Can an element of danger make life safer?

 

photo from ncc.gc.ca

In Ottawa-Gatineau, the standard rules of the road apply to recreational pathways. People cycle, roller-blade or walk on the right. Slower traffic keeps to the far right-hand side of the path. 

 

In theory, it's safe. Everyone knows which side of the pathway they are supposed to be on. Having a yellow line in the middle of the pathway reduces the chances that cyclists will plow into each other, causing a nasty accident. 

But the two-lane system gives cyclists the illusion that they are driving on a cycle-highway. Like a wide road, it creates arrogance of space, and a feeling that it is safe to drive fast. In theory, there is a 20 km per hour speed limit. In practice, the speed limit is often ignored and never enforced. In theory, cyclists yield to pedestrians. In practice, some (though, in fairness, not all) cyclists ring their bells and expect pedestrians to get out of the way. 

Multi-use pathways are not as safe as people think they are, as described in this survey (ungated here):

Results to date suggest that sidewalks and multi-use trails pose the highest risk [of injury to cyclists], major roads are more hazardous than minor roads, and the presence of bicycle facilities (e.g. on-road bike routes, on-road marked bike lanes, and off-road bike paths) was associated with the lowest risk. 

There are problems with a number of the papers on bicycle safety that I looked it, for example, they do not properly control for the fact that more able, confident and experienced cyclists are more likely to be cycling on roads and less likely to be cycling on multi-use trails. But the studies at least suggest that safety on multi-use pathways is an issue. 

There is an alternative: making the pathways a little bit more dangerous. In Exeter, in the south west of England, multi-use trails are divided in two -  cyclists on one side, pedestrians on the other - as shown in the picture below.

It doesn't seem as safe. After all, it's not obvious which side of the pathway one is supposed to be on. Cyclists ride on the left-hand side of the pathway going one direction, and on the right-hand side of the pathway going the other. One has to pay attention, concentrate, and follow the signs to work out which side to be on. When two cyclists approach each other, they have to slow down, otherwise they might just crash into each other.

But is the UK rule really more dangerous? My search for hard data on the comparative safety of different pathway designs turned up nothing. This US review of the literature on minimizing conflicts between trail users advocates painting a centreline and encouraging users to travel on the right, however it does not discuss alternative methods of dividing a single path. While searching I discovered this study on ways of promoting cycling, and the copenhagenize.com blog, but no analysis of the UK v. Canadian/US pathway rules.

However, in preparation for my latest post on the Globe and Mail website, I've been reading about traffic safety, especially Tom Vanderbilt's book Traffic. One of the things that's dangerous about driving is that it seems easy. On some roads, it's boring. People become complacent and stop paying attention. That's when accidents happen. 

An element of danger means people to pay attention. Paradoxically, that can make life safer.

Comments

[Bob Smith]

This was always the knock on mandatory seatbelts and airbags, by making driving "safer" (at least for the driver) they remove the incentive to drive safely. The alternative suggestion was to install mandatory spikes in the center of the steering column and watch car accidends plummet.

[Frances Woolley]

Bob - yup the spike example is the classic one. It would have the additional safety-enhancing effect of killing off the more dangerous drivers!

On seatbelts and airbags, I'm not so convinced. People are really lousy at scientifically estimating the probability of having an accident and getting injured. I suspect most just forget airbags are there - or drive more carefully because they're worried about the airbags exploding accidentally. I think we respond much more to things that activate more primal parts of our brains, e.g. narrower roadways that make us feel as if we're crowded and have to be careful.

Part of this is because a truly remarkable amount of driving is automatic. E.g. when I'm driving at night on a quiet road somewhere in Australia or South Africa, and I see headlights approaching in the distance, my brain automatically puts those lights on the right hand side of the road, and I feel as if the car is driving right towards me in my lane (because, it being Australia or SA, I'm driving on the left). Logically, it isn't, but that's the way that my brain processes the information.

[Shangwen]

And an element of safety can make life more dangerous. One of the limitations with safety regulations is that they are often communicated with great confidence, inducing overconfidence in the intended beneficiaries (see here and here for two studies on helmet laws and injuries).

As an anecdote, years ago I drove for two years with a dysfunctional speedometer (never rose above 20 km/h). It was two years of apparently breaking the law, but it was also a two-year respite from speeding tickets because I was much more mindful of estimating my speed.

