Photo illustration.

Traffic-enforcement cameras have been a common sight in Davenport for 13 years, but now the city is using new cameras for a different purpose: to help prevent and solve violent and property crimes.

Davenport in the past month has begun a pilot project with 18 cameras at four intersections on Washington Street south of Locust Street. The city purchased the cameras for nearly $54,000 as part of a larger neighborhood-revitalization program that also includes street and sidewalk improvements.

“The thing that makes this a little different is the sheer number of cameras,” said Clay Merritt, capital manager for the City of Davenport. “Instead of having one camera in one location just looking at one particular view, ... this is taking a much larger-scale approach.”

The idea is to see to what extent the cameras prevent crime, and how much they assist police in solving crimes that do occur. “If there are cameras in the area that can catch footage, perhaps that would have people ... think twice” before committing a crime, Merritt said. “Hopefully that deters crime ... , and the number of police calls in the area would go down.”

But even if that’s not the case, he said, the cameras might be valuable to, for example, identify suspects: “It’s an added tool. It’s not just looking at it from a crime-prevention standpoint.”

For now, the city is merely dipping its toe into these surveillance waters, and Merritt said the city is still working to get sufficient video quality from the cameras. That problem resulted in police getting no usable footage of a May 10 shots-fired incident.

Still, it’s worth noting some issues with the city’s approach. Merritt didn’t offer any methodology for determining the success or failure of the Washington Street pilot project, and he said there’s no time frame for an evaluation.

“The goal in the long term, if it is proved that it’s successful, is that it can be expanded into other areas,” he said.

The city has already budgeted more than $300,000 for its next fiscal year to add camera locations, Merritt said – probably along high-traffic corridors such as 53rd Street and Kimberly Road, although additional cameras will need to wait for fiber-optic cable to be installed.

This suggests the city has, to some extent, already reached a conclusion about the effectiveness of surveillance cameras without answering key questions. What results will indicate the project is a success, and in what period of time? Is the city taking account any displacement effect? (If crime drops in the target area, is it merely being pushed elsewhere?) How will the city determine whether the relatively modest financial cost of cameras is worth whatever benefit it sees? And are 18 cameras in one neighborhood a fair test of the effectiveness of such a system?

That doesn’t even touch on some of the procedural and social questions raised by police video surveillance. “It all comes down to how much of a Big Brother do we want our government to become,” said Veronica Fowler, communications director for the American Civil Liberties Union (ACLU) of Iowa.

Even at the pilot-project stage, she said, cities need to be clear with the public about their rationale, goals, and procedures for surveillance cameras. “The taxpayers are paying for them,” she said. “They should be backed up with facts based on studies, and not just ‘Yeah, it’s going to be a lot safer if we have a lot of cameras.’ ... Even with a pilot project, there should be clear policies in place.”

She said those should include not only technical issues such as how long footage will be kept but also explicit protections against abuse – for example, policies prohibiting monitoring lawful comings and goings, prohibiting stalking, and prohibiting targeting neighborhoods with high concentrations of racial minorities.

“There are all these things that we would love to be sure that the city is being mindful of and has policies in place to spell that out,” Fowler said. “At least that would give people some confidence that they have thought these things through, and there are clear policies on paper. So if any of those things are violated, some action can be taken. And people know what the rules of the game are.”

That sort of discussion is, admittedly, best-suited to the police department or the city administrator. However, an interview request with Davenport Police Chief Paul Sikorski was declined, and a representative of the police department referred all questions to Merritt. City Administrator Corri Spiegel also referred questions to Merritt.

But we know the answer to one core question: Police surveillance cameras, in general, are not particularly good at deterring crime. They can be effective in certain circumstances – specifically in parking lots or garages – but not in the way Davenport is using them.

And while Davenport has cited the example of Dubuque’s extensive surveillance-camera network, that case doesn’t provide clear evidence of deterrence.

The Appeal of Cameras

It’s easy to understand the appeal to police departments of surveillance cameras. Camera technology continues to get better and cheaper, meaning that the expense is minor assuming the data infrastructure is already in place. In Davenport as in many other cities, the cameras aren’t actively monitored, meaning that they don’t create an added staff cost. (Davenport will save its videos for 21 days.)

If the cameras deter crime by their mere presence or help solve cases, the cost could be justified.

Yet despite that, and despite the alarm that’s been sounded – including in these pages – about an ever-increasing surveillance state, it’s a bit surprising that police-department surveillance cameras in public spaces are not more common in the Quad Cities.

Bettendorf, for example, has never even entertained the idea of installing surveillance cameras. “We have not discussed placing cameras in the city,” Chief Phil Redington wrote in an e-mail. “It just has not been considered.”

Similarly, Moline has no permanent surveillance cameras and no plans to install any – although the police department’s public information officer, Detective Scott Williams, wrote that temporary cameras have been used “for a specific narcotics investigation or where we believe a crime may occur.”

