Safe Streets, LivableStreets

Eric Dumbaugh

The danger in supplanting the real measure of safety (i.e., crash frequency andseverity) by surrogates arises when the link between the two is conjectural, when thelink remains unproven for long, and when the use of unproven surrogates becomesso habitual that the need to eventually speak in terms of crashes is forgotten. (Hauer, a, p. )

Beyond simply acting as thoroughfares for motor vehicles, urban streetsoften double as public spaces. Urban streets are places where people walk,shop, meet, and generally engage in the diverse array of social and recre-

ational activities that, for many, are what makes urban living enjoyable. Andbeyond even these quality-of-life benefits, pedestrian-friendly urban streets havebeen increasingly linked to a host of highly desirable social outcomes, includingeconomic growth and innovation (Florida, ), improvements in air quality(Frank et al., ), and increased physical fitness and health (Frank et al., ),to name only a few. For these reasons, many groups and individuals encouragethe design of “livable” streets, or streets that seek to better integrate the needs ofpedestrians and local developmental objectives into a roadway’s design.

There has been a great deal of work describing the characteristics of livablestreets (see Duany et al., ; Ewing, ; Jacobs, ), and there is generalconsensus on their characteristics: livable streets, at a minimum, seek to enhancethe pedestrian character of the street by providing a continuous sidewalk networkand incorporating design features that minimize the negative impacts of motorvehicle use on pedestrians. Of particular importance is the role played by road-side features such as street trees and on-street parking, which serve to buffer thepedestrian realm from potentially hazardous oncoming traffic, and to providespatial definition to the public right-of-way. Indeed, many livability advocatesassert that trees, as much as any other single feature, can play a central role inenhancing a roadway’s livability (Duany et al., ; Jacobs, ).

While most would agree that the inclusion of trees and other streetscapefeatures enhances the aesthetic quality of a roadway, there is substantive disagree-ment about their safety effects (see Figure ). Conventional engineering practiceencourages the design of roadsides that will allow a vehicle leaving the travelwayto safely recover before encountering a potentially hazardous fixed object. Whenone considers the aggregate statistics on run-off-roadway crashes, there is indeed

Transportation safety is a highly con-tentious issue in the design of cities andcommunities. While urban designers,architects, and planners often encouragethe use of aesthetic streetscape treatmentsto enhance the livability of urban streets,conventional transportation safety practiceregards roadside features such as streettrees as fixed-object hazards and stronglydiscourages their use. In this study, Iexamine the subject of livable streetscapetreatments and find compelling evidencethat suggests they may actually enhancethe safety of urban roadways. Concernsabout their safety effects do not appearto be founded on empirical observationsof crash performance, but instead on adesign philosophy that discounts theimportant relationship between driverbehavior and safety. This study traces theorigin and evolution of this philosophy,and proposes an alternative that may bet-ter account for the dynamic relationshipsbetween road design, driver behavior, andtransportation safety.

Eric Dumbaugh is a doctoral candidatein the School of Civil and Environmen-tal Engineering at the Georgia Instituteof Technology. His research focuses oncontext-sensitive solutions, nonmotor-ized travel, and transportation systemsafety.

Journal of the American Planning Association,

Vol. , No. , Summer .

© American Planning Association, Chicago, IL.

cause for concern. In alone, there were over ,

fatalities involving roadside objects such as trees and utilitypoles on U.S. roadways, accounting for more than % ofthe total fatalities for that year (National Highway TrafficSafety Administration [NHTSA], n. d.). Correspondingly,designing livable streets is often more difficult than simplycounterbalancing the needs of motorists with those of pe-destrians. How is the transportation designer to conscien-tiously incorporate design elements that may result in theloss of life?

This study details existing design guidance and liter-ature, as well as the historical evolution of contemporarysafety practice, and reports the results of an empirical testof the professional assumptions that guide the currentapproach to addressing safety through design. It concludesby outlining an approach to urban roadway design thatmay better address the twin goals of safety and livability.

Considering the Literature onRoadside Safety

The initial motivation behind this research effort wasan attempt to understand the safety impacts of livablestreetscape treatments on urban roadways. On this issue,the design guidance is clear: “for all types of highwayprojects, clear zones should be determined or identifiedand forgiving roadsides established” (American Associationof State Highway and Transportation Officials [AASHTO],, p. ). In practice, this entails providing a clear road-side adjacent to the vehicle travelway, with a preferredwidth of feet. In terms of how to best accomplish thisgoal, AASHTO’s () Roadside Design Guide, the centralauthority on the design of safe roadsides, is also clear:

Through decades of experience and research, theapplication of the forgiving roadside concept hasbeen refined to the point where roadside design is anintegral part of transportation design criteria. Designoptions for reducing roadside obstacles, in order ofpreference, are as follows:

1. Remove the obstacle.2. Redesign the obstacle so it can be safely traversed.3. Relocate the obstacle to a point where it is less likely

to be struck.4. Reduce impact severity by using an appropriate

breakaway device.5. Shield the obstacle with a longitudinal traffic barrier

designed for redirection or use a crash cushion.6. Delineate the obstacle if the above alternatives are

not appropriate. (pp. –)

Journal of the American Planning Association, Summer , Vol. , No.

Figure . Livable streetscape treatments: urban amenities or roadsidehazards?

While the Roadside Design Guide cites “decades ofexperience and research,” there is very little information onthe use of aesthetic streetscape features, and much of theexisting literature on the application of clear zone policies inurban environments is problematic, at best. The definitivework on the subject is a study that describes the physicalcharacteristics of trees involved in crashes within the Citylimits of Huntsville, Alabama (Turner & Mansfield, ).This study found that most crashes involving trees occurwithin feet of the roadway, that % of the reportedcrashes involved trees with a caliper width inches or more,and that almost % occurred on a horizontal curve. Whilesuch information is useful for understanding the character-istics of tree-related crashes, it does not lead to the conclu-sion that eliminating trees with any or even all of thesecharacteristics will have any effect on a roadway’s safety.Such conclusions can only be made by examining the actualcrash performance of eliminating trees in urban areas, asmeasured by changes in crash frequency and severity.

Indeed, there is a growing body of evidence suggestingthat the inclusion of trees and other streetscape features inthe roadside environment may actually reduce crashes andinjuries on urban roadways. Naderi () examined thesafety impacts of aesthetic streetscape enhancements placedalong the roadside and medians of five arterial roadways indowntown Toronto. Using a quasi-experimental design,the author found that the inclusion of features such as treesand concrete planters along the roadside resulted in statis-tically significant reductions in the number of mid-blockcrashes along all five roadways, with the number of crashesdecreasing from between and % as a result of the street-scape improvements. While the cause for these reductionsis not clear, the author suggests that the presence of a welldefined roadside edge may be leading drivers to exercisegreater caution.

Ossenbruggen, Pendharkar, and Ivan () examinedsites with urban, suburban, and residential characteristicsin New Hampshire and hypothesized that the urban “vil-lage” areas, with greater traffic volumes and more pedes-trian activity, would be associated with higher numbers ofcrashes and injuries. Instead, they found the opposite: thevillage areas, which had on-street parking and pedestrian-friendly roadside treatments, were two times less likely toexperience a crash event than the comparison sites. Theauthors associate these crash reductions with the character-istics of the roadside environment, which included side-walks, mixed land uses, and other “pedestrian-friendly”roadside features. The authors also attributed the safetyperformance to reduced speeds, noting that “since nospeed limit signs are erected at village sites, it suggests[speeds] are self regulating” (p. ).

A study of two-lane roadways by Ivan, Pasupathy, andOssenbruggen () found that while shoulder widthswere associated with reductions in single-vehicle, fixed-object crashes, they were also associated with a statisticallysignificant increase in total crashes, with multiple-vehiclecrashes offsetting safety gains achieved through reductionsin fixed-object crashes. The authors comment that “the pos-itive coefficient on right shoulder width is troubling; onenormally expects a wider shoulder to be a safety feature”(p. ).