[Phil]

I ring my bell and expect pedestrians to get out of the way only when they're taking up both lanes or drifting laterally without noticing. That's usually couples, groups, and people with dogs.

If they're obviously limiting themselves to one lane, I just pass in the other.

I have no evidence, but my gut says that unpredictable pedestrians are the cause of most of the danger. There are cyclists going fast, but they're usually going in a straight line, staying in the same lane, and signalling their intentions.

I'd love for the NCC to put up signs reminding everyone (cyclists, pedestrians, and geese) to stay in the right lane.

[Bob Smith]

"On seatbelts and airbags, I'm not so convinced."

Nor am I, if only because, while seatbelts and airbags may save your life in an accident, having to rely on them is still an unpleasant experience. I suspect, to the extent people give any thought to the risks of driving, they see a crash as being "bad" and to be avoided without any regard to whether you'd only be slightly hurt or killed.

But the argument seems more plausible in other contexts. For example, you heard the same arguments with respect to snow tires when Quebec made them mandatory a few years ago (i.e., people won't slow down in bad weather, they'll be over-confident, yada, yada, yada). Ditto for the preference for SUV's because they "feel" safe. To the extent that people believe (rightly or wrongly) that they are better able avoid the "bad" event (because they have snow tires or four-wheel drive), they may not take other precautions (how often have I heard "I don't need snow tires, I have 4-wheel drive"? And I confess, I've driven in weather with snow tires that I wouldn't go out in all-seasons - that's why I got snow tires). In some sense the fact that people are lousy at estimating risk works both ways.

[Jacques René Giguère]

Income effect is rarely bigger than subsitution effects. If seat belts reduces mortality by 5 times, it's almost impossible to drive more than five times more recklessly.
I have to go back to the exam pile so no time to research the exact numbers ( sorry Determinant...) but it seems that snow tires reduced accidents.

[Paul Friesen]

I also wonder about some of the newer industrial safety initiatives. This has changed a lot in recent years. There is a real push on to make all industrial machinery completely idiot proof. Of course, much of this is both expensive and in the way, causing people to try to disable it or work around it. I also wonder if it might make people too complacent.

[jt26]

When I lived in Germany, during a conversation, one German asserted that driving in Germany wasn't safer than driving in France, even with all the German rules (which everyone obeys; + more rigorous driver training etc.) vs. the French (who always broke the rules, but was always on the watch for everyone else breaking the rules). Not sure if the facts bears out his assertion, but it is similar to your proposal.

[Frances Woolley]

Phil: "I ring my bell and expect pedestrians to get out of the way only when they're taking up both lanes or drifting laterally without noticing."

Cyclist/pedestrian interaction is really hard. Cyclists go so much faster than pedestrians, and pedestrians are unpredictable, so there is a real risk of one injuring the other. Pedestrians often won't hear a cyclist if the cyclist doesn't ring his bell, but pedestrians get annoyed if the cyclist does ring his bell. I don't think we've got it sorted out in Ottawa by any stretch of the imagination. The pathways are, for the most part, safe and comfortable for cyclists to ride, but are really not that pleasant for pedestrians.

Imagine what it would be like if the British rules were in place on the Ottawa River pathway or the pathway alongside Colonel By Drive. It would be lot harder to cycle quickly. But because the pedestrians and cyclists wouldn't be sharing the same lane, I would think there would be less conflict. The pedestrians could just amble along taking up an entire lane, until they met some other pedestrians. At that point *because they'd see the other pedestrians coming* they'd be able to get out of the way without conflict or agro.

Some parts of Vancouver/Burnaby that suffer from serious CAnada Goose problems have started to allow dogs to roam free (and chase away the geese). I don't know if Ottawa's hungry and homeless could be persuaded to hunt down Canada goose? I would run them down but I'm worried about wiping out/damaging my bike.

[whitfit]

I think we should do the same with cars in the city!

Give them half of the street, let them sort it out amongst themselves and leave cyclists and pedestrians with their own spaces. To be fair, pedestrians already have their own space, but it is shared with patios, signs, post boxes and cars pulled onto the curb/driveways.

This is another discussion, but I read your column in the Globe today. I also read the book "Traffic" and enjoyed it. I think the answer is robot cars. Humans are not well wired for driving. I can't wait until robot cars take over, and they actually obey speed limits (which I care about more for urban and neighborhood streets) and don't hog the middle lane on the highway, or run over pedestrians and cyclists when turning.

