Review of Charles Komanoff's White Paper,
Getting Motorists to Share the Road,
for the National Bicycle Safety Conference of 2000

home                    safety

1 Introductory Paper: "Getting Motorists to Share the Road"

Charles Komanoff was the leader of section Educating Motorists to Share the Road of the National Bicycle Safety Conference of 2000. He wrote the paper on "Getting Motorists to Share the Road" that served as the initial outline for the work of his section in this conference. This is a critical review of that paper.

Komanoff's paper is not well organized. This review starts with a set of quotations from it, in the approximate order of the paper, to serve as a reference for the evaluation.

2 List of Quotations to be Discussed

2.1 Affirm the Right of Citizens to Ride Bicycle in Traffic

Komanoff asserts that "American cyclists cannot be certain that their legal right-of-way will be observed by drivers, enforced by the police, or upheld by the courts. In [Komanoff's] view, the practical lack of this right is the single greatest impediment to widespread cycling and cycling safety. ... The first step to help cycling thrive is to affirm the right of citizens to ride bicycles in traffic."

2.2 Improving Cycling Safety Must, First and Foremost, Address Driver [Motorist] Behavior.

"Efforts to substantially improve cycling safety must, first and foremost, address driver behavior."

2.3 Case Study: New York City

Fatality rate per trip, 3 times for cyclists than for motorists.

"Fewer than 1% of streets have bicycle lanes."

"New York police rarely enforce the cyclist's right of way."

Cyclists adopt unlawful tactics in search of self-preservation, not to break the law.

2.3.1 NYC Fatalities

By Komanoff's analysis, "driver actions were more than three times more likely than the cyclist actions to be at fault in a crash that killed a cyclist."

Motorist overtaking cyclist is a major type of fatal car-bike collision.

Police reports are not accurate; we need more Cross type studies.

Komanoff concludes that "The NYC PD attributed 3/4 of fatal bicycle-auto crashes to cyclist error, while [Komanoff's] study reached the completely opposite conclusion."

2.3.2 NYC Example

Here is Komanoff's single example of a car-bike collision caused by the motorist. "One fatal New York City crash involv[ed] an 80,000 pound tractor-trailer turning left into the path of a commuter cyclist proceeding in the same direction, the New York City police record states that the `bicyclist struck left rear side of vehicle' rather than `the driver turned left into the path of the cyclist.' This very likely led the State Motor Vehicles Department to miscode "Bike's Error-Confusion" as the primary apparent factor in causing the crash, thus blaming the lawfully proceeding cyclist rather than the truck driver."

2.4 Praise of Dutch Urban Bikeways

Referring to Holland, Komanoff praises the "excellent cycling infrastructure" whose "road engineering gives bicycles a high priority."

2.5 Praise of Dutch Traffic Laws

Dutch and German laws "require motorists to anticipate unsafe walking and cycling." If a car-bike collision involves a child or a elderly person, "the motorist is usually judged to be entirely at fault." "When a crash is caused by an illegal move by a cyclist or a pedestrian, the motorist is almost always judged to be partly at fault."

2.6 Increasing the Number of Cyclists

Komanoff asserts that increasing the number of cyclists on the roads will make motorists act in safer ways toward them.

2.7 Require Motor Vehicles to Yield to Non-Motorized Vehicles

"Motorists should be required by law to yield to bicycles, and bicycles in turn to pedestrians, in the absence of any overriding traffic rule or directive, such as a traffic signal. ... Adherence to this NYC might have averted as many as half of the cyclist fatalities in NYC between 1995 and 1998."

Supposededly, but wrongly, attributed as primary recommendation by the Toronto coroner in a study of fatal car-bike collisions.

2.8 Elevate the Status and Treatment of Cyclists by Police and the Courts

"That the legal system treats cyclists as inferior to [motorists] is an article of faith among cyclists."

2.9 Traffic-Calmed Zones

"Residential neighborhoods to foster safe walking, cycling, and playing by children."

2.10 Treat Bicyclists as Inferior

Sounds like the exact opposite to what Komanoff thinks that he is saying elsewhere. Komanoff claims that children [<12] cannot learn how to cycle properly, and that they play in the streets. Therefore, "education should be provided that teaches ... [motorists] how to avoid ... bicyclists and pedestrians."

2.11 Bike-Safety Education Doesn't Work

"Children younger than 12 years old rarely possess the cognitive, perceptual and behavior capability to learn and employ traffic-safety skills in real-world situations, especially when they are engrossed in play. .. there is no evidence that teaching children to be better pedestrians, or cyclists, reduces their risk of injury on the roads. Instead, education should be provided that teaches those most amenable to change [motorists] how to avoid the must vulnerable road users, bicyclists and pedestrians. ... Although education of children has been a major, even dominant approach for preventing traffic injuries in child pedestrians and cyclists in the US, it is time to revisit the value of such programs for that age group."

2.12 Big Solution

"Establishing in practice the legal right to ride a bicycle is key to making cycling widespread and safe. Let's get on with it."