Finally, Lee and Mannering () examined run-off-roadway crashes along a -km section of an arterial road-way in Washington State that traveled through both urbanand rural environments. Using a negative binomial model,the authors sought to associate crash frequencies with thecharacteristics of the roadside environment. While theirmodel for rural areas performed as expected, with trees andother features being associated with a statistically significantincrease in the number of roadside crashes, their model forurban areas produced radically different results (see Table). Not only were trees not associated with crash increases,but the model coefficients entered negatively at a statisti-cally significant level, indicating that the presence of treesin urban areas was associated with a decrease in the proba-bility that a run-off-roadway crash would occur.

The authors attribute these unexpected crash reduc-tions to the fact that there are fewer trees in urban areasthan in rural ones, but this begs the question: even if thereare fewer trees in urban areas, which suggests that theirpresence would violate driver expectancy, why are theyassociated with statistically significant crash reductions?

Other roadside features proved to be statisticallyrelated to crash reductions as well. The number of signsupports was associated with crash reductions, as was thepresence of miscellaneous fixed objects, a variable thatincluded such roadside features as mailboxes. Further,wider lanes and shoulders were associated with statisticallysignificant increases in crash frequencies.

Interestingly, clear zones are not the only design fea-ture for which such safety anomalies appear. Hauer (a)reexamined the literature on lane widths and found thatthere was little evidence to support the assertion thatwidening lanes beyond feet enhances safety. Instead, theliterature has almost uniformly reported that the safetybenefit of widening lanes stops once lanes reach a width ofroughly feet, with crash frequencies increasing as lanesapproach and exceed the more common -foot standard.

Further, in a series of broad-sweeping and profoundlyimportant studies, Noland (, ) and Noland andOh () consistently found that when one controls forintervening factors such as time-series effects, seat belt use,

Dumbaugh: Safe Streets, Livable Streets

and the demographic characteristics of the population,conventional design “improvements” result in increases incrashes and fatalities.

Indeed, there are a host of safety anomalies in theexisting design literature, but as Noland and Oh ()stated, the problem is that

Studies that find unexpected or unconventional resultstend to dismiss these results as aberrations and havenot examined them in further detail. . . . The resultsof many of these studies lead us to conclude that theimpact of various infrastructure and geometric designelements on safety are inconclusive. Most studies usingsophisticated statistical techniques either find noassociation, or an unexpected association from infra-structure changes assumed to be beneficial. (p. )

Thus, a key question emerges: why does contemporarydesign guidance recommend practices that the best avail-able evidence suggests may have an ambiguous or even neg-

ative impact on safety, and paradoxically, to do so underthe auspices that they constitute a safety enhancement?

The Passive Safety Paradigm

While safety has been a concern for the transportationdesign profession throughout its history, the current ap-proach to addressing transportation safety received itsphilosophical basis as part of the transportation safetymovement of the s, a movement that resulted in theNational Highway Safety Act, the creation of the NationalHighway Traffic Safety Administration (NHTSA), theadoption of the Roadside Design Guide, crash testing, andthe development of air bags, among other features of thecontemporary transportation safety landscape. This move-ment, led by William Haddon and promoted by figuressuch as Daniel Patrick Moynihan and Ralph Nader, soughtto apply the principles of epidemiology to transportationsafety issues (Gladwell, ; Kratzke, ; McLean, ;Viano, ; Weingroff, ).

As a profession, epidemiology is based on the work ofJohn Snow, an English physician who sought to address anoutbreak of cholera that plagued London in the s. Whilecurrent medical theory asserted that the spread of cholerawas associated with “vapors,” Snow hypothesized thatcholera was not airborne, but was instead transmittedthrough polluted water supplies. Using what was at thetime a highly elaborate, data-driven analysis, Snow mappedout the locations of affected households, and determinedthat these households were indeed sharing a common watersource. In an episode that has since become legendary,Snow sought to resolve this problem in one particularlyhard-hit neighborhood by implementing a strategy thatwas both simple and radical: rather than encouraging resi-dents to adopt behavioral modifications, such as boilinginfected water or using an alternative water supply, Snowsimply removed the handle from the pump of the affectedwell, thereby neutralizing the environmental cause of thehazard (Rosenberg, ).

William Haddon, an epidemiologist trained at theHarvard School of Public Health during the s, likewisebelieved that it was difficult, if not impossible, to preventpeople from engaging in behaviors that lead to trafficinjuries and fatalities. Instead, Haddon proposed a passiveapproach: rather than relying on behavioral modificationsto prevent crashes from occurring, Haddon believed thedesign objective should be to enable a “crash without aninjury” by physically engineering safety features into vehi-cles and their environments (Gladwell, ). The ideawas compelling: what if transportation professionals could

Journal of the American Planning Association, Summer , Vol. , No.

Table . Negative binomial estimation results for crash frequencies inurban areas (Lee & Mannering, ; reprinted by permission).

Estimated Variable coefficients t-statistic

Constant −.

Roadway characteristicsBroad lane indicator ( if lane is greater

than . meters, otherwise) . .

Median width (meters) −. −.

Roadside characteristicsBridge length (meters) . .

Distance from outside shoulder edge to guardrail (meters) . .

Fence length (meters) . .

Number of isolated trees in a section −. −.

Number of miscellaneous fixed objects in a section −. −.

Number of sign supports in a section −. −.

Shoulder length (meters) −. −.

Dispersion parameter . .

Restricted log likelihood −.Log likelihood at convergence −.Number of observations ,

design vehicles and roadways to eliminate the injuriesassociated with a crash event?

Clear Zones, Highways, and Passive SafetyThe life safety implications of the passive approach

were not lost on Ralph Nader. In , Nader publishedUnsafe At Any Speed, a critique of the auto industry basedon Haddon’s passive safety philosophy. Nader’s book gen-erated a public outcry to address the “designed-in” dangersof the nation’s automobiles and transportation system,leading both congress and AASHO (later AASHTO) tohold special hearings on the subject in . Figures suchas Nader and Haddon reported to these committees, buttestimony by Kenneth Stonex played perhaps the centralrole in formulating the contemporary perspective on saferoadway design.

One of the key problems identified by the AASHOcommittee was the large number of fatalities associatedwith single-vehicle, run-off-roadway crashes. To addressthis issue, they heard testimony from Stonex, a GeneralMotors employee responsible for designing the “ProvingGround,” an experimental “crashproof” highway that had-foot clearances on either side of the travelway (McLean,; Weingroff, ). Based on the test performance ofthe Proving Ground, Stonex was of the opinion that “Whatwe must do is to operate the % or more of our surfacestreets just as we do our freeways . . . [converting] the surfacehighway and street network to freeway and Proving Groundroad and roadside conditions” (Weingroff, , p. ).

With respect to fixed-object crashes specifically, Stonexreported that most vehicles on the Proving Ground cameto a stop within feet after leaving the roadway. Thus,the committee concluded that eliminating fixed objectswithin feet of the travelway would eliminate most fixed-object crashes and, in a conjectural leap, that the roadwaywould therefore be safer as a result. The -foot clear zonestandard (with adjustments for sideslope) was thus incor-porated into AASHO’s publication Highway Designand Operational Practices Related to Highway Safety, as wellas the revised edition, and remains in the subsequenteditions of the Roadside Design Guide (AASHTO, ,; McLean, ; Weingroff, ).

