June saw the deaths of two people riding bicycles in Cambridge. Both occurred at intersections and involved a turning truck. While we’ve seen a continuous decrease in the rate and severity of bike crash injuries in the city, these incidents underscore that there is more we should be doing, particularly at intersections.
While separated bike lanes are critical for safety on streets outside of intersections, intersections can also benefit from infrastructure safety improvements. In fact, the state’s Department of Transportation has a guide to designing safer intersections, and the National Association of City Transportation Officials also has a guide aptly named “Don’t Give Up at the Intersection.”
As the landmark Cycling Safety Ordinance is implemented, city staff have worked with stakeholders to design, refine and build separated bike lanes across the city. Among other safety benefits, these lanes can eliminate “dooring” crashes such as those that caused the deaths of Amanda Phillips in 2016 and Stephen Conley in 2022 in Somerville. When separated bike lanes arrive at most intersections in Cambridge, they often lose most of their physical protection.
City engineers can mitigate potential risks for vulnerable road users (people walking, cycling or using other non-car mobility devices) in several ways, with techniques specific to intersections. For example, they have used:
- Daylighting, in which parking is moved away from areas close to intersections and crosswalks to make vulnerable road users more visible to people driving.
- Turn hardening, in which corners of an intersection are extended to reduce their turning radius, encouraging turning vehicles to take turns more carefully.
- Signal phasing, which can separate bike and vehicle movements in different signal phases to reduce conflicts between cyclists and vehicles.
- Leading signal intervals, which Cambridge uses for pedestrian crossing lights and, in a few cases, for bicycles.
While these strategies are proven interventions that help reduce crashes, individually they are not enough to achieve the only acceptable number of serious injuries and deaths on the road: zero. According to the Safe Systems approach developed by the U.S. Department of Transportation, humans can be expected to make mistakes, and therefore redundancy is needed so one failure does not cascade into a tragic outcome.
A road user may not see or may ignore a red signal or prohibited turn sign. A driver may choose to park their car illegally in a bike lane or daylighting zone. While efforts should be made to reduce these occurrences, the design of our streets should prevent one failure or poor decision from becoming fatal. Cambridge should prioritize using a combination of these strategies in to add these critical layers of redundant safety.
Improved implementation: There are opportunities to improve some current implementations to increase their effectiveness. For example, bicycle signals must be visible and obvious to people traveling through an intersection for the first time. This is challenging, given that intersections vary widely across the city, so standardizing layouts as much as possible will be helpful.
Protected intersections: These extend the physical protection of the bike lane as far into the intersection as possible while incorporating elements such as signal separation, a reduced turning radius and better lateral separation between the bike lane and motor traffic.
Choice of materials for turn hardening: The city could make turn hardening more effective. For areas in which larger vehicles do not need to be accommodated, even more durable materials may be used. Daylighting and turn hardening materials must be durable and do their job of preventing vehicles from occupying the space. Flex posts may become damaged or destroyed after being driven over repeatedly. Many quick-build materials need constant monitoring and repair, or they will lose their effectiveness.
If you’d like to learn more about these techniques, see our more detailed summary with diagrams and photos.
We call on Cambridge to implement safer biking infrastructure across the city, including intersection safety improvements and the separated bike lanes needed to prevent deaths and injuries outside of intersections. Improved infrastructure can help prevent fatal crashes such as the ones this month – we must not wait for more deaths to make our streets safer.
Kevin Moses, on behalf of Cambridge Bicycle Safety
A stronger
Please consider making a financial contribution to maintain, expand and improve Cambridge Day.
We are now a 501(c)(3) nonprofit and all donations are tax deductible.
Please consider a recurring contribution.
These are great, but two more I’d add that are even simpler.
FOLLOW the rules of the road, all cars, bikes pedestrians.
ENFORCE the rules of the road on all cars, bikes, and pedestrians.
(We could also do with lower speed limits outside of the major roads.)
Yes the city needs to adjust the lights phase cycles… too many are set too fast and truck drivers have a tendency to jump the gun as it were at lights. Lengthening the light cycles by 10 seconds at intersections would also benefit pedestrians as many are problematic for folks with mobility issues or pushing carts or baby strollers or traveling with small children on foot.
q99: My wife was doored while biking (someone in a parked car opened their door) and thrown into the street. The person who did it was from Vermont, wasn’t used to people biking. And plenty of people who live here don’t even _know_ they need to be careful.
The City will give out tickets in these situations, but it’s a drop in the bucket.
