Safety Science 48 (2010) 1175–1181

Contents lists available at ScienceDirect

Safety Science

journal homepage: www.elsevier .com/locate /ssc i

Beyond ‘‘best practice” road safety thinking and systems management– A case for culture change research

Ian Johnston *

Monash University Accident Research Centre, Monash University, Victoria, Australia

a r t i c l e i n f o a b s t r a c t

Article history:Received 1 September 2009Received in revised form 26 November 2009Accepted 12 December 2009

Keywords:Road use cultureInstitutional behaviourSafe system modelRoad safety best practiceRoad safety management system

0925-7535/$ - see front matter � 2009 Elsevier Ltd. Adoi:10.1016/j.ssci.2009.12.003

* Tel.: +61 419 350 930.E-mail address: irj111@bigpond.net.au.

Forecasts of an ‘‘explosion” in the level of road trauma as many of the world’s most populous nations rap-idly motorise have contributed to the urgency of documenting best practice thinking and managementsystems to improve road safety performance. By examining the success stories of nations with eitherthe lowest fatailty rates per head of population or with the largest recent improvements a number of ‘‘les-sons” have emerged. Data-driven problem identification and the development of evidence-driven coun-termeasure packages formalised in a strategy for effective implementation, with ambitious, quantitativetargets and transparent lines of institutional accountability are the critical success factors. Basing thisplanning upon the cornerstones of the safe system conceptual model constitutes best practice thinking.This paper argues that critical elements of the safe system model are in discord with behavioural mores inthe cultures of many western motorised nations and that this hampers the adoption of the most effectivesafety programs within key institutions and within political systems. A case is made for a systematicexamination of car use and safety cultures in order to add a further dimension to best practice learnings.

� 2009 Elsevier Ltd. All rights reserved.

1. Introduction

Most western motorised nations have made substantial advancesin reducing death and serious injury from road crashes over the lastfew decades. In some, but by no means in all, the gains have been inthe absolute number of fatalities, not just in fatality rates, with sub-stantial international variation in the degree of progress. Table 1illustrates the performance gulf between Australia and the UnitedStates of America over the last decade or so. It is an update of datathe author has previously published (FHWA, 2006).

Performance variations of this kind, a rising sense of urgencyresulting from a decade-long period of little progress in most na-tions (FlA Foundation, 2006) and the implications of the rapidmotorisation occurring in some of the most heavily populated na-tions in the world (WHO, 2004) appear to have driven an interestin identifying and documenting best practice. Several recent at-tempts have been made (WHO, 2004; FHWA, 2006; Koornstraet al., 2002; Wegman et al., 2006; OECD, 2002, 2008).

2. Which are the best practice road safety managementnations?

The European research focussed initially on the ‘‘SUN” countries –Sweden, United Kingdom and the Netherlands (Koornstra et al.,

ll rights reserved.

2002; Wegman et al., 2006) – which, consistently over more than adecade now, have recorded the lowest population-based trafficcrash fatality rates among motorised countries. While this is an emi-nently sensible starting point, the lowest rate does not necessarilyimply the existence of best practice road safety thinking and man-agement. Trinca et al. (1988), and more recently Johnston (2006),examined the difficulties with international comparisons, evenamong western motorised nations, highlighting the importance ofrelative level of motorisation, nation size, topography, road usermix and the nature of the prevailing land transport system. WithinAustralia, for example, the Australian Capital Territory has a mark-edly lower rate than the SUN countries (http://www.infrastruc-ture.gov.au/roads/safety/road_fatality_statistics) but it is small,almost entirely urban, has a higher income level than other States(leading to a more modern vehicle fleet) and has a very high standardroad system. These structural factors strongly influence the ex-pected safety performance within that jurisdiction.

The SUN countries have similar motorisation rates (between540 and 580 vehicles per 1000 population in 2007) and these ratesare at the low end of the scale among western motorised nations(Australia and New Zealand have rates over 700 and the USA rateis over 800). Higher motorisation results in higher exposure to riskfor most classes of road user and this must partially explain theinternational differences in rates which, as Table 2 shows, have ex-isted for a long time.

In one sense it does not matter whether the lowest rates trulyreflect the existence of best practice. The initial SUNflower study

Table 2Road crash fatality rate improvements over time for four European and four‘‘new world” nations. Source: www.infrastructure.gov.au/roads/safety/road_fatality_statistics.

Road crash fatality rates per 100,000 population

1995 2007 % Improvement (rounded)

European nationsFrance 15.3 7.5 51%Great Britain 6.4 5.0 22%Netherlands 8.6 4.3 50%Sweden 6.5 5.2 20%

New world nationsAustralia 11.2 7.6 32%Canada 11.4 9.1 20%New Zealand 15.9 10.0 37%USA 15.9 13.6 15%

Table 1The absolute number of deaths and the population-based death rates from trafficcrashes in the United States of America and Australia. Source: www.infrastruc-ture.gov.au/roads/safety/road_fatality_statistics.