Side Effects of the Passive TreatmentPrior to the s, transportation safety had been

addressed primarily through strategies aimed at encourag-ing drivers to engage in safe behavior, an approach that ledto the development and codification of the nation’s signingpractices and motor vehicle laws. Yet, as Nader testified,focusing on behavior was not an adequate solution to theproblem:

Even if people have accidents, even if they make mis-takes, even if they are looking out the window, or theyare drunk, we should have a second line of defense forthese people . . . the sequence of events that leads to anaccident injury can be broken by engineering measureseven before there is a complete understanding of thecausal chain. (quoted in Weingroff, , p. )

Following the hearings, contemporary safetypractice thus became principally concerned with how toengineer this second line of defense, shifting the profes-sion’s focus away from driver behavior and towards thedesign of vehicles and roadside hardware. Passive safetybegins from the perspective that drivers will err, combinedwith the observation that there are fewer crashes on Inter-states than on other roadways. Collectively, this resultedin the conclusion that “Highways built with high designstandards put the traveler in an environment which isfundamentally safer because it is more likely to compensatefor the driving errors he will eventually make” [emphasisadded] (AASHTO, , p. ).

This perspective is still evident in the most recentedition () of AASHTO’s A Policy on the GeometricDesign of Highways and Streets (the “Green Book”), whichstates:

The objective in design of any engineered facility usedby the public is to satisfy the public’s demand forservice in a safe and economical manner. The [high-way] facility should, therefore, accommodate nearly alldemands with reasonable adequacy and also shouldnot fail under severe or extreme traffic demands . . .every effort should be made to use as high a designspeed as practical to attain a desired degree of safety.(pp. –)

Thus, the passive approach attempts to enhance aroadway’s safety by designing it to accommodate the safetyneeds of high-speed, “extreme” driving behavior. Thisapproach hinges on a critical assumption, however: itassumes that drivers who already drive safely will continueto do so when forgiving design values are used, therebyenhancing the overall safety of a roadway by making it safefor not only “average” drivers, but also extreme drivers aswell.

While the logic behind the passive approach has ahigh degree of face validity, it overlooks several importantquestions: how do average drivers adjust their behavior toforgiving design values? What about specific at-risk sub-populations? Is it possible that by widening lanes andshoulders and eliminating roadside objects, designers are

Dumbaugh: Safe Streets, Livable Streets

encouraging “non-design drivers” to adopt behaviors thatresult in crashes and injuries?

A Simple Empirical Test

With respect to determining the appropriate clear zonefor a roadway, the most recent guidance states that “thewider the clear zone, the safer it will be” (TransportationResearch Board [TRB], , p. V-). If this is true, thenone would expect livable streetscape treatments to be lesssafe, in terms of crash frequency and severity, than road-ways adopting more forgiving values for lane widths andclear zones. To test this assertion, as well as to build anunderstanding of the safety effects of livable streets moregenerally, I examined years (–) of crash data forColonial Drive (State Road ), a state-owned arterial thatconnects the north end of downtown Orlando, Florida, toits eastern and western suburbs.

While none of Colonial Drive would be regarded as aparticularly representative example of a livable street, the.-mile segment that constitutes the northern edge ofdowntown Orlando (between Orange Avenue and MillsAvenue) includes many of the design features desired bylivable streets advocates. Roadside development abuts thesidewalk, which is often uncomfortably narrow at points,but continuous throughout the section. Lane widths arenarrower here ( feet) than on much of the remainder ofthe roadway, and the segment includes on-street parkingand roadside objects that buffer the pedestrian environ-ment. The cross section, curb-to-curb, is roughly feet,including four -foot travel lanes, a -foot painted me-dian, and two .-foot parking lanes. Roadside objects areoffset by . to feet from the curb (see Figure ).

To evaluate the safety performance of this segment, itwas matched with the nearest .-mile section of ColonialDrive that was similar in terms of cross-sectional character-istics (four lanes and a painted median), posted speed limit,and average daily traffic (ADT) but which used more for-giving values for lane widths and clear zones.

The nearest comparison length was located slightly lessthan miles east of the livable section described above. Asshown in Table , the roadways are almost identical in allrelevant characteristics of interest—section length, ADT,number of lanes, and median width—while differing prin-cipally in terms of lane widths and roadside object offsets.It should be observed that the posted speed limit is slightlyhigher for the comparison section ( mph vs. mph),but not substantially so. There is little difference in theaverage number of crashes per intersection or the mean ageof at-fault drivers. Further, the use of a nearby comparison

section on the same roadway helps control for the uniquecharacteristics of the driver population, which in this casewill include many of the exact same drivers. Holding all ofthese features constant, if the passive safety assumption holds,there should be fewer mid-block injuries and fatalities onthis comparison section.

Comparing Crash PerformanceAs shown in Table , the livable section is safer in all

respects. By any meaningful safety benchmark—total mid-block crashes, injuries, or fatalities—there can be littledoubt that the livable section is the safer roadway.

A second area of interest for this study is the specificdistribution of crash types across the roadways. What arethe types of crashes that result in mid-block fatalities andinjuries? There were no fatal mid-block crashes on thelivable section during the -year evaluation period, while occurred along the comparison length of the roadway, ofwhich involved pedestrians. Harmful event data were notprovided on the other fatalities.

Pedestrian and bicyclist injuries were likewise higheron the comparison section (see Table ), which may bepartly attributable to the fact that the livable section pro-vides parked cars and fixed objects to buffer pedestriansfrom oncoming traffic. But do the benefits in pedestriansafety outweigh the hazards these features may pose toerrant motorists?

For the livable section, there were injurious crashesinvolving roadside objects, one involving a tree, and a sec-ond involving a parked car. Comparatively, there were injurious crashes involving pedestrians and bicyclists on thecomparison section, of which were fatal.

What about the relative roadside hazards these designsmight pose? From a passive safety perspective, the compari-son section, with a -foot clear zone, should be the saferof the two in terms of total injurious collisions with fixedobjects. Yet this is also not the case. Not counting pedes-trians and bicyclists, there were roadside object-relatedinjuries on the comparison section, versus in the livablesection.

Finally, what about motor vehicle collisions? Theaverage number of crashes per intersection were similarbetween the two study sections. Might the comparisonsection be at least as safe in terms of two-vehicle mid-blockcrashes? Again, the answer is no. For rear-end crashes, thecrash type most likely to be associated with on-streetparking, there were fewer injuries for the livable section.Likewise with injuries associated with head-on crashes, turn-related crashes, and sideswipe crashes. Only angle crasheswere comparable, with both sections reporting suchinjurious crashes during the -year evaluation period. In

Journal of the American Planning Association, Summer , Vol. , No.

total, more multiple-vehicle mid-block injurious crashesoccurred on the comparison section than the livable one.

Comparing the Livable Section to BaselineRoadway Safety Performance

To understand how the livable section of ColonialDrive performed against urban operating conditions alongState Route , I further compared its crash performance

to -mile sections of Colonial Drive located on either sideof the livable section, thereby capturing the majority ofurban and suburban travel along this roadway. While thisapproach does not control for specific design variationsalong individual roadway segments, it is useful for deter-mining whether the livable section is more or less safe thanone would expect, on average, from the urbanized portionof the roadway as a whole.

Dumbaugh: Safe Streets, Livable Streets

Figure . Colonial Drive: livable and comparison sections.

Crashes were normalized by determining the numberof crashes per million vehicle miles traveled (VMT),

thereby developing a measure of exposure that could beused to directly compare safety performance. Nevertheless,a problem with VMT-based measures is that the relation-ship between VMT and crashes is not linear (Ivan et al.,). This finding may be attributable to the fact that highlevels of congestion occurring during peak periods can havethe dual effect of increasing the denominator of the meas-ure while simultaneously reducing free-flow travel speeds,the combination of which may result in underestimates ofa roadway’s actual hazard during low-volume, free-flowoperating conditions, such as late-night travel. To addressthis concern, I also evaluated safety performance based onthe number of mid-block crashes per mile, a measure thatmakes no assumption about the relationship betweentraffic volumes and crash performance.

As shown in Table , the livable section of ColonialDrive is safer, by either measure, than one would expectwhen examining the -mile urban and suburban compar-ison section as a whole, reporting fewer total mid-blockcrashes than the comparison length of Colonial Drive, andsubstantially fewer injurious and fatal crashes.