Calling for people to “FOLLOW the rules of the road” as the _only_ solution is magical thinking. Yes, it would be good, yes we can do much better (red light cameras!), but designing a system around the assumption that individual human behavior will be perfect is a recipe for disaster.
The first engineering discipline to figure this out were organizations building large scale weapon systems in the 1950s, e.g. https://appel.nasa.gov/2008/06/01/an-introduction-to-system-safety/.
This broader perspective on systems safety has since spread beyond that. In this case, the US Department of Transportation calls for a Safe System approach (https://www.transportation.gov/NRSS/SafeSystem), and one of the tenants is “People will inevitably make mistakes and decisions that can lead or contribute to crashes, but the transportation system can be designed and operated to accommodate certain types and levels of human mistakes, and avoid death and serious injuries when a crash occurs.”
So if you want to prevent dooring, you can tell people to “FOLLOW the rules”… and it will never work. Or you can build separated bike lanes and just prevent the problem in most cases, and prevent the worst case scenario (being thrown in front of a truck) in the remaining cases.
Safe infrastructure renders enforcement superfluous.
All important but the segment selection for building the safe network is also a critical part. It we want to design safe systems accepting that people driving, biking, and walking will all make mistakes then choosing lower ADT roads and intersections with low vehicular turns should also be considered. The route selection is the most important part and an alternatives analysis for safety should always be done to assure the safest route for all age and ability users has been selected to prevent tragedies like the two that recently occurred understanding that no matter what we design mistakes will be made by people and reducing the number of potential conflicts is an important part of the analysis.
Itamar, I said I agreed with the solutions suggested, and ALSO added my own suggestions in addition, not as the ONLY solution. Enforcement in this city is a joke.
We also need reformed driver education and better signage to prevent what happened to your wife too. Hope she’s recovering.
And slaw, unfortunately, there never will be or can be 100% perfect infrastructure- we need better infrastructure AND better enforcement to approach our goals of safety.
Absolutely agree q99. Some of the worthwhile infrastructure and rule changes we are making are but window dressing without enforcement
“Lengthening the light cycles by 10 seconds at intersections would also benefit pedestrians” I would really like some evidence to back this up. From my understanding shorter cycles are generally seen as better for pedestrians (as long as there is adequate time to cross) because you don’t have to wait as long to cross the street. Are there locations where people don’t have enough time to cross?
More time for drivers is less time for pedestrians and should not be framed this way.
@q99 there can never be perfect enforcement, saying nothing of actively counter productive racist enforcement and murder by police, which you can’t just pretend doesn’t exist.
“Some of the worthwhile infrastructure and rule changes we are making are but window dressing without enforcement”
Exactly the opposite is true. Countries that focus on enforcement as the primary mechanism of street safety do not see good results. Countries that focus on infrastructure with aviation zero style approach do. The European countries that have made the most progress on reducing street deaths spend way less on policing and pretty much entirely focus on infrastructure. Our focus is off in the US but in exactly the opposite way as you suggest. We still generally design roads that encourage dangerous behavior and then tell people to do the opposite: wide roads with speed limits that don’t correspond to design speed, wide intersections designed for turns without slowing down in heavy pedestrian areas with signs saying to look, etc.
We already have plenty of enforcement. It generally means police pulling over cyclists doing things that keep them safer and ignoring the dangerous behavior of drivers at the same locations. Enforcement will always be subject to the biases and prejudices of the enforcers. Infrastructure that prevents or makes dangerous behavior extremely uncomfortable in the first place simply isn’t.
Otherwise enforcement means things like speed traps where police fill quotas and raise funds by catching people following the design of the road. Safe infrastructure actively prevents that sort of trap that not improving infrastructure and relying on enforcement does. The government simply shouldn’t be setting up traps like that. I believe it is fundamentally wrong and a pushing off of their responsibility for unsafe design onto the individual (second, third, and forth articles below talk about this directly).
Some relevant links:
https://smartgrowthamerica.org/traffic-enforcement-cannot-do-the-job-of-better-roadway-design/
https://www.bbc.com/future/article/20240517-vision-zero-how-europe-cut-the-number-of-people-dying-on-its-roads
https://www.strongtowns.org/journal/2022/11/14/mdf2022-speed-traps-have-no-long-term-effect-on-speeding
https://www.strongtowns.org/journal/2021/3/30/automated-enforcement
https://www.fastcompany.com/3064333/design-is-better-than-enforcement-to-make-cities-safer-for-everyone
https://visionzeronetwork.org/re-thinking-the-role-of-enforcement-in-traffic-safety-work-city-to-city/
*a vision zero
not aviation zero
You have it backwards. Enforcement is a drop in the bucket. People will misbehave regardless, and enforcement can only do so much. Despite heavy enforcement, people still speed on highways.