Number ofdeaths

Death rate per 100,000population

United States of America1995 41,798 15.92007 41,059 13.6Percentage improvement 1.8% 14.5%

Australia1995 2013 11.22007 1603 7.6Percentage improvement 20.3% 32.1%

1176 I. Johnston / Safety Science 48 (2010) 1175–1181

examined the interventions applied and the evidence of effective-ness, together with operational issues in application and concep-tual methods of approach (Koornstra et al., 2002). Seekinglessons from the safest nations is an excellent starting point.

A supplementary method for identifying potential best practicenations is to identify those with the greatest gains in safety overthe last decade or so, the assumption being that these nations haveintroduced new, effective management systems. Table 2 comparesfour European nations (including the three SUN countries) and four‘‘New World” nations. The Netherlands and France stand outamong European nations as dramatic improvers along with NewZealand and Australia among the ‘‘new world” nations. Of course,as Wegman et al. (2006) point out, the greatest rates of improve-ment are more likely to be found in those nations with a poorerstarting point. Nevertheless, useful road safety management les-sons can be found in the experience of the rapid improvers.

3. Scope and purpose

This paper explores the lessons from the ‘‘best practice” re-search and considers their implications in the context of (assumed)prevailing car use and safety cultures, seeking to promote system-atic future research to fill an important current void. It is confinedto western, motorised nations where the bulk of the relevant re-search has been done. It should be noted that little scientific workon safety culture in road use has been conducted (Wiegmann andvon Thaden, 2007) so the descriptions of extant culture arespeculative.

4. The broad culture around road safety

In most western motorised nations from the mid 1950s throughthe early 1970s road safety strategic thinking and planning was theprovince of government-appointed coalitions (‘‘councils”) of con-

cerned citizens; police forces; and road and traffic, particularlytraffic, engineers. Crashes tended to be viewed simplistically ashaving one primary cause, blameworthy human behaviour, andthe countermeasures were reactive, small-scale, isolated, directedlargely at symptoms and with a clear tendency to be punitive.

Gradually road crashes became the subject of scientific study.Ironically however, the early in-depth, crash research, while pro-foundly influencing the thinking of the researchers, appears tohave helped perpetuate the simplistic view of the public and thebureaucratic and political decision makers and a generation of roadand traffic engineers. The public catchcry was – and still is – that95% of crashes were caused by error or misbehaviour, which is,of course, the principal finding of these pioneering studies. Not sur-prisingly in an open loop system where the human is, in effect, theonly active element, it was indeed rare to investigate scientificallya crash in which no operator behaviours could be found among themultiple causal elements identified. However, the importance ofthe interactions between road, traffic and vehicle features and par-ticular types of behaviour never featured in public discussion of theresearch and rarely appeared in briefings to policymakers. Further,the concept of latent conditions within the road transport systemrendering certain types of error more likely had not found itsway from the study of industrial safety (Reason, 1997).

Coupled with a widespread ‘‘blame the victim” mindset withincommunities was a relative apathy towards the extent of deathand serious injury on the road. Williams and Haworth (2007) offerthree explanations: the constant ‘‘drip feed” of daily events ensuresthe ‘‘road toll” becomes an accepted part of the road transportlandscape; the accurate perception that the (personal) risk of acasualty crash on any given trip is remote; and the belief in one’spersonal ability to control risk, the problem being ‘‘the otherdriver”. While personal safety is valued, community safety is not(Johnston, 2004).

This historical perspective is important to understanding thesocial context in which attempts to advance strategic thinkinghave had to struggle with widespread, accepted views that if poorbehaviour is the principal cause behaviour change must be the pri-mary answer (Howard and Sweatman, 2007). The prevailing com-munity culture around road safety needs to be explicitly addressed.Rose (1992) wrote a superb book over fifteen years ago outliningthe strategy of preventive medicine and while road safety wasnot his primary focus it featured in his illustrative case studies.His thesis – that a relatively low level of individual risk spread rel-atively uniformly across an entire population requires population-wide measures to achieve an effect at the community level – is aperfect prescription for most road safety problems yet it is stilluncommon as a strategy for behaviour change among road users.

5. The transport setting and its cultural influence

Road safety has traditionally been an issue for the transport sys-tem to address and there was, for a long time, no systematicengagement with either the field of public health or the field ofoccupational health and safety. Ironically, the World Health Orga-nisation is now playing a lead role in promoting global action asroad trauma assumes a more prominent position in the global bur-den of disease, yet it lacks the networks meaningfully to influenceinstitutional change among the transport players (WHO, 2004,2009). Watkins has argued that, in emerging nations, governments‘‘. . . measure the success of their roads policies in kilometres of asphaltand the speed at which goods can be carried” and that ‘‘road safety isthe poor cousin of road building” (Watkins, in Ward 2009).

The efficient movement of people and freight is the over-arch-ing goal of road transport with the level of death, injury and prop-erty damage being one of a number of by-products to be managed.