He added: “We understand the arguments and stated purposes of these [permanent] surveillance cameras, but our current and past police chiefs and city leaders have [not] yet found cause to pursue a project such as this. We will be watching Davenport’s pilot program with interest.”

Rock Island, according to Deputy Police Chief Jason Foy, has installed 22 surveillance cameras – downtown in 2006, at the base of the Centennial bridge in 2009, and on 15th Avenue from Fifth Street to Longview Park in 2012. Foy said that “they have absolutely assisted in [solving] some cases since they’ve been put in.”

But he said he wasn’t sure if they’ve prevented crime: “I don’t know how many people even know ... that they’re there. I don’t know how much people pay attention to them.” Even if it were shown that crime has dropped in areas where cameras are present, he noted, one has to take into account, for example, that “the downtown has changed over those 11 years.”

He added that the city has no plans to expand its surveillance-camera program.

If the Quad Cities are moving tentatively or are lukewarm toward surveillance, Dubuque is all-in – with hundreds of cameras throughout a city roughly half the size of Davenport. In praising the value of his city’s surveillance program, City Administrator Michael C. Van Milligen gave the example of a random murder a few years ago that the cameras helped solve. “One of the first things the police do is look if any part of that crime – getting to the crime, leaving the crime, or committing the crime – could be on one of our cameras,” he explained.

When an elderly woman was found dead downtown one morning, he said, “the police ... had nothing to go on. ... They had zero evidence.” But because of camera footage, “by 9 o’clock that night, they had him [the murderer] under arrest and a confession.”

Successes such as that, Van Milligen said, pushed the city to this year give shift commanders and the detective bureau access to the camera system in their offices. “It creates a tremendous amount of evidence,” he said.

Modest and Inconsistent Results

So there’s certainly support for the idea that surveillance cameras can help police departments clear cases. But evidence that cameras prevent crime is weak except in one specific circumstance.

A 2009 study of public surveillance cameras in Philadelphia summarized that research into the effectiveness of cameras as a tool to reduce crime is “mixed. Furthermore, there has been a paucity of rigorous evaluations of cameras located in the USA.” The authors concluded that in Philadelphia, “the introduction of cameras is associated with a 13-percent reduction in crime.”

Merritt agreed that studies are not unanimous that surveillance cameras reduce crime: “You can read it either way.”

But even the Philadelphia study finding a decrease in crime said results were inconsistent: “The evaluation suggests that while there appears to be a general benefit to the cameras, there were as many sites that showed no benefit of camera presence as there were locations with a positive outcome on crime.”

This squares with a 2009 analysis of 44 studies that concluded surveillance cameras “caused a modest (16 percent) but significant decrease in crime in experimental areas compared with control areas.” However, the authors found that “this overall result was largely driven by the effectiveness of CCTV [closed-circuit television] schemes in car parks, which caused a 51-percent decrease in crime. Schemes in most other public settings had small and non-significant effects on crime: a 7-percent decrease in city and town centers and in public-housing communities.”

In other words, surveillance cameras are best deployed as a crime-prevention tool in areas with few people and a large amount of property crime – specifically parking lots and garages.

The Misleading Model of Dubuque

In opposition to that general conclusion appears to be the City of Dubuque.

Representatives of the City of Davenport visited Dubuque because of that city’s expansive surveillance program. According to City Administrator Van Milligen, Dubuque has over the past decade installed nearly 1,000 surveillance cameras along with the underlying infrastructure – primarily at intersections, but with several hundred cameras also in parking facilities. Van Milligen said most cameras have been installed in the past six years.

And our neighbor to the north saw its reported violent-crime rate drop by more than half from 2010 to 2014. “You wouldn’t see the crime reductions that we’re seeing if it didn’t have a deterrent effect,” Van Milligen said.

That’s a strong correlation between cameras and decreased crime, but using Dubuque as a model – or a justification – for a surveillance-camera program would be seriously problematic. In fairness, in our interview Merritt didn’t cite Dubuque’s crime numbers as a rationale for Davenport’s program.

Probably with good reason.

For one thing, that dramatic drop in violent crime is illusory. In 2011, Dubuque started counting its aggravated assaults differently, and as a result the number of reported aggravated assaults in the city dropped from 279 in 2010 to 82 in 2011. Prior to 2011, Assistant Chief Terry Tobin told the Telegraph Herald newspaper, aggravated assaults were over-reported: “We reported any assault with injury, like if they had a scratch, as an aggravated assault.”

Aggravated assaults accounted for nearly 84 percent of Dubuque’s violent crime in 2010, so the tighter reporting criteria naturally but artificially drove the violent-crime rate down.