3 Analysis

3.1 Car-Bike Collisions

3.1.1 Relevance of Komanoff's Data

Komanoff's study of fatal car-bike collisions in New York City is his only use of observed facts. Therefore, that study is very significant in assessing the accuracy of Komanoff's opinions.

The study of only fatal car-bike collisions does not provide valid information about the whole range of car-bike collisions, and none at all about the other types of accidents to cyclists. Car-bike collisions constitute only about 1/8 of accidents to cyclists, and fatal ones constitute only about 1/400 of accidents to cyclists. Fatal car-bike collisions are known to be biased, relative to all car-bike collisions, having higher proportions of nighttime, motorist overtaking, and alcohol involvement. Furthermore, New York City is very unrepresentative of the nation as a whole, with accident statistics very different from those of most of the rest of the nation.

To consider the bicyclist accident situation in New York City as even remotely a valid view of that situation in the nation is absurd. Komanoff does not say that it is, but presents no information about the rest of the nation.

3.1.2 Accuracy of Komanoff's Opinions

Komanoff's main conclusion is that: "The NYC PD attributed 3/4 of fatal bicycle-auto crashes to cyclist error, while [Komanoff's] study reached the completely opposite conclusion."

This would be a serious criticism of the official car-bike collision statistics of New York City, if it were correct. Komanoff chooses to present only one example in which his method of analysis produces results opposite to that of the police department. One must presume that Komanoff thinks that this is an ideal example to illustrate the validity of his analytical procedure. (See description above.)

K's analysis is hokum. A 40-ton tractor-trailer rig is long. The report clearly does not blame the truckdriver for turning left from the right-hand side of the roadway, or Komanoff would have so stated. The truckdriver was turning left from a point on the roadway reasonably furthest to the left of the available space for travel in his direction. The point of impact was on the "left rear side of the vehicle." The cyclist was clearly trying to overtake a left-turning vehicle on its left side. Komanoff fails to state whether or not the street was a two-way street, a lapse that illuminates his incompetence. In any case, it is the duty of the overtaking driver to see that his movement is done with safety, and this the dead cyclist clearly failed to do.

Considering the nature of the collision and the prevalence of one-way streets in NYC, I suspect that this collision was the mirror image of the classic right-hook car-bike collision.

In accidents of this type (right hook or left hook) one cannot assign blame until the two movements are known. If the motorist makes his turn directly into a cyclist alongside him, the motorist is at fault. However, if the motorist is well into his turn and the cyclist insists on overtaking the turning motorist, then the cyclist is at fault. The police report stated that the cyclist ran into the left rear portion of the truck. Considering this fact and the low probable speed of a long tractor-trailer rig turning left in NYC traffic, I would think it more likely that the cyclist was at fault for trying to overtake a turning vehicle.

The only reasoning that Komanoff gives for his opinion about fault is that the cyclist was proceeding lawfully and was hit by a turning vehicle. Was the cyclist proceeding lawfully? He was proceeding to overtake on the left a vehicle turning left towards him, a situation for which the law permits overtaking on the right-hand side of the turning vehicle. (UVC 11-304a1) While it is true that the law does not specifically prohibit overtaking on the left side of a vehicle turning left, there is no doubt that (except in the case of multiple turning lanes) doing so is so foolhardy that the attempt would violate the requirement that the overtaking driver do so only when it is safe to do so.

With Komanoff providing such an incompetently analyzed accident as his one example of how to make such analyses, and the shallowness and just plain inaccuracy of his analysis of traffic law, there is no reason to place any credibility on the conclusions that he says he reached on the basis of his report.

3.1.3 My Criticism

Komanoff asserts that the overriding requirement is to reassert the legitimacy of cycling. However, analyzing some unrepresentative portion of accidents to cyclists according to an incompetent logical patter provides no justification at all for the legitimacy of cycling.

3.2 Praise of Dutch Bikeways

Komanoff approvingly refers to the work of Prof. Pucher that praises the Dutch bikeway system. 

3.2.1 My Criticism

Prof. Pucher's work that praises the Dutch bikeway system has been resoundingly discredited in my article "The Bikeway Controversy" in the current Transportation Quarterly, Spring 2001. Komanoff praises the "excellent cycling infrastructure" that "gives bicycles a high priority." These statements are utterly false. The Dutch bikeway system prevents cyclists from operating as drivers of vehicles and requires them to operate as rolling pedestrians. The Dutch bikeway system gives cyclists the lowest priority by prohibiting their use of the adjacent normal roadways, and requiring them to obey traffic signals that delay them much more than if they were operating on the normal roadway.

In short, although Komanoff asserts that the overriding requirement is to reassert the legitimacy of cycling, he recommends the system that entirely removes the legitimacy of cycling on the normal roadway and removes cyclists to facilities with lower priority and greater delays.

3.3 Praise of Dutch Traffic Law

Komanoff praises provisions of Dutch traffic law that usually assigns the fault in a car-bike collision to the motorist. In contrast, US law does not consider the types of vehicles involved, only whether or not each one of them was operated according to law. Komanoff's reasoning is that since cyclists are incompetent and vulnerable, motorists should be required to act always as if cyclists will operate unlawfully.