I. Johnston / Safety Science 48 (2010) 1175–1181 1177

This conception of death and injury as an undesirable but neces-sary by-product of road transport appears to have dominatedthinking in most motorised nations in the early days of rapidmotorisation, still dominates in some (notably the USA), and is acommon characteristic of emerging nations (Sleet et al., 2007).

Rarely were safety targets set and when they were it was interms of the rate of death and serious injury (typically death alone)per unit of road use. As road use continues to grow, nations havecontinuously to improve their road and transport systems to copewith the need for more capacity and the need to minimise delay.Efficiency-inspired improvements in vehicles, roads and trafficmanagement systems contribute to an ongoing improvement insafety per unit road use, over and above the effect of any specificsafety measures; for example converting a two lane road to a fourlane divided road to meet capacity demands dramatically improvessafety by eliminating the risk of head-on crashes. However, if thesafety gains do not outweigh the growth in road use then the abso-lute number of deaths must increase. That has been the case in theUSA (FHWA, 2006 and Table 1) where the absolute number of cit-izens being killed has not meaningfully decreased.

Public policy around road transport in the USA appears to ac-cept that a commensurate price must be paid for what is, by a con-siderable margin in terms of vehicles per head of population, thehighest level of personal mobility in the motorised world. Judge-ments of priority among competing goals are, of course, the dailybusiness of governments on behalf of their communities. The USpublic policy around road transport where unfettered personalmobility appears the primary goal implies the absence of a wide-spread, public constituency for safety (Kissinger, 2007).

Most western motorised nations have moved from using, astheir main performance measure, deaths per unit road use todeaths per head of population; the common public health measureof the burden of disease, in this case the disease of mobility. Themove appears to have had two drivers – the need for a measurethat facilitated direct comparison with other public health issuesand a desire to break from the mindset that assumes that a pricefor personal mobility in death and serious injury is inevitable.The USA remains one of the few nations to rely on deaths per unitroad use (100 million miles driven) as its principal safety perfor-mance measure.

Elvik (2003) has shown that what matters most in creatingcommitment and action is the setting of ambitious, quantitativetargets. The nations considered as being at the leading edge ofsafety thinking and performance set their safety performance tar-gets in terms of reductions in the absolute number of deaths andserious injuries, not merely as reductions in population-basedrates. This forces a shift in the institutional mindset from one ofmanaging a by-product to one of viewing safety as a fundamentaloutcome of a sustainable road transport system.

Many nations continue to rely heavily upon economic analysesof cost and benefit when choosing not only whether to introduce aparticular safety measure but the optimum level of resource to beallocated. A monetary value is placed on travel time (particularlycongestion and delay), on environmental impacts (fuel usage,emissions, noise) and on safety (death and injury) as a methodfor making economically rational trade-off decisions betweenmobility, safety and environmental impacts among competingroad and traffic investment proposals.

No distortion is introduced if the choices to be made are be-tween competing safety projects – or congestion ameliorating pro-jects – as such decisions are about best safety – or congestionreduction – value for money but complex socio-political issuesarise when the decisions involve trading safety for mobility. Swe-den is the only nation to date to have explicitly excluded the useof monetary values for death and serious injury from benefit-costanalyses of the impact of investment options. Its Vision Zero phi-

losophy requires that there be no increase in adverse safety out-comes from any proposal (Tingvall, 2004).

Debates around the measurement of the so-called externalitieswithin the road transport system, and what to include and exclude,are conducted within academe and the institutional players. Noevidence of a public debate has been uncovered, yet if the societalculture around road safety is to become an important target forchange then the debate needs to enter new arenas. The recent at-tempts by Wegman et al. (2008) to develop macro measures of anation’s road safety progress have begun this journey.

6. Best practice strategic thinking is not aligned with prevailingculture

The term safe system has come to represent the current consen-sus of what constitutes best practice strategic thinking in roadsafety (OECD, 2008). It builds upon the Swedish evolution of roadsafety strategies from their Vision Zero and the dutch developmentof the principles of sustainable safety. Its key elements, drawn lar-gely from Wegman and Aarts (2006), are:

� The vision is to prevent crashes and, if this is not possible, toreduce crash severity to a point where (severe) injury risk iseliminated (since the human body has limited tolerance tokinetic energy impacts system design should be such as to keepthe anticipated crash forces below human tolerance limits).

� All members of the population are road users, the full range ofhuman abilities and personalities exist within the population,humans are fallible operators and mistakes must be anticipatedas commonplace in the open-loop road transport system. (Vehi-cles, roads and traffic systems must be designed and operated tobe error-tolerant, particularly through increased predictabilityin system operation, and the performance levels of users mustbe raised).

� Latent errors in what is an inherently dangerous traffic systemmust be reduced.

� All of the elements in the road safety system are interrelated andresponsibility for safety must be shared across all players andefforts must be coordinated.

It is sometimes claimed, incorrectly, that adopting a safe systemmodel transfers responsibility from the individual road user to thevehicle, road, and traffic system designers. In reality, it shares theresponsibility in a way not previously achieved. The safe systemseeks error-tolerance, it does not demand designs which seek toeliminate deliberate misbehaviour. The individual road user is ex-pected to comply with prescribed (and proscribed) behaviours – tobe well-trained and licensed, to be fit to take part in traffic (unaf-fected by alcohol, other drugs or fatigue) and to comply with theoperational traffic rules such as speed limits and movement prior-ities throughout his journey.