Take that reporting change out of the picture, and Dubuque’s crime trends look similar to Davenport’s – a substantial drop in both violent crime and property crime, but not anything out of the ordinary. From 2010 to 2014, Davenport’s violent crime rate dropped by more than 26 percent and its property-crime rate dropped more than 14 percent, according to FBI statistics. In the same period, Dubuque’s reported violent crimes (outside of aggravated assaults) dropped by 13 percent, while its property-crime rate dropped by more than 15 percent.

It’s also important to note that the law-enforcement surveillance function of Dubuque’s cameras was basically an accident. The city put the cameras in for one purpose: to automate intersection traffic signaling.

“The initial installation of the camera system at the traffic lights was purely to control traffic,” Van Milligen said. Then, “as we saw this parade of police officers going upstairs almost every day to look at the traffic-camera system, we said, ‘Wait a minute. There’s something bigger here that we ought to consider.’ And the council decided to make it a priority that we would supplement our traffic-camera system with a security-camera system.”

So law-enforcement use of the system was essentially a tangential benefit.

“If you look at our system in isolation of security cameras, you’d be doing a huge disservice to whoever’s considering doing this,” Van Milligen said. “If they were only for that [law-enforcement] purpose, you could say, ‘Wait a minute, that serves a single purpose for that level of expense.’ But ... it’s for a much greater purpose. Moving traffic is extremely important. ... The backbone of the system is around traffic.”

Within three years, he added, Dubuque will have an entirely self-adjusting traffic-control system based on cameras.

In Davenport, Merritt said the city’s surveillance cameras will also be used by traffic engineers for things such as traffic counts, but that’s a far cry from Dubuque’s multifaceted deployment.

Social Costs

Beyond the cost-effectiveness of surveillance cameras, Davenport’s pilot project – and plans for additional cameras – beg for discussion of the social costs of widespread public surveillance. Merritt said he’s not heard any concerns about privacy or potential abuse of surveillance cameras from Davenport residents, but those issues should still be considered.

The ACLU has offered four answers to its question of “What’s Wrong with Public Video Surveillance?”

In summary, it says: “Although the ACLU has no objection to cameras at specific, high-profile public places that are potential terrorist targets, such as the U.S. Capitol, the impulse to blanket our public spaces and streets with video surveillance is a bad idea.”

The first reason is its lack of proven effectiveness.

The remaining reasons mostly relate to potential abuse of the technology and the likelihood of a chilling effect. These are two sides of the same coin. On one side, “the growing presence of public cameras will bring subtle but profound changes to the character of our public spaces. When citizens are being watched by the authorities – or aware they might be watched at any time – they are more self-conscious and less free-wheeling.”

On the other side, surveillance cameras give authorities the tools to monitor not just crime but also peaceful protests and individual behavior – both perfectly legal. The ACLU asks: “Do we want the authorities installing high-resolution cameras that can read a pamphlet from a mile away? Cameras equipped to detect wavelengths outside the visible spectrum, allowing night vision or see-through vision? Cameras equipped with facial recognition, like those that have been installed in airports and even on the streets of Tampa, Florida? Cameras augmented with other forms of artificial intelligence, such as those deployed in Chicago?”

It further noted that “during the Civil Rights movement and the Vietnam War, ... the FBI – as well as many individual police departments around the nation – conducted illegal operations to spy upon and harass political activists who were challenging racial segregation and the Vietnam War. This concern is especially justified since we are in some respects enduring a similar period of conflict today.”

The organization advocates for “legally enforceable rules for the operation of such systems”: “While the Fourth Amendment to the U.S. Constitution offers some protection against video searches conducted by the police, there are currently no general, legally enforceable rules to limit privacy invasions and protect against abuse of CCTV systems. Rules are needed to establish a clear public understanding of such issues as whether video signals are recorded, under what conditions, and how long are they retained; what the criteria are for access to archived video by other government agencies, or by the public; how the rules would be verified and enforced; and what punishments would apply to violators.

“There have long been well-established rules governing the audio recording of individuals without their consent. (There is a reason surveillance cameras never have microphones.) It makes no sense that we don't have equivalent laws for video recording.”

Taking all this together suggests certain best practices for surveillance cameras. First, they should be deployed where they’re known to be effective: in parking areas.

Second, if they’re used in other areas, the cameras should serve clear functions beyond law enforcement, such as Dubuque’s use of cameras for traffic signaling.

In Davenport, where neither of those things is true, the pilot project needs to have a rigorous evaluation to demonstrate what, if any, benefits it produces. Only then will it be possible to determine if the financial and civil-liberties price tags are worth it. While the cameras were cheap in the grand scheme of the city budget, they still cost taxpayers more than $50,000, with plans for six times that amount in future spending despite no track record of effectiveness.

As the ACLU put it in an article on terrorism: “The key question to ask is not ‘Do cameras provide any potential benefit ... .’ The questions are ‘Do cameras provide enough benefit to justify their cost?’ and ‘What are those costs?’”