Trend IdentificationTo determine whether the safety performance of Co-

lonial Drive might perhaps be part of a broader safety trend,I further examined the crash performance of state arterialroadways traveling through the National Register–desig-nated historic districts of DeLand and Ocala, Florida. Eachcity has two .-mile sections of state roadways entering itshistoric district, with all four roadways having dense devel-opment adjacent to the travelway, minimum (.– feet)fixed-object offsets, and, for the two DeLand roadways,on-street parking as well.

Journal of the American Planning Association, Summer , Vol. , No.

Table . Design characteristics of livable and comparison sections of Colonial Drive.

Characteristic Livable section Comparison section

Length (miles) . .Average daily traffic (vehicles) , ,

Posted speed limit (mph)

Lanes x ft. x . ft.Median ft. painted ft. paintedShoulder . ft. parking lane ft. paved shoulder ± ft. runout zoneAvg. crashes per intersection

Mean age of at-fault driver

Table . Mid-block crashes on Colonial Drive, –.

Livable Comparison DifferenceMid-block crashes section section (%)

Total −%Injurious −%Fatal −%

Table . Injurious mid-block crashes on Colonial Drive, –.

Livable ComparisonCrash type section section Difference

Rear-end −

Head-on −

Angle

Left-turn −

Sideswipe −

Parked car

Pedestrian −

Bicyclist −

Tree/shrubbery

Other fixed object −

Ditch/culvert −

Other/unreported −

Total −

These roadways also included posted speed limit re-ductions of mph or more, thus preventing one-to-onecomparisons with adjacent sections. Nevertheless, under-standing their safety performance with respect to the urbansections of these roadways as a whole does allow one to evalu-ate whether such treatments are safer than one would expectfrom baseline roadway averages, and is extremely useful fordetermining whether the safety performance of the livablesection of Colonial Drive is anomalous, or whether it mightperhaps be part of a broader safety trend.

In the absence of detailed field observations, the historicdistrict boundaries were used to determine the boundariesof the livable sections of these roadways, and these sectionswere then compared against the crash performance of -milesections of the same roadway located on either side of thehistoric district, thereby permitting a consistent compari-son of these roadways both against each other and againstColonial Drive. The individual performance of theseroadways, as well as their averages, are reported in Table .

Like Colonial Drive, the livable sections were generallysafer than their comparison roadways. On average, the his-toric roadway sections reported somewhat fewer total crashesand substantially fewer injurious crashes. Perhaps most not-ably, not a single fatal crash was reported for any of thesehistoric roadway sections during the -year analysis period.

Individually, two specific results warrant noting. First,while the historic section of Woodland Avenue shows re-ductions in total and injurious crashes on a per-mile basis,it reports substantially higher crashes and injuries whenusing a VMT-based metric. In this case, the relatively lowlevel of VMT observed for the historic section when com-pared to its comparison roadway (, vs. , ADT)may overestimate its relative hazard. In absolute terms, thenumber of mid-block injurious crashes on the historic sec-tion of Woodland Avenue is identical to that for the otherthree roadways, with all of the historic sections reporting

exactly injury crashes over the -year analysis period, or injurious crashes per mile.

Pine Avenue, while safer than the comparison roadwayoverall in terms of injuries and fatalities, nevertheless reportsa higher number of total mid-block crashes than the road-way as a whole, although this may in part be attributableto cross-sectional differences. While the majority of StateRoute , of which Pine Avenue is a part, is a four-laneroadway, the .-mile section that travels through Ocala’shistoric district briefly switches to a six-lane cross section, afactor that has been shown to result in higher numbers ofcrashes and injuries (Noland & Oh, ). The fact thathigher total crash rates were not accompanied by higherrates of injurious or fatal crashes is an interesting andpotentially important finding.

Reconsidering the RelationshipBetween Safety and Design

To reject one paradigm without simultaneously substitutinganother is to reject science itself. (Kuhn, , p. )

While these results seem to contradict conventionaldesign practice, they confirm a trend that many researchersand practicing engineers have observed for some time, butwhich has received little substantive elaboration: specifically,that clear zones and other forgiving design practices oftenhave an ambiguous relationship to safety in urban environ-ments, and may be associated with declines in safety per-formance. The best possible explanation for the enhancedsafety performance of the livable sections considered in thisstudy is that drivers are “reading” the potential hazards ofthe road environment and adjusting their behavior inresponse.

Dumbaugh: Safe Streets, Livable Streets

Table . Mid-block crash performance of Colonial Drive, livable section vs. -mile urban comparison section, –.

Mid-block crashes per 100 million VMT Mid-block crashes per mile

Livable 10-mile Difference Livable 10-mile Differencesection comparison (%) section comparison (%)

Mid-block crashesTotal . −.% . −.%Injurious . −.% . −.%Fatal . −.% . −.%

The reason why this subject has not received greaterattention in design literature and guidance appears to bethat it contradicts the prevailing paradigm of what consti-tutes safe roadway design. Nevertheless, a behavior-basedunderstanding of safety performance is supported byresearch and literature in the field of psychology, which hasfocused on the subject of traffic safety as a means for under-standing how individuals adapt their behavior to perceivedrisks and hazards.

Risk homeostasis theory, as developed by Wilde(, , ), asserts that individuals make decisionson whether to engage in specific behaviors or activities byweighing the relative utility of an action against its perceivedrisk. While all actions involve some risk, risk homeostasistheory asserts that individuals will adjust their behavior tomaintain a static level of minimum exposure to perceivedhazard or harm. With respect to driving behavior, risk ho-meostasis theory posits that drivers intuitively balance the

relative benefits of traveling at higher speeds or engaging inother higher-risk driving behavior against their individualperceptions of how hazardous engaging in such behaviormight be. Where hazards are present and visible, such as inthe case of livable streetscape treatments, risk homeostasistheory would expect drivers to compensate for this per-ceived environmental hazard by adjusting their behaviorto minimize their exposure to risk.

Nevertheless, risk homeostasis theory would also assertthat, ceteris paribus, the relative crash performance of aroadway should remain constant along its length, regard-less of specific design variations, since any change in per-ceived hazard will be offset by corresponding adjustmentsin behavior. Thus, according to risk homeostasis theory,the livable street sections should be no more or less safethan their comparison roadways overall.

Yet as Hauer (b) describes, there is an importantdistinction between safety, which is (or should be) an em-

Journal of the American Planning Association, Summer , Vol. , No.

Table . Mid-block crash performance of roadways in historic districts and -mile comparison sections, –.

Mid-block crashes per 100 million VMT Mid-block crashes per mile

Historic 10-mile Difference Historic 10-mile DifferenceRoadway district comparison (%) district comparison (%)

Pine Ave., Ocala (SR )Total . . .% . . .%Injurious . . −.% . . −.%Fatal . . −.% . . −.%

Silver Springs Blvd., Ocala (SR )Total . . −.% . . −.%Injurious . . −.% . . −.%Fatal . . −.% . . −.%

New York Ave., DeLand(SR )Total . . −.% . . −.%Injurious . . −.% . . −.%Fatal . . −.% . . −.%

Woodland Ave., DeLand(SR )Total . . .% . . −.%Injurious . . .% . . −.%Fatal . . −.% . . −.%

AverageTotal . . −.% . . −.%Injurious . . −.% . . −.%Fatal . . −.% . . −.%

pirical measure of crash performance, and security, which isan individual’s subjective perception of safety (or conversely,perceived exposure to harm). The presence of features suchas wider lanes and clear zones would appear to reduce thedriver’s perception of risk, giving them an increased butfalse sense of security, and thereby encouraging them toengage in behaviors that increase their likelihood of beinginvolved in a crash event. If so, this explains why the livablestreetscape treatments examined in this study resulted innot only fewer fixed-object crashes, but fewer multiplevehicle and pedestrian crashes as well. Such treatmentsappear to help balance drivers’ sense of security with thereal levels of risk in their environment, providing themwith more accurate information on the appropriate level ofcaution, and resulting in behavioral adjustments that betterprepare them for the potentially hazardous vehicle andpedestrian conflicts that one encounters in urban environ-ments. From the perspective of risk homeostasis theory,the use of high design values is not “forgiving,” but isinstead “permissive.”