[Frances Woolley]

whitfit -
"we should do the same with cars in the city"

I wonder how many cycling advocates there are out there who are just fine with the current design of the recreational pathways but who loathe having to ride on roads governed by essentially the same rules. It's all about whether one is in the majority or the minority, the fast or the slow means or transport.

"I think the answer is robot cars."

They're getting closer and closer - cars can do more and more things automatically e.g. slam on the brakes when you're about to hit something at high speed, parallel park. The danger point is the car that can do *almost* everything by itself - leaving the human driver free to gaze out the window or chat on her cell phone - but can't cope with sudden unexpected situations.

Thanks for commenting on the G&M post here - I don't ever look at those comments.

[Chris J]

Hmm... I am about to give the safety talk at work today. (Seriously)

The four mental states that most contribute to errors are rushing, frustration, fatigue and complacency. I can see how a cyclist going fast on a trail could have all four going on at once: trying to go fast; frustrated because they can't; tired because it's the end of the ride; and complacent because they think it's safe.

Complacency is a big issue. Although in an industrial setting complacency is mostly likely to come about as "I've ridden this trail every week all summer. I know what's around the corner."

[whitfit]

"I wonder how many cycling advocates there are out there who are just fine with the current design of the recreational pathways but who loathe having to ride on roads governed by essentially the same rules. It's all about whether one is in the majority or the minority, the fast or the slow means or transport."

I agree, though I would add two things:

(1) I used to hate riding a bicycle on multiuse pathways because they are not good for riding fast, and I think that one needs to abandon the idea that these paths are useful for such riding; and

(2) I have become a much more accommodating and less aggressive driver, cyclist and pedestrian because I realize that the framework of importing the highway mentality to all forms of transit is foolish - each sidestreet, pathway and sidewalk becoming a mini-highway with rigid rules, emphasis on speed and aggressive interaction makes the world dangerous and unpleasant. Streets that aren't highways should really be dominated by people, not the fastest mode of transit that are using the road/path/whatever (within reason). See reason #1 - I am more relaxed riding on such paths now - I don't expect to crank out a consistent high speed (though I am bothered still by people walking on one side of the path with their dog leash stretched across to their dog on the other side of the path, setting up a "clothesline" and by people that are standing around on the path oblivious to the fact that they are on the path (why not stand around in a group on the grass beside the path???)).

With respect to the driving, and your G&M commentary, the one thing that I found after reading "Traffic" is that I try to focus much more when I do drive. It is mentally challenging to focus for extended periods of time, especially as speeds creep up beyond 100km/h. I am also convinced that most people on the road are only putting in the minimum amount of focus required. I am not surprised that people that drive a lot every day (especially on busy roads like the 401) don't focus as much as they should consistently - if you spend a couple hours each day doing that, you'd seriously reduce your mental reserves.

[Dan]

I thought the bell was more "I'm passing you, don't jump to the side for no reason" rather than "Get out of my way."

[Leo]

"I think the answer is robot cars."

As far as my safety as a cyclist is concerned, what would be useful would be robot car doors.

[Frances Woolley]

Dan "I thought the bell was more..."

This raises an interesting issue - the difficulty of communicating with bells and horns. Once I spent a day driving a prominent Indian economist around Barbados. She was unimpressed by the quality of my driving (quite unfairly - I only wandered over onto the right hand side of the road two or three times), but the only time we came close to coming to blows was over the use of the horn. As someone used to driving in India, she couldn't understand why I wouldn't just give people a little tootle to announce my presence - it seemed to her to be the only polite thing to do. As someone used to driving in Canada, I regard the horn as something to be used only in extreme circumstances - near-collisions, the Sens winning the Stanley Cup. I tried to explain to her "I simply can't do that." She just couldn't see it.

This of course raises all of these issues about trust, social cohesion and traffic accidents - there's a reasonable sized literature out there on it, IIRC.

[Stephen Tapp]

Hi Frances,

You should pitch this to Freakonomics` Stephen Dubner for an upcoming podcast.

[Chris McLeod]

There are several places where there already are two paths. One nearer the river or canal and one nearer the parkways. We could easily designate one for pedestrians and one for cyclists.

