[tokenadult]

The death rate per miles driven in the United States has fallen dramatically in my lifetime, and I can remember when the annual number of deaths was much higher.[1] That said, the article makes a very interesting claim about attitudes that we should all follow to the end of the article for further discussion here: "There’s an open secret in America: If you want to kill someone, do it with a car. As long as you’re sober, chances are you’ll never be charged with any crime, much less manslaughter." My wife bike-commutes year-round (yes, even in Minnesota), and as I mention this among Facebook friends, other friends who are also bike commuters point out that car drivers can basically kill bicyclists in the United States with no legal penalty at all. That's not a good social environment for getting more people out for exercise and energy conservation by substituting bicycling for driving cars.

The history reported in this article is very interesting. There are a lot of contemporary photographs of changing American cities. The quotations from experts provide perspective on the visuals: "'If a kid is hit in a street in 2014, I think our first reaction would be to ask, "What parent is so neglectful that they let their child play in the street?,"' says Norton.

"'In 1914, it was pretty much the opposite. It was more like, "What evil bastard would drive their speeding car where a kid might be playing?" That tells us how much our outlook on the public street has changed."

Indeed. Are we really thinking carefully about how to spread the risk around, when so much of our living space is dominated by cars?

AFTER EDIT: The video link shared by pugz[2] in a reply comment elsewhere in this thread is not to be missed. Car safety standards have improved enormously in my lifetime, but those protect the occupants of cars better than they protect pedestrians and bicyclists who are hit by cars.

[1] https://en.wikipedia.org/wiki/List_of_motor_vehicle_deaths_i...

[2] https://www.youtube.com/watch?v=fPF4fBGNK0U

[nostromo]

I don't think the solution to unsafe roads is to throw everyone that is at-fault in an accident in jail.

I understand the frustration cyclists have with auto traffic, but we've seen a million times that harsher penalties don't always have the intended outcome. And we have a habit in the US of using prison as the answer for every malady.

Instead, I think the answer is better road design, separating cars and cyclists using dedicated lanes, and automated vehicles.

Also, according to this report, pedestrian fatalities are also on a long-term downward trend, even without correcting for increased traffic: http://www.fhwa.dot.gov/publications/research/safety/pedbike...

[revelation]

Sure, lets do all of that. But there is plenty of low-hanging fruit. Maximum of 20 mph everywhere you have pedestrians, certainly in any city or city center - the simple reasoning being that higher speed only reduces time to travel some distance linearly, but the forces at work in a potential collision increase quadratically. So this is a complete no-brainer tradeoff that is directly reflected in survival rates for pedestrians in ped-car crashes:

http://www.streetsblog.org/wp-content/uploads/2008/11_10/Fat...

(There need to be legal repercussions regardless. If you are involved in a collision as a car driver, you can not be allowed to be driving again the very next day, regardless of fault. People that operate multi-ton machinery producing enough kW to provide power for a whole block take on a massive gamble at the cost of other people, and so that in and of itself needs to be remedied legally.)

[freehunter]

I think the premise of the 20mph speed limit is good, but the way car transmissions are geared, doing a sustained slow speed could actually be dangerous as well. Doing 15mph through the parking lot at my office means I'm feathering the throttle quite a bit, and paying close attention to my speedometer and less attention to the things in front of me. I've never driven a car that "wants" to do 15mph or 20mph. I can't just keep the throttle at one position and stay at 20mph, the pedal, the foot controlling it, and the sense of speed related by the brain just aren't designed for fine adjustments.

The point being, I believe that setting low and strict speed limits could be counter productive and could lead to drivers being more distracted and less prepared to stop. If my hunch is right, there would be an increase in car-pedestrian accidents at exactly 20mph, and fewer at 10mph or below due to people being less likely to get their foot on the brake because of the effort required to keep the car moving at 20mph and watching the speedometer.

I've actually hit a pedestrian once. My light turned green as the person stepped into the crosswalk, and I hit them less than a second after I was stopped at 0mph. The pedestrian and the car were both fine with no injuries. Perhaps rather than dropping the city speed limit from the 25mph it already is down to 20mph, we should be focusing on getting pedestrians to cross in locations where cars are naturally going very slowly to begin with. I don't want to shift the blame of an accident from the perpetrator to the victim, but for pedestrian safety, crossing at a light (and following crosswalk signals) is worlds safer because the cars there will already be either stopped or going very slowly.

[sp332]

This sounds like an argument for introducing a separate class of "city car" to the USA. On a highway, I want a big comfy car that can be efficient at 55-65 MPH, and protect me in a collision at those speeds. In a city, I don't really need that.