Researchers attempting to understand safety anomaliesemerging in their work have implicitly suggested a driver’srisk perception accounts for their findings. Ossenbruggenet al. () speculated that the better safety performanceof urban villages may be attributable to the fact that theroadside environment “warn[s] drivers that they must main-tain a low speed and use caution” (p. ). In explainingwhy new roadway improvements were shown to result inan increase in crashes and injuries, Noland () suggestedthat “higher design standards [allow] drivers to increase theirspeeds on roads and reduce their levels of caution” (p. ).

Towards a Theory of Positive Design

. . . competent drivers can be given appropriate informationabout hazards and inefficiencies to avoid errors. (FederalHighway Administration, , p. -)

The idea that safety can be addressed by focusing ona driver’s perception of risk, rather than relying solely onpassive engineering principles, is not without precedent inthe engineering community. Two important byproducts ofthe passive safety approach are the related concepts of posi-tive guidance and driver expectancy, which first emerged inthe Appendix to the second edition of AASHTO’s High-way Design and Operational Practices Related to HighwaySafety () as a means to address crashes associated withnarrow bridges. While emphasizing that the consistent

application of freeway standards is the preferred solutionfor addressing safety at narrow bridges, the guide remarks

that “it would take years and billions of dollars to effectsuch a program” (p. ).

In an attempt to satisfice a lower-cost, more imple-mentable solution, the guidance proposes that “highwaysafety can be considerably improved by restructuring thedriver’s expectancies so that he is prepared for the narrowbridge situation [and] the narrowing of the shoulder and/or roadside . . .” (p. ). The guidance then proceeds todetail how to adequately sign and mark the approach tothe “restricted” condition of a narrow bridge.

To date, positive guidance has focused largely on theuse of pavement markings and signs to convey safety infor-mation, and there has been relatively little advancement inthis area since , when the most recent edition of theFederal Highway Administration’s (FHWA’s) A Users Guideto Positive Guidance was published. Nevertheless, it maybe time to resurrect this concept, particularly as it mayrelate to the physical design of highways and streets.

A positive approach to transportation design wouldseem to explain the emerging safety anomalies the passiveapproach simply cannot account for, such as Naderi’s() findings on aesthetic streetscape treatments or thelivable street examples included in this study. It also ex-plains why narrowings and chicanes, two traffic calmingapplications that modify the roadside in a manner thatpassive safety suggests should increase crashes and injuries,have been shown to result in substantial (%–%) crashreductions (Zein et al., ). Indeed, all traffic calmingmeasures appear to reduce accidents by slowing traffic and/or increasing driver caution (Ewing, ), leading Euro-pean designers to view them not as “livability” features butas safety countermeasures (Skene, ).

A Positive Approach to the Design of UrbanStreets

The passive approach promotes designs intended tosupport high-speed operating behavior, and then attemptsto mitigate a roadway’s hazards through the use of signsand pavement markings. The problem that emerges, how-ever, is that signs and roadways are often communicatingcontradictory information. The result is that the majorityof drivers in urban areas disregard posted speed limits, andseem to learn to disregard road signs altogether, even whenthey display information that is essential to their safety(Chowdhury et al., ; Fitzpatrick, Carlson, et al., ;Fitzpatrick, Shamburger, et al., ; Kubilins, ; Tarriset al., ). Further, even when drivers are deliberatelyattempting to obey speed restrictions, they instinctivelyincrease their operating speed to their perception of a road-way’s safe speed when their concentration is focused onsomething other than actively monitoring their vehicle’s

Dumbaugh: Safe Streets, Livable Streets

speedometer (Recarte & Nunes, ). This latter findingsuggests that even conscientious drivers may be unable tocomply with posted speed limits when roadways are designedfor higher-speed operation.

A key point of departure for positive design is that itopenly recognizes that drivers use the total informationprovided by their environment—not just posted speedlimits—and strives to take advantage of these opportunitiesto provide drivers with the information they need to oper-ate their vehicles safely and appropriately. On this subject,Our European counterparts, with markedly safer roadwaysthan the United States, have developed a potentiallyvaluable alternative.

European designers use an “environmental referencespeed” when designing a roadway, beginning the designprocess by tightly specifying the desired operating speed ofa roadway, and then using this intended operating speed asthe roadway’s design speed, providing posted speed limitsthat match (FHWA, ; Lamm et al., ). Roadwaysare thus designed to be self-explaining and self-enforcing,conveying a single and consistent message to the driver onsafe operating behavior.

Further, European designers view high-speed drivingas incompatible with the safe operation of urban roadways.For all streets with any concentration of roadside develop-ment or anticipated pedestrian activity, design speeds areseverely restricted, rarely exceeding km/h ( mph). As a FHWA scan of European design practice concluded:

[European] countries have very high safety goals (rang-ing from zero fatalities to reduction of more than

percent for all crashes) that guide the design approachand philosophy. To achieve these goals, planners arewilling to provide roadways that self-enforce speedreductions, potentially increase levels of congestionand promote alternative forms of transportation. Thisapproach contrasts with the U.S. design philosophy, inwhich wider roads are deemed safer, there is a heavierreliance on signs to communicate the intended message,and there is a lower tolerance for congestion and speedreduction. (FHWA, 2001, p. viii)

The European approach is achievable because design-ers explicitly recognize that a roadway’s environmentalcontext plays a key role in determining its safe design andoperation, and they have developed design practices aimedat linking specific design values to their correspondingphysical and operational contexts. German designers, forexample, use a -celled functional classification systemthat accounts for not only mobility and access, but alsovariations in a roadway’s design environment and the needs

of a diverse set of user groups (Lamm et al., ). Thus,practicing designers are provided with clear guidance onthe safe and appropriate design of roadways that addressdesign needs for a range of physical and environmentalcontexts.

Conversely, U.S. practice applies an extremely coarse,three-tiered functional classification system that categorizesroadways exclusively according to their vehicle access ormobility functions (AASHTO, ), resulting in theproblem that many of the urban roadways classified asminor arterials serve a variety of purposes other thanhigher-speed mobility functions. Roadways designatedas minor arterials cover a wide range of physical environ-ments, and there is little guidance detailing which designvalues are most appropriate in each context (see Figure ).Indeed, the lowest recommended design speed for an urbanminor arterial in the United States ( mph) is simultane-ously the highest design speed that would be applied on asimilar roadway in a European city (AAHSTO, ; Lammet al., ).

Despite a host of problems associated with applyingthe U.S. functional classification system in an urban envi-ronment (de Cerreno & Pierson, ; Forbes, ; Ku-bilins, ; Meyer & Dumbaugh, ), many designengineers have made notable strides in this area throughthe use of context-sensitive solutions, an approach thatincorporates stakeholders in the project design and devel-opment process (FHWA, ; TRB, ). One problemthat emerges, however, is that determining whether a spe-cific design approach is appropriately safe is ultimately amatter of professional engineering judgment, not an out-come of public involvement activities. On this subject,designers are forced to navigate the uncharted waters ofurban road safety alone.

And increasingly, many practicing designers are doingso. There are a growing number of examples of designengineers who have chosen to thoughtfully strike out ontheir own, moving beyond the conventional definition of“safe” design practice to develop new strategies for address-ing the twin goals of safety and livability. Five such exam-ples are included in this study alone. Yet any success in thisarea has occurred in spite of passive safety practices, notbecause of them. It is the obligation of future researchers tobegin to more fully develop our understanding of how tosafely design urban roadways, and to ensure that this in-formation is better disseminated throughout the profession.A positive approach to transportation design would appearto be a key means of doing so.

