have had, the most they will do is displace crime into areas outside their range. Then the argument will be to put cameras there too. The logical extension will be cameras everywhere.

But even that is only the beginning. Video surveillance is just the first step. As the system stands right now, you are seen, but at least you can remain anonymous unless someone watching the videos knows you.

But what we call "function creep" virtually guarantees that if you put up enough cameras they'll soon be linked to biometric technology that will take away that anonymity.

Biometric facial recognition technology can analyze the image of your face, digitize it, classify it, and link it up to a police database. That can be used to compare your face against the images of known criminals or suspects-or simply to identify you, to associate your face with your name and address.

This isn't science fiction; it's existing technology. It's already being used in some places. It was used at last year's Super Bowl. It was set up on the streets of Tampa, Florida, and a growing number of U.S. airports are adopting it.

Again, you may say, "so what? Let them compare me against anyone they want. As soon as the technology shows I'm not someone they're looking for, that's the end of it and no harm done. "

Trouble is, these biometric systems are far from foolproof. They don't correctly match people to their own photos a lot of the time. They also match people to the wrong photos, photos of other people-falsely identifying them as someone the police can be looking for.

You could find yourself suddenly surrounded by police officers with guns drawn-just because the technology hiccuped. And it won't be long before we see these video surveillance systems linked to our driver's license or passport photos. Before you know it, when you walk along the street, the police will know exactly who you are. Your fundamental privacy right of anonymity will have disappeared.

In Canada, we don't allow the police to stop people at random and force them to identify themselves. You aren't even required to carry any identification. But now the technology will do it for them.

Some of you may be thinking, "So what? I've got nothing to hide." Well, think a little longer.

Think how easily the simple, innocent things you do can be misinterpreted by someone observing you. Someone stops you on the street and asks for directions. You tell him what he wants to know, and maybe chat for a moment. Then he goes on his way.

What you don't know is that on the police screen, biometrics has identified him-rightly or wrongly-as a suspected terrorist. And, of course, your name and address are available too. The watchers have no way of knowing what was said, just that you met and talked.

Next thing you know-or rather, don't know-you're in a police database as a suspect yourself.

Or there's a coffee shop or little restaurant you like to go to regularly. The location's convenient, or you particularly like the food. But you have no way of knowing that the police consider it an organized crime hangout. Because you're observed and identified as someone who goes there all the time, maybe you end up on some police list of suspicious characters-with no chance to ever explain yourself.

Maybe cameras were put up because prostitution is a problem in the area. You're stopped by a prostitute who asks you to protect her-she says someone is stalking her and asks you to walk her down the street to her car.

Because you're a decent person and you care about the safety of a fellow human being, you decide to take her request seriously and you walk with her.

The camera watches you, and you're now on record as accompanying a prostitute. You'd never know about any of these misinterpretations. Except maybe one day you apply for a job that requires a security clearance-and you can't get one, for some unexplained reason.

Or you're refused entry to the United States.

As cameras multiply and observation increases, we'll learn to avoid doing things that could be misinterpreted.

And that's the whole point: If you have to go through life knowing that everywhere you go, everyone you meet, everything you do, may be observed, scrutinized, cross-referenced, judged, maybe misinterpreted and used against you by persons unknown, by authorities of the state-if you have to go through life like that, you are not truly free.

So against this background, let me now come back to those four tests I mentioned: necessity, effectiveness, proportionality and availability of a less privacy invasive alternative.

First, necessity. Is crime so spiralling out of control here in Hamilton, that the enormously invasive remedy of police cameras is necessary? Actually it's not.

Crime rates in Hamilton are decreasing, and have been for some time. Your Superintendent Terry Sullivan was quoted in the Hamilton Spectator as saying a rash of criminal incidents in the late 1990s was what led police to come up with this video surveillance idea. Well, in 1997, there were just over 50,000 criminal code incidents in the city. Since then, that figure has dropped steadily. In 2000 there were 44,000-a 12% drop from just three years previously. That's hardly a crime crisis, is it?

The executive director of one of your business improvement areas said that businesses are concerned about security, property damage, and vandalism. Video cameras are, in her words, "a tool to get a handle on this." But property offences were about 20% lower in 2000 than they were in 1997.

Homicides, assaults, sexual assaults, break-and-enters, car thefts, the very things that most people worry most about, were less common in 2000 than they were three years before. You didn't need video surveillance to bring those crime rates down. Why would anyone think you need it now?

It's the same story for crime rates in Ontario as a whole, and for the rest of Canada as well. So there is no evidence of some huge new crime wave that makes it urgent and necessary to turn to video surveillance cameras as a solution.