[graylights]

That big comfy car that protects you from highway collisions accomplishes that by shifting the impact toward the smaller car. It doesn't increase overall highway safety, just steals it from others in your favor.

[kahirsch]

Large cars are also safer in single-vehicle collisions, which account for 65% of traffic deaths. There is some adverse effect on occupants of other vehicles, but this is much smaller than the advantage in safety from the large vehicle.

It is definitely not true that it "just steals it from others".

[SixSigma]

It definitely is true. In a collision the vehicles absorb the energy of the collision inversely proportional to their relative masses i.e. the lighter car takes the brunt of the crash.

Safety researchers are responding by using active protection systems: in vehicles equippped with v2v transcievers will stiffen or soften their crash protection systems based upon the relative masses of the vehicles.

[malandrew]

Of that 65% that is single vehicle collisions, what causes are responsible for most of those and where do most of those collisions take place? suburbs, rural, city?

I would imagine that the overwhelming majority of single vehicle collisions are in low density suburbs.

I would also imagine that the overwhelming majority of those deaths involved colliding with a stationary, immovable object, such as a tree or traffic pole. In such a case, smaller vehicles will much less kinetic energy, but still as much thought given to their safety designs, should be much safer.

[Dylan16807]

Big doesn't have to mean particularly heavy, and it's ridiculous of you to suggest that a couple extra feet of crumple zone isn't going to make impacts less forceful.

[Gracana]

What about all the safety features found in larger cars? What about single-vehicle accidents? I don't need to drive a logging truck to feel safe, but I'd much rather drive a new corolla than an 80s model. Solid pillars and increased crumple zones and multiple airbags make a car bulkier, but the safety improvement is much greater than a simple function of weight.

[collyw]

I would feel far safer on my bike being hit by one of these larger cars and their safety features.

[mikeash]

I don't think that's true. Consider the two extremes.

1. You somehow manage to get a cardboard box up to 70MPH. You strike another fellow pulling the same trick. How badly are the two of you hurt? I believe the chances are excellent that you both die immediately.

2. You somehow manage to get a container ship up to 70MPH. You strike another fellow pulling the same trick. How badly are you hurt? If you're strapped in and not near the front, I believe chances are excellent that you'll both walk (swim) away.

[seanmcdirmid]

Sounds like a tuktuk, many cities in Asia have them, but they mix with car traffic.

[mikeash]

When people decry using the prison system as a cure-all, it's usually for things like drugs or mental illness. In essence, why punish people for things that either don't harm others, or aren't their fault?

You're applying this to a serious crime, at least manslaughter if not murder. If prison isn't appropriate for that, what is prison appropriate for? I imagine an argument could be made that prison is never the right thing, but I don't think you're making that argument.

Harsher penalties don't always have the intended outcome, sure. But this isn't so much advocating a harsher penalty as advocating a penalty at all.

If I'm driving a car, what is my incentive to pay attention and try to avoid killing cyclists? Simple morals, obviously, but that doesn't work on everybody. If I'm a selfish asshole (lots of those out there) and I know I won't suffer legal penalties, why would I care about cyclists?

If you kill someone while driving a car and you are at fault, why should you not go to jail? That is the standard punishment in that scenario without the car, so why should adding a car make it go away?

[graylights]

Simple, "at fault" takes on a different meaning in driving. Often it's a cluster of mistakes that causes an accident, mistakes by either vehicle or even mistakes by someone not a part of the collision itself. If I'm 51% at fault should I go to jail?

But also everyone makes mistakes occasionally behind the wheel, most often nothing bad happens. And by everyone I mean every single driver without exception. That's why the law mostly operates with concept of Mens Rea, a guilty mind. Negligence only comes into play under a reasonable person standard and since we know those reasonable persons are also making same mistakes... it goes to civil court.

http://en.wikipedia.org/wiki/Mens_rea#Criminal_negligence :

"There is credible subjective evidence that the particular accused neither foresaw nor desired the particular outcome, thus potentially excluding both intention and recklessness. But a reasonable person with the same abilities and skills as the accused would have foreseen and taken precautions to prevent the loss and damage being sustained"

[DougWebb]

and you are at fault

That's the tricky phrase. How do you prove fault in a car collision, to the same degree that's required for a charge and conviction of manslaughter or murder?

In New Jersey, we have "no-fault" laws for insurance, which basically means that when there is an accident there's no way to prove that either party is more at-fault than the other, so legally neither is at fault. In practice, both parties are treated as if the accident is their fault, and both are penalized with higher insurance rates at minimum.