The adoption of this conceptual model requires a level ofchange in institutional mindset that cannot be understated. Whilethere are standards – and warrants for their use – that take safetyexplicitly into account their origins are rarely based on a soundunderstanding of human behaviour and their application is basedtypically on traffic volumes. Hauer (1990), for example, describedthe origin of the method for computing sight distance in geometricroad design, a standard intended to assure adequate braking dis-tance when an unexpected object appeared on the road ahead.The standard had to be regularly modified to increase the size ofthe design object as vehicle design changes regularly resulted insubstantial lowering of driver eye height. The alternative ofrebuilding large lengths of road was not acceptable. Lie and Tingv-all (2009) have opened debate on the development of an ISO stan-

1178 I. Johnston / Safety Science 48 (2010) 1175–1181

dard for traffic safety as part of a process of driving the implemen-tation of safe system thinking through the ranks of system design-ers and operators.

Many discussions of the safe system model commence with adiscussion of an assumed underlying ethical platform in which hu-man life is sacrosanct. The Swedish parliament formally adopted‘‘Vision Zero” in 1997 which, in effect, made the prevention ofdeath and serious injury the over-arching policy objective in themanagement of the road transport system (Tingvall, 1998; Tingvalland Haworth, 1999). It has been argued that this formal publicadoption of policy was fundamental in paving the way for the rad-ical change in institutional mindsets necessary.

While this may have been so in Sweden the question remains asto whether it is a necessary first step in all nations. While the ulti-mate objective is a system in which no-one behaving reasonably iskilled or seriously injured in road traffic the socio-political realitiesprevailing in most countries are more likely to result in incremen-tal steps along a safer system continuum rather than in short-termradical reform. The keys to starting the safe system journey appearto be securing top level political commitment, the setting of ambi-tious safety targets and the adoption of a new design and opera-tional philosophy by the key institutions (Johnston, 2003).

The acceptance of a safer (as opposed to a safe) system mindsetmay appear to run the risk of leading us back into the mire oftrade-off decision making, explicitly balancing the various goals ofroad transport against each other. But, this is not necessarily so. Inmany nations, utilisation of the principles of the safe system in roadsafety strategy formulation and the application of a formal roadsafety management system will be a simpler socio-political taskand, providing the targets set are ambitious and, ideally, in termsof absolute reductions in serious casualties, will lead to substantialgains. Current Australian road safety planning reflects this prag-matic approach; the communications focus with the politiciansand with the professionals in the critical institutions revolves aroundthe need for error-tolerance and for kinetic energy reduction.Through this focus, strategic planning for future road safety has ta-ken a new shape (Western Australian Road Safety Council, 2008).

7. How far has safe system thinking, planning and actionpermeated institutions and society?

There are three issues to consider – the extent to which the safesystem model underpins a nation’s strategic planning for improv-ing road safety; the degree to which the affected institutions haveadopted the operational implications of the safe system; and thelevel of political and community understanding and acceptanceof the consequences of safe system adoption. It is not possible toexamine these issues comparatively across national borders so thissection is confined to observations about Australia with which theauthor is most familiar.

In Australia one needs to examine planning at the state level be-cause Australia is a federation. While there is a national road safetystrategy it is more a consensus document than a model. The cur-rent national strategy (2001–2010) contains no mention of the safesystem (Australian Transport Council 2002). The most recent na-tional Action Plan, however, does refer to the safe system, specifi-cally the need to ‘‘. . . include and apply safe system principles ininfrastructure documents, safety audit processes and professionaldevelopment programs. . .” (Australian Transport Council 2009).

The State of Western Australia has recently issued its strategyfor the period 2008–2020. It is explicitly anchored in safe systemprinciples and its target is a reduction in the absolute number ofdeaths and serious injuries. It is an excellent example of best prac-tice strategic planning. It recognises that the State is not homoge-neous but comprises a large metropolis and vast regional and

remote areas, each requiring specific plans. It identifies the majorproblems in each; examines potential countermeasures for eachmajor problem and evaluates the evidence of their effectiveness;models packages of evidence-based measures applied at differentresource levels and selects a preferred package that optimises po-tential effectiveness in the context of resource opportunity andinstitutional readiness (Western Australian Road Safety Council,2008). This approach is now being followed in the developmentof the new national strategy (Corben, 2009, personalcommunication).

However, understanding and acceptance of the safe systemprinciples seems not to be widespread among institutions suchas road and traffic authorities, transport planning agencies orenforcement agencies. Using speed moderation as an example –because it is so fundamental to future progress (Lie and Tingvall,2009) – road and traffic engineers seem generally not to acceptthe value of reducing speed limits, particularly urban speed limits.Little of the discussion around speed limit setting examines thespeed distribution as a whole. Urban speed limits remain high inAustralia, in comparison with most European nations. However,in terms of infrastructure design, there are promising signs of insti-tutional progress, for example a national roundtable was held inMarch 2009 to share knowledge and experience in seeking toimplement safe system innovations (Turner et al., 2009).