[Neil]

Several Points:
1 - I remember as a kid some multi-use paths in Edmonton had painted markings indicating 3 sections: pedestrians, roller-bladers, and bikes. In the decades since, these have all changed over to more standard center-line markings. I doubt it was about collisions, I strongly suspect it was because nobody followed the markings. They were only wide enough to travel single file, so people just ignored the markings and had a more social outing because of it.

2 - When it comes to multi-use paths, increasing people's sense of danger is likely to just make them not use them. Few Canadians view them as essential to their lives, and if they don't feel safe on them, they'll sit empty.

3 - While the safety study you note has a lot of good points in its methodology, it has a few failings. One of them is not differentiating between road-side multi-use paths (bike-permitted sidewalks) and off-road multi-use paths. The former have more risk with crossing intersections and typically more obstacles, while the latter are likely more like off-road bike paths.

[Frances Woolley]

Leo - robot car doors - Definitely! I hope some car engineer is reading!

Stephen - thanks.

Chris "two paths." The physical separation solution does seem to be the one that makes people happiest, typically - that's essentially what's been done on the Stanley Park Seawall in Vancouver, i.e.. the cycle part of the route has been raised so people can't wander over into the pedestrian part. Unfortunately there are few places in Ottawa where there are separate paths - the Queen Elizabeth side between just north of Pretoria and close to the NAC is one, but pedestrians amble along on the cycle path anyways because the pedestrian path is narrow and hard to get to.

Neil, interesting about Edmonton. Perhaps someone else from there will know more. On your point two - increasing cyclists' sense of danger may reduce pedestrians'. But, yes, paths being under-used is definitely an issue.

[Bean]

I was taught (in a class on radiation protection, of all things) that when, e.g., seatbelts were introduced in cars, everyone just started driving faster, and so were operating at the same level of risk of death as before seatbelts. That would be a completely subconscious calculation.

I notice the illusion-of-safety thing for cages on hockey masks - I play shinny with a group that includes old guys with no cages on their helmets - and I'm less cautious about getting my stick a bit high when trying to manoeuvre around the young guys with cages on, because I know I won't hurt them anyway.

In Italy, which is well-known for chaotic driving conditions, everyone drives as if they expect chaos. They are paying way more attention to all the odd things - pedestrians threading their way across the street, oddly parked cars, people about to cut you off. Accidents happen when someone doesn't follow your expectations - e. g., a pedestrian jumps into the road between intersections - if you have no expectations for "good behaviour" from anyone else then you are a lot more aware of your surroundings and react appropriately. So in spite of the apparent chaos I felt relatively safe as a passenger.

[jb]

Have you ever noticed that anybody driving slower than you is an idiot, and anyone going faster than you is a maniac?

My grand-dad passed away while driving on a motorway and he had a seizure, and paralyzed, could'nt move a muscle. Hit a motorway flyover. Sounds instant, but he was 72 and driving at 20 miles an hour. He saw that flyover coming towards him for eleven minutes. We all had time to bail out.
My uncle even went back for his jacket. I got the Tom-tom, but it's what he would have wanted.

[whitfit]

With respect to the safety from expective chaos argument, I have often thought that the best way to stay safe on my bicycle is to try to appear like I am crazy and might do anything at any moment, but to actually ride in a very predictable way. I haven't worked out how to create the illusion of craziness.

Also, further to that, there is the example of eliminating road signs in Holland, and theory and evidence that the resulting "anarchy" actually makes everyone safer:

http://www.spiegel.de/international/spiegel/controlled-chaos-european-cities-do-away-with-traffic-signs-a-448747.html

I am not sure if this is true everywhere - for instance, I have travelled in Malawi and China, both places that seem to have few road rules and don't make me feel safe!

[Frances Woolley]

jb: which reminds me of one of my favourite jokes "I want to die in my sleep like grandpa, not screaming and terrified, like his passengers."

whitfit: " I haven't worked out how to create the illusion of craziness."

When I was in Toronto last year I saw a cyclist dressed in a really crazy get-up - don't remember whether it was a pink lacy tutu (on a male cyclist), or a disco glitter helmet, but it was something similar. I think if you put aside any notions you might have of shame and/or modesty, you'll be able to come up with something!