The only reliable way to make streets safer is through infrastructure changes. While enforcement helps, the streets themselves need to be safer.
There are dangerous intersections in Cambridge that can be improved with a few simple alterations. We know these changes work; we just need to implement them.
What @Slaw said. We already know the answers to these questions; there is a wealth of data available. There’s no need to rely solely on our own thoughts.
The answer is clear: enforcement has limited impact on safety, while infrastructure changes are highly effective.
However, some people will oppose any changes, even if they save lives or prevent injuries, if it means losing a single parking spot.
Even if we eliminated private cars, we would still have conflicts between public transit, delivery and bikes needing curb access for people with mobility issues which will bring vehicles and cars into shared spaces and necessitates enforcement.
Opposing enforcement in favor of just infrastructure is as dumb as unsafe infrastructure and only enforcement and praying for culture change / personal responsibility for all road users.
We can do better on all fronts.
Its both/and.
Infrastructure is more important and I am strongly in favor of everything laid out in this letter.
Enforcement is also important. I’m not sure where some of these assertions come from Slaw, but I’m from a Northern European country with great infrastructure.
A few things to consider:
Many European countries spend more on policing and tend to have a higher number of police officers per capita than the US does.
Especially automated enforcement is a crucial part of making roads safer in Europe. The prevalence of speed cameras has all but removed speeding in many dense European cities. The reality of the built environment means that you cannot “infrastructure” your way to safety. There will still be stretches of road that allow cars to speed up too fast. People can still run red lights. Catching those offenders consistently is a crucial part of making the environment safer.
No one opposes enforcement. Let’s install traffic and red light cameras on every street. If someone runs a red light, they get a ticket in the mail.
But enforcement alone is not enough. People will still break the rules. We also need infrastructure changes.
Many intersections are dangerous even if everyone follows the rules. For example, having cars parked right up to the corner makes it hard for pedestrians to see around them and for drivers to see pedestrians trying to cross.
A simple measure like removing parking spots near intersections has been proven to save lives.
Yes. Exactly what I said
Yes, but without constant surveillance, people will still break the rules. Traffic calming measures like street diets, removing a few parking spots for better visibility, and adding traffic islands to shorten pedestrian crossings have all been proven to save lives.
However, implementing these measures often faces resistance. A traffic island, for example, elicited loud complaints because “drivers might hit it and pop a tire.” I suppose they think pedestrians are softer.
Yes… Who are you responding to?
And I agree that the most effective enforcement that pairs with safer infrastructure is automated and constant enforcement. Random speed traps don’t do anything. Though out of spite, I wouldn’t mind it if we had more of them!
@cportus. I’m addressing those who argue for “more enforcement” to avoid making our streets safer. They resist losing even one parking spot, despite the potential to save lives and prevent injuries.
@cportus
You said:”Especially automated enforcement is a crucial part of making roads safer in Europe. The prevalence of speed cameras has all but removed speeding in many dense European cities.”
Yes, that would be a great idea. However, Massachusetts does not permit the use of automated enforcement for traffic violations, including speed cameras and red-light cameras.
Stupid… just plain stupid.
Even if the state allowed it, it is doubtful that Cambridge would, because unfortunately, too many people in this city, and the City Council, believe that cameras invade privacy. What nonsense.
We have speed limits in Cambridge. No right hand turns on a red light. The only way to enforce these things is by automation.
The changes to intersections detailed in this letter have been proven highly effective. One example is Mass Ave and Beacon St in Back Bay.
It was known as one of the most dangerous intersections in Boston. A few simple changes greatly decreased the accident rate. Every intersection should be examined and modified, if necessary.
https://www.wbur.org/news/2015/09/01/back-bay-intersection-changes-bike-safety
Yet another accident at Hampshire and Portland, where a cyclist was just killed. We can do better. We can make intersections safer.
https://x.com/CambridgeCrash/status/1810373399597752604
And there were two crashes on 7/1/24. In both cases, the cyclist was taken to the hospital and the driver was cited for a failure to yield.
https://x.com/CambridgeCrash/status/1808199071305199924
https://x.com/CambridgeCrash/status/1808226869797331387