Finally, this study does not suggest that certain urbanroadways can not or should not be designed to addressmobility needs. But it does suggest that we must move

Journal of the American Planning Association, Summer , Vol. , No.

beyond the assumption that the use of “forgiving” designvalues necessarily equates to enhanced safety, and to beginreconsidering the role that driver behavior may have on aroadway’s safety performance, particularly in urban envi-ronments. Substantial opportunities for enhancing bothsafety and livability remain to be explored.

Safe Streets, Livable Streets

At the most fundamental level, the major tension inthe design of urban roadways does not appear to be a matterof balancing safety and livability objectives. There is littleevidence to support the claim that “livable” streetscapetreatments are less safe than their more conventionalcounterparts, and the weight of the evidence suggests thatthey can possibly enhance a roadway’s safety performance.Instead, the more basic problem appears to be that safetyand livability objectives are often in direct conflict with theoverarching objective of mobility, and its proxy—speed.

The passive approach to transportation safety beganwith the observation that the Interstate Highway Systemproduced fewer crashes and injuries than other roadwayclasses, and attributed this safety performance to the use ofhigher-speed, more “forgiving” design values. Yet it mustbe recognized that the safety performance of the Interstatesystem is probably better explained by the fact that theseroadways physically restrict access, channel vehicle move-ments, and limit their use to a single user type—motorists—than because they permit higher operating speeds.

Conventional safety practice attempts to superimposethese high-speed, limited-access design characteristics onother roadway types, but it is not at all clear that thesedesigns are either safe or appropriate in an urban context.At the most basic level, the primary function of cities, andthus the streets that serve them, is to concentrate compati-ble developments and activities together and to encouragea high degree of access between them, traditionally throughnonmotorized modes. High-speed, limited-access roadwaysare inherently antithetical to these purposes.

I have argued that many of the safety concerns thatemerge on urban streets result from design practices thatfail to link a roadway’s design to its environmental context,thereby providing motorists in urban environments with afalse sense of security and increasing their potential expo-sure to crashes and injuries. I have further provided a theo-retical framework that better accounts for the safety anom-alies one observes when examining the literature and dataon the crash performance of urban roadways. Yet theory isonly the first step. There is a clear and demonstrated needto better develop our professional understanding of the

relationship between driver behavior and transportationsafety, as well as to enhance our overall approach to thedesign of urban roadways. This study thus concludes withthe hope that by better understanding the relationshipbetween design, driver behavior, and safety, we can designroadways that are not only safe, but also livable.

Dumbaugh: Safe Streets, Livable Streets

Figure . Three urban minor arterials.

AcknowledgmentsThis article benefited from the comments and recommendations ofmany colleagues. Particularly, I would like to thank Michael Meyer andMichael Dobbins from the Georgia Institute of Technology; SusanHerbel with the TSS Group; Jane Lim-Yap, Ian Lockwood, and WalterKulash at Glatting-Jackson; Jonathan Lewis at Jordan, Jones andGoulding; Stephanie Macari with the Montgomery County PlanningCommission; and Quentin Kruel with Long and Foster. I would alsolike to especially thank JAPA’s editors and anonymous reviewers, whoprovided invaluable comments on an earlier draft of this article.

Notes. In conventional engineering parlance, all roadways are referred to ashighways. Conventional, high-speed highways are referred to as freeways.. While the AASHTO () Green Book permits the use of a .-foot“operational offset” on urban arterials, it is important to recognize thatthis is intended only to prevent motor vehicles from hitting theirmirrors on roadside objects during normal operating conditions, and isnot intended or perceived as having a meaningful relationship to safety.As stated in the Green Book: “Clear roadside design is recommendedfor urban arterials wherever practical” (p. ).. The finding with the potentially most profound influence on roadsidesafety is also the one that receives the least attention. The authors notedthat “curves were highly over-represented in tree accidents. Almost %of tree accidents were related to a curve. This is startlingly high consid-ering that probably no more than % of all street mileage in the city ofHuntsville is curved” (Turner & Mansfield, , p. ). In response tothis finding, the authors recommended prioritizing tree eliminations atcurves. While such an approach may go a long way towards reducinginjuries in run-off-road events, it fails to ask the potentially more im-portant question: why are run-off-roadway events more likely to occurat curves in the first place? It would seem unlikely that this remedialaction—eliminating the tree—will have any effect on eliminating therun-off-roadway event, which may nevertheless result in an injuriouscrash, such as a rollover, regardless of whether a tree is present. Asevidenced in Milton and Mannering (), the problem is not thecurve itself, but a curve that it is preceded by a straight (high-speed)approach.. Hauer () does note that “I am not convinced that if research wasdone on current data, that -foot lanes would be found to be less safethan -foot lanes. Much has changed since then; trucks grew to belarger and research methods improved. Hovever, at the time the Policywas written, the aforementioned findings by respected researchers shouldhave sounded alarm” (p. ).. While Noland and Oh’s () study focused primarily on ruralobservations, two specific findings bear mentioning. Shoulders had anambiguous relationship to safety, with wider shoulders being associatedwith a decrease in total crashes but increases in fatal ones. While thesefindings do not directly address safety in urban environments, they dosuggest that increasing shoulder widths may increase vehicle speeds,thereby increasing crash severity, if not frequency.. My treatment of this topic skirts over a rich and interesting historythat deserves a more thorough treatment than can be given here.Interested readers should are encouraged to see Weingroff () andGladwell (), both of which are not only highly informative, butsurprisingly compelling.. Practicing engineers will undoubtedly recognize the similarity betweenNader’s hypothetical “drunk looking out the window” and the defini-tion of the “design driver” used in contemporary design practice.

. This is evidenced in the fact that while our methods for the crashtesting of vehicles and roadside hardware have become increasinglyelaborate in the past years (see Transportation Research Board, ,for current test standards), there has been little advancement in ourunderstanding of the behavioral factors that cause crashes to occur(Kanellaidis, ; Noland, ).. While a full treatment of the subject of design speed is beyond thescope of this study, the important fact is that a roadway’s design speed isthe controlling element in its design. Once a design speed is selected, allother geometric features, such as lane widths and clear zones, are designedto conform to the adopted design speed. Thus, higher design speedsencourage the use of higher minimum values for all other geometricfeatures as well.. Examining Fatality Analysis Reporting System (FARS) data for minorarterials, collectors, and local roadways is revealing. In , for example,half of all individuals killed in a fixed-object crash on these road classeswere between the ages of and , and fully % of the total crashesinvolved males in this age group. Females account for less than a thirdof the fatalities in all age brackets except the and older group, wheremale and female fatalities equalize, undoubtedly the result of the factthat at these ages, personal motor functions and reaction times beginto decline. When one considers this information holistically, it suggeststhat fixed-object fatalities may not be a design problem as much as theyare a reflection of broader demographic and sociocultural factors, suchas a propensity of young males to engage in higher-risk behavior.. To calculate a roadway’s VMT for the -year study period, I de-termined average ADT for each roadway milepost, and then used themedian ADT to derive an overall estimate of VMT for the road seg-ment. The median was selected as a better measure of central tendencythan the average because several small roadway segments had unusuallyhigh ADTs, thus skewing the overall averages. Once median ADT wasdetermined, VMT was calculated as: VMT = Median ADT × × ×Section Length.. The exception is Pine Avenue (State Road ) in Ocala. Only .miles of roadway were available for the northern comparison sectionbecause a substantial (-mile) segment is currently off the state system.To acquire miles of comparison roadway data, I used averages for a.-mile section to the north and a .-mile section to the south.. Design consistency, a phrase often used by designers to discuss howthey address safety through design, also emerged in the guide, whichstates: “consistency in design standards is desirable on any section of road,because problem locations are generally at the point where minimumdesign treatment is used” (p. ). Restated another way, design consis-tency, as it was originally conceived, encourages the consistent adoptionof high design values.. In the and editions of AASHTO’s Green Book, thesections dealing with these subjects contain no data, nor has a wordbeen changed.. A few statistics bear mentioning. In , the year that passive safetyprinciples first became embedded in contemporary practice, the U.S. hadfewer transportation-related fatalities per capita ( per , popula-tion) than all other countries except Great Britain ( per ,). By, the U.S. ( fatalities per ,) remained behind Great Britain( per ,), but had also fallen behind the entirety of the EuropeanUnion ( per ,), Australia ( per ,), Japan ( per ,),and, indeed, the rest of the developed world (NHTSA, n. d.; WorldHealth Organization, ). While these statistics are alarming, they alsosuggest that promising new opportunities for enhancing transportationsafety remain to be explored.