My concern with any law that holds drivers responsible for deaths in accidents they're involved in by charging them with manslaughter or murder is that the same logic will be applied: every driver in every accident where someone dies will go to jail, unless they're rich enough to bend the rules and escape the charge. That's unfair, because many accidents really are accidents, and many people who die in accidents are at least as responsible as the people who survive.

Besides, there are already additional penalties for drivers in these situations. Their insurance pays out substantially to the victims or their families, and the driver's insurance rates go up, potentially to the point that the driver is no longer insurable and they can't drive anymore. So the article is wrong in saying that drivers can get away with murder with no consequences.

[malandrew]

Increased insurance premiums aren't enough. That just punishes those too poor to handle the increased premiums. Everyone else just gets back on the road after the accident is "resolved" legally. A lifetime ban on driving or at least a ban of 10 years is far reasonable. Moving a multi-ton vehicle at speeds of ~40-70 mph is the ultimate privilege and it should be a privilege that is incredibly easy to lose.

All reckless driving citations should carry with them a minimum 1-5 year loss in driving privilege that no lawyer can get you out of.

[mikeash]

Why would the same logic be applied? Surely a conviction for vehicular manslaughter would be held to the same standard as a conviction for regular manslaughter, which is to say proven guilty beyond a reasonable doubt.

[DougWebb]

If guilt can't be proven for sufficiently to assess which driver's insurance should pay for the costs of an accident, how can guilt be proven for the much stricter requirements needed for manslaughter? I think this would be especially difficult if the courts are overloaded with all accidents that involve deaths, instead of only the ones where drivers today are being charged with manslaughter and murder. (eg: cases where there appears to be intent to kill, rather than just accidental killings.)

[SixSigma]

> If guilt can't be proven for sufficiently to assess which driver's insurance should pay for the costs of an accident, how can guilt be proven for the much stricter requirements needed for manslaughter?

Guilt could be proven for more cases but the cost is prohibitive. It is cheaper for all concerned to just say "we know what happened, let's not bother finding out how and just split the costs of repair", the drivers pay higher premiums for a while, they probably share the blame anyway and both will drive more carefully from now on.

When one of the parties could go to prison for life, society decides to accept the burden and does a full investigation.

[mikeash]

There are many cases where guilt can be proven sufficiently, both to assess insurance responsibility and legal culpability.

Just because New Jersey has decided to punt on the whole question doesn't mean it's impossible to answer in all cases. Sure, there will be cases where somebody kills somebody else due to negligence and they get away with it because it can't be proven. People sometimes go free after murdering people with guns and knives because the crime can't be proven too, but that's not a reason to give up on prosecuting all murders.

[7952]

Only a tiny majority of drivers will ever wilfully put cyclists at risk. Most people will genuinely believe that they are good drivers and nice people because that is exactly how everyone they know also behaves. There is simply no strong social pressure to be careful beyond a basic minimum standards. Cyclists are seen as breaking a basic social convention.

But as a cyclist, you realise how basically selfish people are and how unfair traffic accidents are. Motorists have an unrealistic idea of "fairness" and would feel that excessive penalties would be "unfair" when they made an honest mistake that is commonly made.

Personally I think that the penalties of bad driving should match the unfairness of the effects. Do not punish 99% of people who get a ticket. The other 1% are picked at random and get a very large fine (say $10,000).

[malandrew]

Every driver on his cellphone or sending text messages is willfully putting cyclists at risk. Considering how common both those scenarios are, I would say that many if not the majority of drivers put cyclists (and pedestrians, other drivers and themselves) at risk on a regular basis.

[Houshalter]

There is a difference between human error and deliberate crimes.

[thaumasiotes]

> When people decry using the prison system as a cure-all, it's usually for things like [...] mental illness. [...] why punish people for things that [...] aren't their fault?

Because they're dangerous to society. Whether or not it's their fault, there are some people we can't have on the streets. If you plead insanity in a murder case, we don't want to release you, we want to imprison you in an asylum instead of a jail.

[supersystem]

De Blasio announced an initiative[0] a couple of weeks ago inspired by the Swedish "vision zero"[1:].

[0] http://www.nytimes.com/2014/02/19/nyregion/de-blasio-unveils...

[1] http://www.minnesotatzd.org/events/2013/strategicvisioning/d...

[2] http://www.unece.org/fileadmin/DAM/trans/roadsafe/unda/Swede...

[3] http://www.visionzeroinitiative.com/

[megablast]

> Instead, I think the answer is better road design, separating cars and cyclists using dedicated lanes, and automated vehicles.

One way of keeping the status quo is to suggest things that are incredibly expensive and time consuming will never be done.

Making driving more expensive however, does reduce traffic and fatalities.

[malandrew]

    "Making driving more expensive however, does reduce traffic and fatalities."