Police officers are another group of players who seem not to ac-cept the population-wide, preventive medicine approach toenforcement. Using speed moderation again as an example, thereis little acceptance of the case for reducing enforcement toleranceswhen enforcing speed limits. The majority of traffic infringementsissued through speed camera detection are for speeding less than10 km/h over the limit; almost none of those issued by police offi-cers are for less than 15 km/h over the limit. The author’s personalexperience in presenting in-house training to large numbers of po-lice officers in Victoria, Australia and in New Zealand demonstratesthat the vast majority of front-line officers and their immediatesupervisors do not understand the importance of shifting the speeddistribution to the left. They remain convinced that their principaltask is to apprehend those speeding at very high levels.

While in-house training can, and should, be used to increase theunderstanding of officers around the critical importance of shiftingthe speed distribution to the left, a more immediate process is avail-able. Assigning line officers the target of changing the speed distribu-tion on nominated high-risk sections of the network in their patchand holding them accountable for achieving this outcome will alterenforcement practice. It is rare at present for police to set targets forother than hours of speed enforcement or number of tickets issued,both of which sidestep the objective of shifting the speed distribu-tion rather than just impacting the tail of the distribution. Changingthe speed distribution operationalises the safe system kinetic energymanagement principle in a realistic way for police officers.

It is clear that the spread of safe system thinking in key institu-tions is still embryonic in Australia. How critical and how urgent isits widespread transfer in changing institutional behaviour? It isextremely important in the longer term and will likely requirethe targeting of future generations of practitioners as an integralpart of the curriculum for their qualifications training. In the shortterm, however, the key to changing institutional behaviour is thecreation of appropriate targets, sound strategies, effective perfor-mance monitoring and clear lines of accountability; in short theadoption of an effective road safety management system.

There have been significant changes in institutional behaviourwith regard to institutional co-operation and coordination, whichcaptures that element of the safe system model related to the inter-connectedness of all elements in the road transport system. The tra-ditional silo approach in which each institution did its own thingindependently is disappearing. The synergies across the efforts of

I. Johnston / Safety Science 48 (2010) 1175–1181 1179

agencies are not only reasonably well understood but in manystates are being overtly and explicitly managed. In Victoria, thechief executives of the road and traffic agency, the judicial agency,the police and the compensation insurer are jointly accountable toparliament for achieving the state’s road safety targets (FHWA,2006). Again, it is the setting of targets, the clear strategies andthe effectiveness of the lines of accountability that have proven cru-cial in driving institutional cooperation and coordination.

8. Seeking culture change

Road safety progress in the last decade appears to havestemmed from an increased emphasis on strategic planning – com-prising the data-driven selection of the major problems to address,the setting of objective but ambitious targets and a focus on effec-tive implementation of programs of measures through institutionalco-operation and coordination (OECD, 2008). Nevertheless, asWegman et al. (2006) argue, the culturally-anchored unsafe behav-iours of drink-driving and speeding continue to prove barriers tothe achievement of sustainable safety. It is worth noting here thatthe success stories of Australia (FHWA) and France (FlA Founda-tion, 2006) stemmed primarily from massive investments in roadpolicing – both equipment and enforcement intensity – to reducethe incidence of these unsafe behaviour by extreme levels of deter-rence. It is an open question as to whether the underlying culturearound these behaviours has changed fundamentally as a result.

It is not possible to generalise across national boundaries andarguably of dubious validity to do so across even state boundariesin a federation like Australia. Within Australia, however, it seemsclear that there is little understanding within the media or thepublic-at-large of the safe system principles. The media reportingof new state road safety strategies has focussed on the targetsand on specific measures, particularly those that involve changesto regulation and enforcement intensity.

How important is it that the public understand the thinking be-hind the strategies? It likely matters only when public support isneeded for strategy implementation. In the early 2000s in the Stateof Victoria, the lowering of general urban speed limits, the intensi-fication of speed limit enforcement, the reduction of enforcementtolerances, and the use of automated enforcement such as fixedand mobile speed cameras generated considerable public contro-versy. Accusations of revenue-raising dominated the newspapersand radio airwaves for many months. One political party in thelead up to an election promised to reduce the number of speedcameras and to restore enforcement tolerances, based on theirinterpretation of public sentiment. Interestingly, official surveysof public opinion reported a majority of citizens in favour of in-tense speed enforcement but this was not apparent in politicalreaction or media coverage (FHWA, 2006).

There is an underdeveloped constituency for safety within theAustralian public. For example, both government and corporatefleet managers have resisted specifying, in their purchase specifica-tions, vehicle safety features that are not required by regulation.Thus proven collision avoidance innovations such as electronic sta-bility control and proven injury reduction features such as curtainside airbags are not specified. Employees have not pressed theiremployers to provide safety above the regulated minimum, despiteduty of care provisions in occupational health and safety laws thatcould be brought into play.

