My wife has never run a red light in her life. The city has two photographs that say otherwise.
Both were taken at the same intersection — Walnut Lane and Henry Avenue — while she was turning left onto Ridge. It's a turn we've made a hundred times, the same turn millions of Americans make every day. Both times there was no oncoming traffic. Both times the light ran its entire cycle while the car was already committed to the turn: green, a blink of yellow, red — quicker than any honest yellow should be, and faster than felt fair. No screech of brakes. No near-miss. Nobody in the crosswalk, nobody anywhere close to harm. Just a flash from a box on a pole and, a few weeks later, two $100 demands in the mail.
Two hundred dollars poorer for a maneuver that hurt no one.
Here's the part I can't shake: I still get a knot of anxiety every single time I drive through that light. Not because the intersection is dangerous — it isn't — but because I know the machine is up there, waiting, and I know there is nothing I can say to it. No officer who watched it happen and weighed whether we'd actually done anything wrong. No hearing where I get to point out that the yellow was too short to clear the turn legally. No one to argue with, no discretion to appeal to, no human who decided we were a danger that day. Just a threshold, a photograph, and a bill.
Then, a few months later, they got her again — a speed camera this time, on Broad Street. Her offense: driving one of the widest, straightest, most forgiving stretches of road in the city at the speed the road all but invites. The limit on Broad is 25 — a number set so far below how the street actually drives that just about everyone on it is a ticket waiting to be printed. We didn't even know Broad was being clocked; there was no real sign, nothing that registered as a warning that a machine was measuring our speed. Just, a few weeks later, the same envelope: another $100, another photograph, another moment that put no one in danger.
Set a limit that far under the natural speed of a road and you're not calibrating for safety — 35 would do that honestly. You're stocking the pond.
We paid them — all three, three hundred dollars for three moments that hurt nobody. I was angry. And then I did the thing I always tell other people to do: I went and read the actual legislative record. I run a site that tracks every bill Philadelphia City Council has passed since 2000, so I have the receipts. I expected to find a contentious fight — surveillance versus safety, the kind of debate a city is supposed to have before it points cameras at its own residents.
I found almost nothing. And that is what turned my anger into opposition.
The receipts: 25 years of yes
Philadelphia runs two automated-enforcement programs, and the legislative trail is remarkably one-sided.
The red-light camera program came first. In November 2000, Council passed the enabling ordinance creating civil liability for camera-caught red-light running — a $100 fine, issued from a photograph. Over the next two decades, Council came back to amend, extend, and expand that program dozens of times, intersection by intersection. I went through the votes. They are almost uniformly 16–0 and 17–0. As recently as December 2025, Council was still adding new red-light cameras, this time at School House Lane and Henry Avenue.
The speed-camera program is newer and moving faster. In 2019, Council created a new chapter of the city code — "Use of an Automated Speed Enforcement System" — to put cameras on Roosevelt Boulevard, with tiered fines of $100, $125, and $150 for drivers caught going 11 mph or more over the limit. For a few years that was the whole program: one notoriously deadly road.
Then, in the span of about eighteen months, it became everything:
- May 2024 — Speed cameras on Broad Street
- October 2024 — Speed cameras expanded across Philly streets
- February 2025 — Speed cameras in school zones
- September 2025 — Speed cameras on state routes
Every one of those passed unanimously, or close to it.
Nobody said no
Here is the single fact that bothers me most. In twenty-five years of Council building out an automated surveillance-and-fine apparatus that now spans red lights, boulevards, school zones, and state routes, I can find essentially one dissenting vote.
On that original 2000 red-light bill, the vote was 15–1. The lone "no" belonged to Councilmember Cohen. After that, the record goes quiet. Bill after bill, year after year, the cameras expanded and the tally was unanimous.
I want to be fair about what that does and doesn't mean. Unanimity can signal genuine consensus on something that obviously works. But it can also signal that a program stopped being debated — that "more cameras" became a default, a routine line-item nobody campaigns against because the alternative is to look soft on a child getting hit on the Boulevard.
When a city expands surveillance of its own residents for a quarter century without a single sustained argument against it on the record, that's not a settled debate. That's an absent one. So consider this my no vote, cast late.
The honest case for the cameras
I'm not going to strawman the other side, because the other side has real points and I'd be lying if I pretended otherwise.