3.3.2 My Criticism

Komanoff asserts that the overriding requirement is to reassert the legitimacy of cycling. However, his recommendation is to require that traffic law presume that cyclists operate incompetently. Consider the first demand of motorists when confronted with such a law. Get those damned bicycles off the roads, because they cannot be operated properly and we are being held responsible for the consequences of their incopetence.

3.4 Traffic Calming

Komanoff praises the Dutch voonerven, the traffic-restricted play streets. These are so constructed and used that vehicular travel becomes extremely dangerous. Therefore, the speed limit is very low.

3.4.1 My Criticism

Komanoff fails to realize that these streets are more dangerous for cyclists than for motorists, since cyclists who hit running pedestrians can even be killed, while motorists who do so lose only a little paint.

In short, while Komanoff asserts that the overriding requirement is to reassert the legitimacy of cycling, he recommends that they operate on playgrounds instead of normal streets.

3.5 Cyclist Education

Komanoff concludes that the bike-safety education program has not worked. (I agree, but for different reasons.) Komanoff argues for motorist education that places more responsibility on the motorist to avoid car-bike collisions.

3.5.1 My Criticism

I estimate that approximately 30% of US car-bike collisions are caused by the cyclist acting as he is encouraged to act by the bike-safety instruction. The bike-safety instruction has not worked because it teaches that the cyclist is inferior to motor vehicles, and, therefore, has to operate by more dangerous rules.

Komanoff asserts that the overriding requirement is to reassert the legitimacy of cycling. However, his recommendation is exactly the opposite, in that he wants to teach motorists that cyclists are incapable of operating by the rules of the road for drivers of vehicles.

3.6 Changing Traffic Law for Right-of-Way

Komanoff recommends that traffic law be rewritten so that there is a hierarchy of yielding of right of way: motorists yield to cyclists, who, in turn yield to pedestrians. He states: "Motorists should be required by law to yield to bicycles, and bicycles in turn to pedestrians, in the absence of any overriding traffic rule or directive, such as a traffic signal."

3.6.1 My Criticism

Komanoff's proposal has two parts. The first is the hierarchy of yielding. Present traffic law provides that yielding and right-of-way are specified for the conflicts arising out of each movement. Komanoff asserts that yielding should be in accordance with his vulnerability scale. If this were to be enacted into law, consider the results. Suppose a cyclist, followed by a motorist, intend to make a left turn. The cyclist has the right of way over the opposite-direction motor traffic, but not over the opposite-direction bicycle traffic. So the motor traffic stops for him, while he collides with the opposite-direction cyclist. Meanwhile, although the opposite direction motor traffic has stopped for the left-turning cyclist, it has the right-of-way over the left-turning motorist. If two motorists collide, who had the right-of-way, those who had just restarted, or those who were in the midst of the turn? Utter chaos. Komanoff clearly has not thought about the consequences of such a claim.

However, Komanoff might consider himself saved by his qualification clause: "in the absence of any overriding traffic rule." In the left-turn situation, the left-turning vehicles are required to yield to the opposite-direction vehicles, which, presumably is the kind of "overriding traffic rule" to which Komanoff's assertion refers.

However, the traffic rules for drivers of vehicles have been so worked over through the decades that there are rules for every pair of conflicting movements. Therefore, there is no room in which Komanoff's proposal could operate. In any case, if we did discover such a gap in the traffic laws, it would be best to make a rule that fits the movement, in the same manner as all the other rules, rather than to make a rule that is based on an entirely different system, which will surely cause trouble.

Komanoff asserts that the overriding requirement is to reassert the legitimacy of cycling. However, consider the results of giving bicyclists right-of-way legitimacy over that of motorists. What's the first thing that motorists would demand? Keep those damned bicyclists off the roads, because it is impossible to operate properly and safely when their presence fouls up the traffic laws.

3.7 Changing Traffic Law, Liability Considerations

Komanoff praises laws that "require motorists to anticipate unsafe walking and cycling." If a car-bike collision involves a child or a elderly person, "the motorist is usually judged to be entirely at fault." "When a crash is caused by an illegal move by a cyclist or a pedestrian, the motorist is almost always judged to be partly at fault."

3.7.1 My Criticism

Komanoff asserts that the overriding requirement is to reassert the legitimacy of cycling. However, his recommendation states that motorists shall be held responsible in liability for car-bike collisions, because the bicyclist is legally presumed to be allowed to operate incompetently. The immediate demand of motorists faced with such liabilities is that bicyclists be prohibited from using the roads and causing motorists to be held liable for the bicyclists' incompetent and unlawful operating actions.

4 General Conclusion

Have you noticed that although Komanoff asserts that the overriding requirement is to reassert the legitimacy of cycling, every recommendation that he makes tends to destroy the legitimacy of cycling? It is difficult to place much credibility in the thought processes that result, in every case, with recommendations that contradict the prime premise.

Many people have criticized my description of the emotional condition that produces such thought processes as the cyclist-inferiority phobia. Well, what other explanation can be provided for this example of completely self-contradicting reasoning?

Return to: John Forester's Home Page                               Up: Safety