The situation appears more acute in the United States. The AAAFoundation for Traffic Safety has lamented that ‘‘. . . one of ourgreatest challenges (isn’t) a specific safety issue, but rather a way ofthinking about all of the issues, or our ‘‘safety culture”. There are sev-eral compendia in the literature detailing known effective safetymeasures and providing the evidence for each (Elvik et al., 2009)

but the challenge is choosing measures that the community andhence the political and institutional players will accept.

It is important to establish acceptance by both politicians andthe public of the rationale for the road safety strategy to be imple-mented. Where folk lore and official strategy are at odds – as is thecase with speed moderation across the entire speed distribution –public understanding is vital (Johnston, 2004). There can be noclearer example of a mismatch between strategic thinking and cul-tural mindset.

There is a total lack of action by governments worldwide and bythe automobile industry to limit the power, top-speed and acceler-ation capability of cars and trucks. Injury reduction is a matter oflimiting the (potential) transfer of kinetic energy to the humanbody to a level which can be borne. Industry has certainly madegreat progress in reducing energy transfer through seat belts, airbags, collapsible steering columns, improved side door strengthand so on across a long list of crashworthiness measures but im-pact speeds are frequently above human tolerance levels. The prin-cipal determinant of the amount of kinetic energy to be managedin the event of a crash is impact speed and speed control has neverbeen on the vehicle design or regulatory agenda. There is not onevehicle safety design rule that pertains to top-speed capability oracceleration performance. Moreover, the speedometer throughwhich the driver manages his speed of travel contains a scale thatdisplays speeds of double or more that permitted on all but a hand-ful of the world’s roads. A manufacturer can get a five star safetyrating for a model that can exceed 250 km/h!

The automobile industry markets and sells on power and per-formance far more so than on any other feature; there is a vibrantglobal motor sports industry that links manufacturers (if tenuouslyin reality) to publicly available brands; there is a thriving after-market in speed-related accessories; and motoring journalists tendto promote speed and performance in their reviews of new models.The fact that governments have not intervened illustrates the glo-bal economic importance of the industry in all its forms as well asan apparent belief that the community (the market) does not be-lieve that speed limiting is important at the vehicle design level.

A wider issue is the relative roles of public versus politicalacceptance in the adoption of safe system principles and, moreimportantly, in their effective implementation. The chicken andegg argument of whether the politicians develop public policy thatreflects public opinion or whether public policy is used to shapepublic opinion is often heard when safety is discussed.

It is at least arguable that the politicians in most Australianstates have not accepted the safe system as such but have merelyaccepted the package of measures that comprise the particularstrategy put to them, after weeding out any specific measuresdeemed unacceptable politically. The new strategy adopted bythe government in Western Australia, for example, excluded therecommended reductions in speed limits because of their per-ceived unpopularity (Western Australian Road Safety Council,2008). None of this is surprising, or even disappointing; it is simplypart and parcel of the functioning of a democracy. What matters isthat the politicians accept a strategy that contains challenging tar-gets of reductions in the absolute numbers of serious casualties, apackage of evidence-based measures that are likely to achievethose targets and transparent mechanisms for public accountabil-ity for achieving those targets. This requires leadership, especiallyfor those critical measures in the package that may prove contro-versial. The safety strategists have a crucial role in providing poli-ticians with an understanding of the evidence underpinning thestrategy and its package of measures.

The safe system model requires shared responsibility. Roadusers must comply with the prescribed and proscribed behavioursand norms of road use while system designers and operators mustprovide a system that is error-tolerant and manages potential ki-

1180 I. Johnston / Safety Science 48 (2010) 1175–1181

netic energy for a compliant road user population. How well are wemanaging compliance with appropriate behavioural norms? Thecornerstones are requiring self-protection behaviours, such as seatbelt wearing; controlling entry to the road system (for other thanpedestrians and bicyclists) by licensing; promoting fitness to drivethrough the prescription of blood alcohol and similar limits; andprescribing a set of road rules governing precedence at conflictpoints and maximum permitted speeds under variouscircumstances.

It is with regard to behavioural norms that differences in culturalunderpinnings across nations assume paramount importance. Theauthor was requested to prepare, for the US Department of Trans-portation, an account of how Victoria achieved a halving of roadcrash fatalities in a 15 year period (FHWA, 2006). Victoria’s successstemmed largely from a preparedness to legislate, regulate and en-force, supported by extensive public education specifically tounderpin these measures. In France, the dramatic improvementsnoted in Table 2 followed the adoption at the end of 2002 of ‘‘a com-prehensive plan to automate the ‘‘control and sanction” chain for driv-ing offences and particularly for speeding offences” (FlA Foundation,2006). The USA has been unable, given its cultural norms, to sustainbasic self protection legislation requiring helmets or seat belts to beworn. Only 21 states require all vehicle occupants to wear seatbelts, one state has no belt wearing law at all. A large number ofstates do not require motorcyclists to wear helmets. Random breathtesting, a proven effective measure, is not acceptable. Thus, the Uni-ted States is severely handicapped in terms of its strategies for roaduser behaviour compliance under the shared responsibility princi-ple. The need for individual nations to develop tailored approachesto culture change to reflect their unique patterns is crucial.