Roosevelt Boulevard earned its cameras. It has been one of the most dangerous roads in Pennsylvania for decades — a twelve-lane river of traffic running through dense neighborhoods, with a death toll to match. After automated enforcement went in, speeds and crashes on the corridor fell sharply. If you believe a policy should be judged by whether it stops people from dying, the Boulevard is the strongest argument the cameras have, and it is a strong one.
The deeper point is this: a camera doesn't get tired, doesn't profile, and doesn't escalate a stop into a tragedy. We have spent years, rightly, worrying about what happens when an armed officer pulls over a driver for going 9 over. A camera never pulls anyone over. It never asks you to step out of the vehicle. For a city trying to reduce dangerous police-citizen encounters, "let the machine handle speeding" is not a crazy answer. It might even be a humane one.
I take all of that seriously. It's why this is hard.
Why I still oppose it
And yet.
The money never sleeps. A fine is a tax that we only call a fine because a machine caught you. Once a city wires part of its budget to violations, it acquires an interest in violations continuing — in thresholds staying low, in yellow lights staying short, in "just one more corridor." I'm not alleging a conspiracy. I'm pointing at an incentive, and incentives win in the long run. A safety program that would celebrate its own obsolescence is trustworthy. A revenue program that needs you to keep speeding is not.
Mission creep is the whole story. Look back at that timeline. We started with one deadly boulevard and a narrow justification almost no one could argue with. We ended, within a few years, with cameras as the city's general-purpose enforcement tool. The May 2024 vote that put a speed camera on Broad Street — the same one that eventually mailed my wife a $100 ticket for driving Broad the way Broad is built to be driven — passed 17–0. Not one member rose to ask whether a 25 mph limit on that road was protecting anyone or simply printing money. There was even a 2020 proposal to add automated license-plate readers to the mix, which is a different thing entirely: not catching a dangerous act, but logging where your car is. Each step was small. Each step used the credibility of the last. That's how surveillance infrastructure always gets built — not in one vote you can rally against, but in twenty you never hear about.
And the cameras still don't fix the road. Slower cars are safer cars, and the Boulevard's numbers are real — I won't pretend otherwise. But a camera polices speed, not design, and a twelve-lane highway run through row-home neighborhoods stays lethal no matter how many photographs it takes. On August 28, 2025 — five years into the program — a single crash on the 9000 block of East Roosevelt Boulevard killed a driver and a pedestrian and left a third person in critical condition. Meanwhile the deadliest road in the city in 2024 wasn't the Boulevard at all — it was Broad Street, at more than a fatality a month: the same Broad Street where the city's answer was to mail my wife a ticket for doing a few miles an hour over an invented limit. When enforcement clusters where the fines come easy instead of where the dying is worst, you have to ask what problem it was really built to solve.
Automated justice removes the human. The thing I hated about those tickets was never really the money. It was that there was no one in the loop. Due process used to mean a person had to look at you and decide. Now the decision is a threshold and the appeal is a web form. We should be slow — genuinely slow — to remove human judgment from the act of the government penalizing a citizen, even when the human was imperfect. Especially then.
And the debate never happened. Even if every camera in this city is justified, we deserved the argument. We deserved a Council that made the case out loud, weighed the privacy cost in public, and put a real check on expansion. Instead we got unanimity and silence, and a program that grew while no one was watching it watch us.
What I'd actually want
I'm not anti-safety, and I'm not naive enough to think the Boulevard should go back to being a speedway. So here's what accountable automated enforcement would look like to me:
- Revenue-neutral by design. Fines fund traffic safety and nothing else; any surplus triggers a rate cut, not a new corridor. Break the budget's dependence on violations.
- Sunset clauses. Every camera authorization expires and must be re-justified with crash data, not renewed by default. Make Council vote yes again on purpose.
- A hard wall against the LPR creep. Speed and red-light enforcement is one debate. Mass location tracking is another, and it should never ride in on the first one's coattails.
- A real appeal — to a person. If the city is going to penalize you by photograph, you get a human being to contest it in front of, easily and for free.
None of that is anti-camera. It's pro-debate. It's the conversation Philadelphia skipped twenty-five years ago and has been skipping ever since.
We paid our tickets. I'd just like to think the city had to earn them — and that next time, somebody on Council says what I'm saying here, before the vote instead of a resident saying it after.