Journal of the American Planning Association, Summer , Vol. , No.

ReferencesAmerican Association of State Highway Officials. (). Highwaydesign and operational practices related to highway safety. Washington,DC: Author.American Association of State Highway and Transportation Officials.(). Highway design and operational practices related to highway safety.Washington, DC: Author.American Association of State Highway and Transportation Officials.(). A policy on geometric design of highways and streets (rd ed.).Washington, DC: Author.American Association of State Highway and Transportation Officials.(). Highway safety design and operations guide. Washington, DC:Author.American Association of State Highway and Transportation Officials.(). A policy on geometric design of highways and streets (th ed.).Washington, DC: Author.American Association of State Highway and Transportation Officials.(). Roadside design guide. Washington, DC: Author.Chowdhury, M. A., Warren, D. L., Bissell, H., & Taori, S. (,February). Are the criteria for setting advisory speeds on curves stillrelevant? ITE Journal, –.de Cerreno, A. L. C., & Pierson, I. (). Context sensitive solutions inlarge central cities. New York: Rudin Center for Transportation Policyand Management.Duany, A., Plater-Zyberk, E., & Speck, J. (). Suburban nation: Therise of sprawl and the decline of the American dream. New York: NorthPoint Press.Ewing, R. (, March). Pedestrian and transit friendly design. Talla-hassee: Public Transit Office, Florida Department of Transportation.Ewing, R. (). Traffic calming state-of-the-practice. Washington, DC:Institute of Transportation Engineers.Federal Highway Administration. (, September). A user’s guide topositive guidance (rd ed.; Report No. FHWA-SA--). Washington,DC: U.S. Department of Transportation.Federal Highway Administration. (). Flexibility in highway design(Publication No. FHWA-PD--). Washington, DC: U.S. Depart-ment of Transportation.Federal Highway Administration. (). Geometric design practices forEuropean roads (Publication No. FHWA-PL--). Washington, DC:U.S. Department of Transportation.Fitzpatrick, K., Carlson, P., Brewer, M., & Wooldridge, M. D. ().Design speed, operating speed, and posted speed limit practices. InTransportation Research Board, nd Annual Conference Proceedings[CD-ROM]. Washington, DC: Transportation Research Board.Fitzpatrick, K., Shamburger, B., & Fambro, D. (). Design speed,operating speed, and posted speed survey. Transportation ResearchRecord, , –.Florida, R. (). The rise of the creative class: And how it’s transformingwork, leisure, community and everyday life. New York: Basic Books.Forbes, G. (, December). Urban roadway classification: Beforethe design begins. In Urban Street Symposium Conference Proceedings(Transportation Research E-Circular Number E-C), B/–B/.Frank, L. D., Engelke, P., & Schmid, T. (). Health and communitydesign: The impact of the built environment on physical activity. Washing-ton, DC: Island Press.Frank, L. D., Stone, B., & Bachman, W. (). Linking land use withhousehold vehicle emissions in the Central Puget Sound: Methodologi-cal framework and findings. Transportation Research D, , –.Gladwell, M. (, June ). Wrong turn: How the fight to makeAmerica’s roadways safer went off course. New Yorker, –.

Hauer, E. (a). Safety in geometric design standards. Retrieved January, , from http://members.rogers.com/hauer/Pubs/Safety%in%Geometric%Design.pdfHauer, E. (b). Overview. In Institute of Transportation Engineers,The traffic safety toolbox: A primer on traffic safety (pp. xi–xvi). Washing-ton, DC: Institute of Transportation Engineers.Ivan, J., Raghubhushan, N., Pasupathy, K., & Ossenbruggen. P. J.(). Differences in causality factors for single and multi-vehiclecrashes on two-lane roads. Accident Analysis and Prevention, , –.Jacobs, A. B. (). Great streets. Cambridge, MA: MIT Press.Jacobs, J. (). Death and life of great American cities. New York:Modern Library.Kanellaidis, G. (, January/February). Human factors in highwaygeometric design. Journal of Transportation Engineering, –.Kubilins, M. A. (, December). Designing functional streets thatcontribute to our quality of life. In Urban Street Symposium ConferenceProceedings (Transportation Research E-Circular Number E-C),B/-B/.Kuhn, T. S. (). The structure of scientific revolutions (rd ed.). Chicago:University of Chicago Press.Kratzke, S. R. (, February). Regulatory history of automatic crashprotection in FMVSS . Paper presented at the Society of AutomotiveEngineers Congress and Exposition, Detroit, MI.Lamm, R., Psarianos, B., & Mailaender, T. (). Highway design andtraffic safety engineering handbook. New York: McGraw Hill.Lee, J., & Mannering, F. (, December). Analysis of roadside accidentfrequency and severity and roadside safety management. Olympia: Wash-ington State Department of Transportation.McLean, J. (). Review of the development of US roadside designstandards. Road Transport and Research, (), –.Meyer, M. D., & Dumbaugh, E. (, July). Institutional and regula-tory factors related to non-motorized travel and walkable communities(Report Commissioned by the Transportation Research Board and theInstitute of Medicine). Available at: http://trb.org/downloads/srpapers/srpaperstoc.pdfMilton, J., & Mannering, F. (). The relationship among highwaygeometries, traffic-related elements and motor-vehicle accident frequen-cies. Transportation, , –.Nader, R. (). Unsafe at any speed: The designed-in dangers of theAmerican automobile. New York: Grossman.Naderi, J. R. (). Landscape design in the clear zone: The effects oflandscape variables on pedestrian health and driver safety. In Transpor-tation Research Board, nd Annual Conference Proceedings [CD-ROM].Washington, DC: Transportation Research Board.National Highway Traffic Safety Administration. (n. d.). FatalityAnalysis Reporting System (FARS) web-based encyclopedia. Available at:http://www-fars.nhtsa.dot.gov/Noland, R. B. (). Traffic fatalities and injuries: Are reductions theresult of “improvements” in highway design standards? In Transporta-tion Research Board, th Annual Conference Proceedings [CD-ROM].Washington, DC: Transportation Research Board.Noland, R. B. (). Traffic fatalities and injuries: The effects of changesin infrastructure and other trends. Accident Analysis and Prevention, ,–.Noland, R. B., & Oh, L. (). The effect of infrastructure and demo-graphic change on traffic-related fatalities and crashes: A case study ofIllinois county-level data. Accident Analysis and Prevention, , –.Ossenbruggen, P. J., Pendharkar, J., & Ivan, J. (). Roadway safetyin rural and small urbanized areas. Accident Analysis and Prevention, ,–.