What we have not done, at least not systematically, is addressthe cultural mores that underpin behaviour. Earlier, the conflictssurrounding the speeding issue were discussed as a prime exam-ple. A population-wide speed behaviour change (a general slowingdown) in the interest of the common good is hardly likely withoutsignificant cultural change. Individuals perceive a personal benefitfrom setting their own travel speed, do not regard low level speed-ing as dangerous and seem not to accept that if everyone moder-ated their speed by a small margin there would be significantsafety gains for the community.

The Dutch have started examining the theoretical principlesunderpinning what they call ‘social forgivingness” to begin tounderstand how to formulate a strategy for changing the prevailingcultural more of personal gain to one of community gain in dictat-ing road user behaviour (Houtenbos, 2009). The concept is a gen-eral one of how we might engender a culture around road safetythat would facilitate a preventive medicine approach to behaviourchange (Rose, 1992). Howard and Sweatman (2007) use the term‘‘social contract” to describe the bringing together of prevailingcommunity views and strategic thinking around road safety. Fu-ture road safety gains demand increased scientific attention tothe issues around changing the culture of road safety.

Urban geographers concerned with social cohesion and com-munity well-being (May et al., 2008) have issued a call for a shifttowards an ethic of care, a call which to date appears to have es-caped the attention of mainstream road safety professionals. Co-operative research between public health, urban geography, envi-ronmental sustainability and road safety professionals may wellbe a promising way forward in developing a culture change re-search agenda.

9. Conclusions

Having an appropriate strategy and action plan is one thing,achieving implementation is quite another. There are several

excellent road safety management systems in the literature (forexample Bliss and Breen, 2009) but they cannot be implementedeffectively without a political and community willingness to treaddown a safe (or even a safer) system path. The recent OECD reporton achieving ambitious road safety targets (OECD, 2008) stressesthe importance of political commitment and public support beforegoing on to list the detailed requirements for institutions and gov-ernments, requirements which are necessary but not sufficient.

The keys to current best practice can be summed up in the fourCs:

� Constituency – unless the public demands action, appropriateresource will not be applied. In Victoria, for example, each ofthe new programs introduced in 1970, in the period aroundthe mid 1970s, in the mid 1980s and in the early 2000s followedpublic ‘‘outrage” at apparent, dramatic increases in death totals(FHWA, 2006). However, ‘‘outrage” generates only an immediateresponse. A constituency based upon a cultural underpinning ofan ethic of care for community is likely to be required for sus-tained progress.

� Commitment – without political will from the top, success mustbe limited. The outstanding demonstration is in France whichwent from laggard to best practice within three years after thePresident personally intervened (FlA Foundation, 2006). Com-mitment and constituency are bedfellows, with political leaderstending to move with publicly held behavioural norms.

� Cooperation – road safety is institutionally very complex withthe actions of numerous agencies impacting upon progress.Without meaningful cooperation integrated packages of mea-sures cannot be successfully applied (FHWA, 2006; Wegmanet al., 2008).

� Coordination – the necessary bedfellow of cooperation, coordina-tion is vital to integration and synergy across institutionalefforts (FHWA, 2006; Wegman et al., 2008).

It is instructive that the latest work out of Europe on how tomeasure road safety progress recommends three indices – onebased upon policy performance (e.g. the quality of the strategicand operational planning and the political will); one based uponimplementation performance (e.g. process indicators like resourceapplied); and one based on more conventional safety outcomemeasures (e.g. deaths and injuries) with the three linked causallyin a chain from policy to outcome (Wegman et al., 2008).

Ideal best practice is when the principles of the safe systemhave been adopted within all of the institutions impacting the de-sign and operation of the road transport system and appliedthrough an effective management system with transparent linesof institutional accountability for the achievement of tough targetsfor reductions in death and serious injury. Institutional acceptanceis related to community understanding and acceptance.

We must systematically address several critical socio-culturalissues around car use and road safety and explore ways of movingtowards community acceptance of a duty of care. It may well bethat a duty of care in road use can best be developed through pub-lic engagement in sustainability. This will require political leader-ship, duty of care role-modelling by governments and largecompanies, and the creation of partnerships with others concernedwith the sustainability of urban society. A community constituencyneeds to be created and the beginning likely lies in vigorous andwell-informed public debate (Howard and Sweatman, 2007). Theethic of care (conservation) created in nations like Australia aroundwater and waste needs extending to road transport.

As Wiegmann and von Thaden (2007) conclude: ‘‘While theconcept of safety culture has been fundamentally applied to organ-isations or groups, it has yet to be systematically applied to thepopulation at large”. Given the discord between some of the key

I. Johnston / Safety Science 48 (2010) 1175–1181 1181

behavioural norms of road use and the principles of safe systemthinking addressing the research void is likely to make a significantprogress to ongoing safety gains.

