Cameras Only Deter People Who Believe They’re Being Watched
In 1785, the philosopher Jeremy Bentham designed a prison built as a ring of cells around a central tower. A single guard in the tower could see into every cell, but the prisoners could never see him. No one inside knew when they were being observed, so they behaved as though they were always being watched. The guard hardly mattered. The mere possibility of the guard did the work.
Bentham never built the prison as he imagined it, but the idea outlived his design. Michel Foucault picked it up almost two centuries later and argued that modern institutions, from schools to workplaces, rely on the same silent pressure. People police themselves when they think they might be seen.
A camera on a tower doesn’t physically stop a thief; it changes what they think will happen next.
Independent evidence shows this works, though less than the industry likes to admit. A review of CCTV research summarized by the College of Policing found crime was around 13% lower in areas with cameras than in comparable areas without them. That is a real reduction, and I would not dismiss it, but it hides a lot of variation. In car parks, which are contained, often unattended and full of vehicles and equipment, crime fell by 37%. In town and city centers, the review found no statistically significant effect overall. The setting matters as much as the camera.
The same research says something more interesting about how cameras are used. Schemes with active monitoring produced bigger effects than passive systems, and schemes that ran alongside other measures did better than those relying on cameras alone. The authors of the largest review warned against treating CCTV as a standalone answer. A camera that is watched, and sits within a wider effort, is a different thing from one that quietly records.
First, people must know the cameras are there. A device hidden in a ceiling tile may be useful for evidence, but it deters nobody. Deterrence is a message and it needs to be received to have impact.
Second, the threat must be credible. Bentham’s guard could not be seen, but he did exist. A camera that nobody monitors, or that records to a drive nobody checks, is a bluff. Offenders talk. Word gets around that a particular site is all show and then the bluff is called. I have seen sites that spent money on visible equipment and then wondered why the thefts carried on. The answer was usually that nobody was ever behind the lens.
Finally, the people being watched must trust the reason. The research on how surveillance affects those on the receiving end is not comfortable reading. Some studies link being monitored with stress, unease and, for people who are already vulnerable, paranoia. Anyone who has worked in an office where nobody explained why the cameras went up will recognize the mood. Watching without explaining breeds resentment and that is not a foundation for a safer place.
So, my view is that visibility and honesty must travel together. In practice, that means clear signage saying who is watching and why. It means cameras placed in proportion to the risk, not simply because the budget allowed for more of them. It means a trained person, or a well-configured system with a person behind it, responding when something happens. It means handling the footage properly under UK GDPR and the Data Protection Act 2018, with policies you would be comfortable showing to a regulator or a member of the public who asked.
None of this makes surveillance less effective. If anything, it makes it more so, because a camera people understand and accept is far easier to defend than one they suspect. It also keeps the industry on the right side of a public debate that is only getting louder. With police forces and retailers expanding live facial recognition on high streets and in shops, the question of whether the public understands why they are being watched has become impossible to ignore.
Bentham believed that the sense of being seen would make people behave better. He was partly right, but he was writing about a prison, where the watched had no say. Those of us putting cameras into workplaces, streets and building sites are working in a world where people can ask questions, complain and walk away. The technology has moved a long way since 1785. The test of whether we are using it well has not: would the people being watched think it was fair?