Citation? I'd love to see exactly how this correlation looks.

[bunderbunder]

I'm actually not fond of separating cars and cyclists using dedicated lanes. My experience has been that roads with bike lanes tend to be more dangerous - cars stop paying as much attention to bikes, so you get a lot more instances of them forgetting you're there and making a right hand turn right in front of you. Not to mention that they often force cyclists to ride within swinging distance of car doors - which is a far more popular way to injure cyclists than actually running into them is.

My preference would be to have designated bike thoroughfares where cars are permitted, but the speed limit is kept to something low enough that cars and bikes are going at approximately the same speed so bikes can ride in the lane without (reasonable) motorists feeling obstructed.

[benjiweber]

Bike lanes are often more dangerous, but that is generally due to poorly designed bike lanes .

There's lots of informative explanation on this blog http://www.aviewfromthecyclepath.com/search/label/subjective...

[ygra]

Since bike lanes integrated into the road are common enough, I think when people demand dedicated bike lanes, they mean »not part of the road, but instead physically separated«. And those are indeed better, safety-wise.

[malandrew]

At the very least loss of privilege is warranted. If you are responsible for a motor vehicle accident, you should not be allowed to drive again. This at least provides some permanent hazard to being careless. If you can cause an accident, clean it up because you can afford a good lawyer and be back on the road with business as usual, you're far less likely to be careful.

[lutorm]

Your edit is an important point: As far as I am aware, there are no safety standards or ratings for how likely a vehicle is to injure occupants of other vehicles, let alone unprotected pedestrians or bicyclists. If anything, the focus on occupant safety to the exclusion of others has supported the "race to the top" in terms of vehicle mass and size.

[sukuriant]

Modern European tests include pedestrian safety ratings:

https://www.theaa.com/motoring_advice/euroncap/crash_tests.h...

[pugz]

The Australasian New Car Assessment Program provides a website howsafeisyourcar.com.au. A recent interesting addition is precisely what you say: it measures pedestrian safety in collisions with specific car models. For example, the 2013 VW Polo:

http://www.howsafeisyourcar.com.au/2013/Volkswagen/Polo/Tren...

"The bumper provided mixed protection for pedestrians' legs. The front edge of the bonnet was marginal to poor. The centre of the bonnet offered good protection to a child's head but poorer protection towards the edges. In most areas likely to be struck by an adult's head, poor protection was provided."

EDIT: Seems three of us posted in the same minute. :)

[saraid216]

Yup.

http://www.kidsandcars.org/back-overs.html

[sokoloff]

Holy crap. That's the most heart-breaking web page I've ever read. I literally had to stop reading after just a couple so that I'd have any chance of sleeping tonight.

[DougWebb]

I don't know this for certain, but I think there are at least some safety standards. One thing I've noticed while driving around my town is that school buses have an odd design. In most vehicles the tires are placed fairly close to the corners of the vehicle, which I assume increases stability and improves maneuverability.

But in school buses the tires are set far back from the corners, especially the rear tires. In a large bus the tires are 5-10 feet from the rear bumper. School buses also tend to have a very high ground clearance. I assume that these design features are for safety: if the bus hits a child while starting to move, either backward or forward, the driver will have a lot more time to react to the thud against the bumper before the tires reach the child. It takes a moment for the driver to notice the thud and move his foot from the gas to the brake; if the tires were at the corners that moment is all it would take for the bus to roll over the child.

So, I think this is an example where the vehicle is designed explicitly to protect pedestrians, even though it probably makes the vehicle less stable and harder to maneuver.

[dredmorbius]

That's a damned interesting observation on school bus design, though I'd really like to see if that is in fact the case.

The other thing that moving the wheelbase in does is to make the vehicle more maneuverable overall -- it has a shorter turning radius.

Edit: And so far as I can tell, turning radius is in fact the principle concern. There are numerous versions of school bus design specs online, typical: http://ww2.mackblackwell.org/web/research/ALL_RESEARCH_PROJE...

[mikeash]

I was recently in Beijing, where the supremacy of the automobile over the pedestrian is just about taken to its logical conclusion.

The US puts cars on a high footing compared to pedestrians, with pedestrians restricted to crossing roads at certain points, generally better signaling and routes for cars, etc. Yet cars are still required to yield to pedestrians when e.g. turning or at a marked crosswalk.

In Beijing, the idea of yielding to pedestrians when turning doesn't seem to exist. Crossing a major intersection, even with full signaling, is an adventure and an exercise in attempting to look in all directions simultaneously. It's even worse than you might think, because the sidewalk often doubles as a parking lot, so somebody might by trying to drive up where you're waiting, or worse, come up behind you! And you are expected to yield. It got to the point where I measured the difficulty of any walk not in terms of distance but in terms of how many intersections I'd have to cross.