Bean - see Jacques-Rene's comments above on this. He argues, and I think he's right, that it's almost impossible to drive fast enough and recklessly enough to counter-act the safety effects of wearing seatbelts. One of the persuasive stats on this, I think, is the organ donor numbers - see e.g. the section on supply and demand in this article http://www.guardian.co.uk/world/2013/jan/09/mass-donor-organ-fraud-germany.

[Jacques René Giguère]

Frances:a further rejoinder.
3 examples about road safety.
I've been travelling a lot on the QC 138 for the last 35 years. Compared to what it was, it is now, in good weather, a very easy road to drive.Some of the worst parts have been ameliorated, going from two lanes to 3(passing lane uphill)and even 4 and steep descent have been smoothed out with breathtaking carving into mountainsides. Instead of slowly winding our way at 70-80 km/h, some of thes downhill can (and are usually) taken at 130-140. When I look either up or down at the former road, the only thought is "I was young and foolish but I drove on that at night in a snowstorm?!". Though we go twice as fast on those segments, the ambulance at Port-Cartier is now waiting at the station instead of the bottom of the Rivière-Vachon curve as it used to...

In 1994, in Les Éboulements in Charlevoix ( the place Nick want to visit) a bus crashed in the Grande-Côte, killing 52. Although the cause was badly maintened brakes and the locals shoed on tv how school buses were driven there 4 times a day safely, the public outcry ( from outside the area) led to a rebuilding. It was feared that it would be tempting and that accidents would increase. At first, people drove fast but soon saw that 140 on a cliff above the St_Laurent was not a good idea and the place is safe.

Last summer, in Arizona, I visited the Kitt Peak National Observatory
http://www.noao.edu/kpno/
You get there via AZ 386. You climb 5800ft in 12 miles at a constant 8% grade. The road is carved into the sides and essentially has no security barriers. I've never seen such careful driving in my life...

[Bob Smith]

"Bean - see Jacques-Rene's comments above on this. He argues, and I think he's right, that it's almost impossible to drive fast enough and recklessly enough to counter-act the safety effects of wearing seatbelts. One of the persuasive stats on this, I think, is the organ donor numbers."

I wonder if there are any statistics the effect of seat belts on injuries to people outside of cars? Jacques point, that you'd be hard-pressed to drive so recklessly to offset the benefits of a seat belt is a fair one, but only in respect of drivers and car passengers. It wouldn't take much, though, to increase the harm to pedestrians/cyclists, who don't have the benefit of seat belts/air bags.

I like the use of organ donations as a proxy, although it only speaks to the severity of accidents, rather than the accidents themselves (what's the relative disutility of a fatal accident versus one that merely wrecks your body? I don't know), and suggests that introducing seatbelts/airbags has a hidden cost - i.e., increased suffering for would-be organ recipients.

It's dark, but to the extent organs can be "re-used" doesn't that reduce the social disutility of a fatal car accident? Heck, arguably a fatal car accident, where the victim's organs might save the life of someone else (or perhaps multiple people), might be preferrable to one in which the victim barely survives, leaving both the victim facing a life of misery and the would-be organ recipient facing death? Perhaps the rule should be that you only have to wear seatbelts if you're not young and healthy.

[genauer]

Here in Dresden we have no separation lines on the trails / bike lines along the Elbe, neither direction nor bike/walkers
(From Czech to the North Sea) and frequented by thousands on the weekends in Summer in Dresden. Both cyclists and pedestrians know they have to arrange, and I see this as working well.

What we have in 2 other german states, is still a lot of alley trees, which still kill more people per capita than gun shots in the US, but are defended religiously. They look innocent, but hitting them at normal speed is very often deadly.

Green alley trees vs 2nd amendment rights are weighted very differently there and across the pond. Comments from Canada?

Finally, it is still freezing cold here, but even if it is very icy, when you watch your feed walking, some folks ride the bike, and if one crashes, hard, full body contact with the surface, at full speed, and you tell the guy 50 meter behind, it is the 3rd on just one kilometer, you see, one gets the comment "but I ride carefully" ... at 15 - 20 km/h : - ) How do you help this kind of people?

Even the easy observation of clear and present danger doesnt change some people.

[genauer]

looking at the biomedcentral link above, I would like to point out, that there seem to be massive differences between nations, that if you calculate death rates (for Netherlands) per hour, and not per kilometer, the numbers for bikes and cars are very comparable , until age > 65 for bikes, and the age dependence follows closely general mortality numbers.