But even if there were, that still brings us to the next test-are video surveillance cameras on public streets effective in reducing crime? The answer is no. All the available evidence indicates that at very best they displace it-move it from where the cameras are, to where they aren't.

Even a spokesman for the RCMP detachment out in Kelowna-where they've already put up a camera-was quoted last summer as saying the cameras would mostly displace crime.

He said crime would be displaced from downtown to Kelowna's residential neighbourhoods-and that would be a good thing because homeowners would be quicker to call the police.

In December, I asked the RCMP Commissioner for figures on the number of arrests brought about by the Kelowna camera since its installation-and for statistics comparing the overall crime rate in Kelowna during the months the camera has been used to the rate for the same period the previous year. I was told the information doesn't exist-even though this is supposed to be a pilot project.

In the U.S., the picture is no better. In Long Island, in Newark, and in Charleston, West Virginia, they set up video surveillance systems and found that they made no difference to crime rates.

In Mount Vernon, New York, they ran one of these systems for three years and then dismantled it because it didn't lead to a single arrest.

Last June, the city of Tampa in Florida installed a video surveillance system that included a biometric facial recognition system. The system never identified a single individual in the police's database of photographs. What it did do was make a lot of false matches, including people of the wrong gender and with big differences in age or weight.

In fact, the system was so inaccurate that the police department stopped using it a couple of months after it was installed.

The United Kingdom has more video surveillance cameras than any society in the world-more than two million of them, and the number is increasing every year.

And the facts speak for themselves.

In Glasgow, they evaluated their surveillance system one year after it was set up. At first glance, it looked as though the crime rate had fallen. But a harder look at the statistics showed that crime had actually increased by 9%.

London has roughly 150,000 video surveillance cameras. Last year, it had more cameras than ever before. And guess what? Last year, street crime in London increased by 40%.

My Office spoke just 2 weeks ago with Jason Ditton, a professor of criminology in the Faculty of Law at the University of Sheffield in England. Professor Ditton is widely recognized as perhaps the world's leading expert on the effectiveness of open-street video surveillance. He's been studying it since 1993.

Here's what he told us:

"There isn't convincing evidence that open-street closed circuit TV reduces either crime or the fear of crime. Indeed, for every reliable study that shows a benefit, I can show another that contradicts it. If evidence of success is a prerequisite of installation, I can confirm that no such evidence exists."

And, you know, it really isn't surprising that video surveillance can at best displace crime, not reduce it. Logic tells us the same thing.

A whole lot of factors-social, psychological, economic, you name it-underlie criminal behaviour. Putting up a bunch of cameras isn't going to make drug addicts quit cold turkey and go sip a latte at Starbucks instead.

It isn't going to make prostitutes sign up for secretarial college. And it isn't going to make muggers or purse-snatchers repent their life of crime. It might just make them move away from the cameras by switching to home invasions or car-jackings.

And unlike a police officer on the beat, a video camera can't make you safer anyway. If you're being mugged or beaten up or stabbed, it can't jump down off the pole and rescue you.

By the time the police get there, the crime will probably be long over and the damage to you will be done-these things usually happen in a flash. So, in summary, there is no convincing reason to believe that video surveillance cameras here in Hamilton will reduce crime or make you safer.

That brings us to the third test-proportionality. I don't think I need to belabour this one. Since there are no demonstrable safety benefits, the harm these cameras cause can hardly be proportional.

There is no reason to believe they will make you safer. There is every reason to believe-to know-that they will reduce privacy, and contribute to creating a society where our fundamental rights and freedoms are greatly and permanently diminished.

And now for the final test-are there any less privacy-invasive alternatives available?

Sure, there are lots of alternatives to look at-things like improved street lighting, neighbourhood watch programs, or greater police presence on the streets.

And of course, there are broader initiatives to consider that may go deeper towards the roots of the problem-initiatives in social policy, economic policy, education and rehabilitation programs, urban development, you name it.

So I suggest: Forget about the cameras, and focus on those other crime prevention measures.

I'd like to be able to say to the rest of Canada, when I go around the country speaking against the spread of these cameras: "Look at Hamilton. The people of Hamilton took the time to think about the facts, to look at the evidence, to consider what they are about to lose, and they said no to the cameras."

And I'm going to say to them what I say to you now: Be an example of strength, not timidity. Base your decisions on facts, not on scare-mongering. Build a genuinely safe free society, not a falsely safe police state. Show your commitment to the values that make this country great. Show that you think our country, our society, is worth preserving and stand up for privacy.

Thank you.