Dumbaugh: Safe Streets, Livable Streets

Recarte, M. A., Nunes, L. (). Mental load and loss of control overspeed in real driving: Towards a theory of attentional speed control.Transportation Research Part F, , –.Rosenberg, C. E. (). The cholera years. Chicago: University ofChicago Press.Skene, M. (). Traffic calming on arterial roadways? In Institute ofTransportation Engineers, th Annual Meeting Compendium of Techni-cal Papers [CD-ROM]. Washington, DC: Institute of TransportationEngineers.Tarris, J. P., Mason, Jr., J. M., & Antonucci, N. (). Geometric designof low-speed urban streets. Transportation Research Record, , –.Transportation Research Board. (). Recommended procedures for thesafety performance evaluation of highway features (NCHRP Report ).Washington, DC: National Research Council.Transportation Research Board. (). A guide to best practices forachieving context sensitive solutions (NCHRP Report ). Washington,DC: National Research Council.Transportation Research Board. (). Guide for addressing run-off-road collisions (NCHRP Report , Vol. ). Washington, DC: Na-tional Research Council.Turner, D. S., & Mansfield, R. R. (). Urban trees and roadsidesafety. Journal of Transportation Engineering, (), –.Viano, D. C. (). Lap-shoulder belts: Some historical aspects. In D.Viano (Ed.), Seat belts: The development of an essential safety feature (pp.–). Warrendale, PA: Society of Automotive Engineers.Weingroff, R. F. (). President Dwight D. Eisenhower and the federalrole in highway safety. Retrieved January , , from http://www.fhwa.dot.gov/infrastructure/safetypr.htmWilde, G. J. S. (). The theory of risk homeostasis: Implications forsafety and health. Risk Analysis, , –.Wilde, G. J. S. (). Risk homeostasis theory and traffic accidents:Propositions, deductions and discussion of dissension in recent reactions.Ergonomics, (), –.Wilde, G. J. S. (). Target risk. Toronto: PDE Publications.World Health Organization. (). World report on road traffic injuryprevention. Retrieved January , , from http://www.who.int/violence_injury_preventionZein, S. R., Geddes, E., Hemsing, S., & Johnson, M. (). Safetybenefits of traffic calming. Transportation Research Record, , –.

Counterpoint

J. L. GattisGattis is a professor at the Mack-Blackwell TransportationCenter, University of Arkansas, Fayetteville.

It has taken many decades for roadway designers tobegin to recognize that how road and road environ-ments are designed affects safety, and to identify what

particular features enhance or detract from safety in a givenenvironment. This process is still evolving. Historically, therehas been a tendency for those funding research to focus onrural or high-speed environments, and on pavements and

structures. One outcome of the resulting underemphasis onsafety and urban design concerns, as author Eric Dumbaughidentified in this article, is the problem with the publishedresearch about urban roadside design in the U.S.: it is lim-ited in both scope and quantity. With these limitations, itis understandable that city street designers extrapolate fromprinciples learned in a rural highway environment. Theymay not be standing on the firmest ground when they dothis, but they judge that it is the best ground they have.

While the author’s initial focus is on urban roadsidedesign, he eventually broadens the scope of the article toconsider underlying philosophies and assumptions of whydrivers choose a certain speed or exhibit other behaviors.He and I agree on certain points, such as that currentresearch is inadequate and a better understanding of urbanroadway design and driver behavior interactions is needed,and that other paradigms may bear consideration. How-ever, I question some of his statements in the article:

• In discussing the Huntsville study (Turner & Mans-field, ), the author states “. . . it does not lead tothe conclusion that eliminating trees . . . will have anyeffect on a roadway’s safety. Such conclusions can onlybe made by examining the actual crash performanceof eliminating trees in urban areas. . . .” Would not acomparison of roadways similar except for the absenceor presence of roadside objects such as trees constitutea valid comparison?

• Does the Toronto study (Naderi, ) actually showthat trees in concrete planters resulted in crash reduc-tion, or that they were instead associated with a reduc-tion? While inferences from association can be valid,they do not always equate with causality.

• Contrary to the author’s claims, I believe that growingawareness of the need to engineer safety into roadwaysdid not necessarily shift the focus away from driverbehavior, but rather expanded the focus to be moreinclusive.

• The author suggests that an active approach to safety—constraining the roadway to communicate the needto slow down—might be more effective than thepassive approach, which accommodates and perhapsencourages “extreme driving behavior.” It is mislead-ing to state that the passive approach attempts toaccommodate high-speed, extreme behavior. Speedstudies typically reveal that most drivers on a givenroadway fall within a rather narrow band of speeds.When roadway designers design for the th or thpercentile speed, they are designing for a speed that iswithin a few miles per hour of what most drivers chooseto drive at, which is hardly extreme. For example, refer

Journal of the American Planning Association, Summer , Vol. , No.

to the accompanying graph, showing the plottedcumulative speed distribution from five city streetsin Fayetteville, Ft. Smith, and Little Rock, Arkansas.There is approximately a - to -mph difference be-tween the th percentile speed and the mean speed,which means there is very little difference betweendesigning for the th percentile driver and designingfor the average driver.

• The author’s appeal to risk homeostasis theory em-ploys an a priori assumption. First, we need to ques-tion whether this theory applies to driving behavior.Observation would suggest there is elasticity in theamount of risk drivers accept, since drivers seem to bewilling to increase risk to achieve some reward, such assaving time. Used in this context, this theory almostsuggests purposefully increasing the driving hazard inorder to improve safety. If we wanted to discouragepeople from running in grocery store aisles, would wethrow down more banana peels?

It should be noted that some recent American researchhas tried to examine why drivers choose a certain speed inan urban environment, and what factors might commu-nicate to them that they should slow down, but this is adifficult issue to study, much less to resolve. For the pres-ent, we are left with conflicting concepts about what effectscertain design features will have on the safety of a given

urban street. To better understand where and under whatconditions various urban aesthetic streetscape treatmentsare benign or even helpful, resources must be reallocated toimprove both urban roadway data systems and safety analysis.

A problem one has with trying to draw conclusionsfrom limited and sometimes seemingly contradictorystudies is that of comparing apples to oranges. Researchsuggests that crash rates are affected by multiple factors,such as traffic volume, width, speed, presence of parking,type of roadside development, and access frequency. Crashstudies do not always consider these nuances, but due to theeffort needed to have a suitable sample size, it is certainlyunderstandable that only some factors are accounted for.Even in the author’s own data, there were somewhat mixedresults (e.g., Woodland Avenue in DeLand, Florida). Ifaccess frequency for these roadways were to be factored in,still a different finding could have appeared. In short, morecontext-sensitive research is a prerequisite for context-sensitive design.

Also not to be overlooked in a discussion of the author’sinitial issue is the fact that there can be other problems withroadway landscaping. Motorists pulling out of side streetsand driveways encounter landscaping that has been installedin such a way that it obstructs their view of oncoming traf-fic. Also, landscaping in the wrong place can restrict motor-ists’ ability to see pedestrians or traffic-control devices.

Perhaps there is another lesson to learn from theStonex report. What is most often referenced from the

Dumbaugh: Safe Streets, Livable Streets

Cumulative speed distribution for five city streets in Arkansas.

%

%

%

%

%

%

%

%

%

%

%

Breckenridge

Brooken Hill

Mall-Shiloh

Pine Valley

Salem

Speed (mph)

Percentile

report of GM Proving Ground experience is the -footclear zone, but what is overlooked is perhaps more signifi-cant. Stonex reported that even with professional driverson a closed course, with a statistical probability driverswould sooner or later lose control and have an accident. Asociety that understands this lesson recognizes that driversare human and sooner or later make mistakes, and tries toincorporate safeguards into roadway design.

Livable streets advocates sometimes ignore the fact thatmobility is also a part of quality of life. In the American

cities I am familiar with, only a small fraction of the streets(the major and minor arterials) are intended for highervolumes and speeds, while the great majority of the streetmiles are for lower volumes and speeds. This small percent-age of arterial streets is all that people have to rely on to getto their jobs, schools, and other destinations safely andwithout delay. Policies and design features that impedetravel on the few available corridors of mobility, or maketravel more dangerous, adversely affect the quality of lifefor all people.

Journal of the American Planning Association, Summer , Vol. , No.