Beijing is pretty walkable overall but at the same time this can make it extremely unpleasant, depending on where you are. I wonder if that might not be where the US is heading if we're not careful about it.

(The Chinese are strangely cavalier about cars in general. Essentially nobody wears seat belts, for example, and drunk driving is common. No surprise that their death rate per vehicle is three times higher than the US's. Seat belts would be such an easy way to reduce that. It's hard to understand.)

[saraid216]

> In Beijing, the idea of yielding to pedestrians when turning doesn't seem to exist.

It's odd. I remember when people were bemused by the bicycle city, and yet I also remember when I learned that the way to cross the street in Beijing was never to indicate that you could see a car coming, because then they'd assume you'd stop. Instead, you watch for traffic out of your peripheral vision and keep your eyes forward so that they'd feel you couldn't see them.

I wonder if that's still common wisdom.

> It got to the point where I measured the difficulty of any walk not in terms of distance but in terms of how many intersections I'd have to cross.

One of the possibly most frustrating moments I had was recently in downtown San Jose. My cousin and I (and others) had finished lunch and left into an alleyway. Her four-year-old son cheerfully ran out to the sidewalk, and we panicked because it would have been so easy for him to keep going out to the street itself.

There was no harm (he indignantly pointed out that he had not gone into the street), but it perplexes me why we find these kinds of risks acceptable. Every block, we've essentially put down a death trap for the young. I could understand it if this was the edge of the untamed wilds and we couldn't really do anything about the wolf pack in the area, but... we built these cities.

> The Chinese are strangely cavalier about cars in general. Essentially nobody wears seat belts, for example,

I've noticed this is prevalent elsewhere in Asia: there are often similar claims made about Mumbai or Bangkok.

[mikeash]

> I wonder if that's still common wisdom.

Personally, I think that not indicating you can see cars coming and crossing that way would be a good way to get killed. However, I haven't actually tried it. I haven't observed the natives using this tactic either as far as I can tell.

[WildUtah]

It's ironic. Peking was considered a beautiful city of clean air (except dust storm season) full of bicyclists during the oppressive communist era.

Now in the time of industrial crony capitalism, it's famous as the most polluted capital in the world. The streets are dangerous and cars have replaced bikes making the powerful insiders even more oppressive against individuals because now they can kill people with impunity.

Surely this is better than communism, but unlike the Taiwanese the Red China capitalists have imported every Western vice tenfold.

Only during the Dèng era was Peking open to new ideas and businesses, full of pedestrians and bicycles free from fear, and able to look up to clear skies.

[Zigurd]

Yes! Having seen the "Nixon in China" images of a surging wave of bicycles at every green light at an intersection, I was shocked to see in person how completely cars have taken over, and how badly Beijing drivers treat cyclists, driving within inches of them. The only reason there isn't carnage on every block is that nothing moves faster than 10mph.

Taipei, meanwhile, is overrun with gas scooters. These pollute worse than cars. The mainland Chinese could not allow gas scooters because of their pollution output. Every office in Taipei has racks in the stairwells for hanging up your very necessary rain poncho.

A great irony is that the Chinese have mastered the ability to make cheap electric bikes. Some Chinese use them. Very few Taiwanese, because they can't keep up with gas scooters. But you'll see lots of Chinese electric bikes in New York.

[vacri]

In Beijing, the idea of yielding to pedestrians when turning doesn't seem to exist.

In parts of the US, too. Travelling across the US, there were a couple of places where as a pedestrian I had the right of way (pedestrian crossing light illuminated) and I actually had to stop crossing otherwise I would walk into the side of the car turning right in front of me, cutting me off.

I also noticed wandering around some inner suburbs of Austin that it was hostile to pedestrians - some houses just decided to stretch their gardens all the way to the road, blocking the footpath. Pedestrian traffic had to step out onto the road, in one case a secondary thoroughfare, to get past the property.

[freehunter]

Is a lawn legally allowed to block a sidewalk? I know the right-of-way stretches well past the boundaries of the road. I've always walked on the edge of people's lawns when there's no sidewalk, because I'm fairly certain it's legal to do so. If there's no sidewalk, pedestrians walk on your lawn within a few feet of the road, and there's nothing you can do about it.

One thing I would never do is step into the street to walk, regardless of how angry a homeowner might be about it.

[happyscrappy]

Pretending kids were better protected in any context in 1914 is absurd.

[pyrocat]

That quote was specifically about attitudes towards cars, not statistics for mortality rates.