References

Australian Transport Council, 2002. National Road Safety Strategy 2001–2010. ATC,Canberra.

Australian Transport Council, 2009. National Road Safety Action Plan for 2009 and2010. ATC, Canberra.

Bliss, T., Breen, J., 2009. Implementing the Recommendations of the World Reporton Road Traffic Injury Prevention. Country Guidelines for the Conduct of RoadSafety Capacity Reviews and the Related Specification of Lead Agency Reforms,Investment Strategies and Safety Projects. World Bank Global Road SafetyFacility, Washington, DC.

Elvik, R., 2003. An overview of target setting in Europe. In: Best in Europe 2003:Targeted Road Safety Programmes in the EU. European Transport SafetyCouncil, Brussels.

Elvik, R., Vaa, T., Erke, E., 2009. The Handbook of Road Safety Measures, second ed.Emerald Publishing.

FHWA, 2006. Halving Deaths From Road Traffic Crashes – A Case Study FromVictoria, Australia 1989–2004. US Department of Transportation, Washington,DC.

FIA Foundation, 2006. Road Safety in France – Reflections on Three Decades of RoadSafety Policy. FIA Foundation for the Automobile and Society, London.

Hauer, E., 1990. The engineering of safety and the safety of engineering. In: Rothe,J.P. (Ed.), Challenging The Old Order – New Directions in Traffic Safety TheoryTransactions. New Brunswick.

Houtenbos, M., 2009. Social Forgivingness: A Theoretical Exploration SWOV ReportR-2009-8 SWOV, Leidschendam (in Dutch with an English summary).

Howard, E., Sweatman, P., 2007. Road safety culture development for substantialroad trauma reduction. In: AAA Foundation for Traffic Safety (2007) ImprovingTraffic Safety Culture in the United States – The Journey Forward. AAAFoundation for Traffic Safety, Washington, DC

Johnston, I.R., 2003. Improving road safety in the longer term – finding the rightbuttons to push. In: Proceedings 2003 Road Safety Research, Policing andEducation Conference. Monash University Accident Research Centre,Melbourne.

Johnston, I.R., 2004. Reducing injury from speed-related road crashes – towards apopulation-based preventive strategy (guest editorial). Injury Prev. 10 (5), 257–259.

Johnston, I.R., 2006. Highway safety. In: Fwa, T.F. (Ed.), The Handbook of HighwayEngineering. Taylor and Francis, New York.

Kissinger, J.P., 2007. Preface. In: AAA Foundation for Traffic Safety (2007) ImprovingTraffic Safety Culture in the United States – The Journey Forward. AAAFoundation for Traffic Safety, Washington, DC.

Koornstra, M., Lynam, D., Nillson, G., Noordzij, P., Petterson, H.-E., Wegman, F.,Wouters, P., 2002. SUNflower; A Comparative Study of the Development of RoadSafety in Sweden, The United Kingdom and the Netherlands. SWOV,Leidschendam.

Lie, A., Tingvall, C., 2009. What Can an ISO-Standard for Traffic Safety Do. SwedishRoad Administration, Borlange.

May, M., Tranter, P.J., Warn, J.R., 2008. Towards a holistic framework for road safetyin Australia. J. Transport Geogr. 16, 395–405.

OECD, 2002. Road Safety: What’s the Vision? OECD, ParisOECD, 2008. Towards Zero: Ambitious Safety Targets and the Safe System Approach.

OECD, Paris.Reason, J., 1997. Managing the Risks of Organizational Accidents. Ashgate, UK.Rose, G., 1992. The Strategy of Preventive Medicine. OUP, Oxford.Sleet, D.A., Dinh-Zarr, T.B., Dellinger, A.M., 2007. In: AAA Foundation for Traffic

Safety Improving Traffic safety Culture in the United States – The JourneyForward. AAA Foundation for Traffic safety, Washington, DC.

Tingvall, C., Haworth, N., 1999. Vision Zero: an ethical approach to safety andmobility. In: 6th Institute of Transport Engineers International Conference onRoad Safety and Traffic Enforcement, Melbourne, September 1999.

Tingvall, C., 1998. The Swedish Vision Zero and how parliamentary approval wasobtained. In: Proceedings 1998 Road Safety Research. Policing and EducationConference Monash University Accident Research Centre, Melbourne.

Tingvall, C., 2004. The Swedish Vision Zero. In: WHO (2004) World Report on RoadTraffic Injury Prevention WHO, Geneva.

Trinca, G.W., Johnston, I.R., Campbell, B.J., Haight, F.A., Knight, P.R., Mackay, G.M.,McLean, A.J., Petrucelli, E., 1988. In: Reducing Traffic Injury – A GlobalChallenge. Royal Australasian College of Surgeons, Melbourne.

Turner, B, Tziotis, M, Cairney, P, Jurewicz, C., 2009. Safe System InfrastructureNational Roundtable Report. Report ARR 370, ARRB Group Ltd., Melbourne.

Watkins, K., 2009. The world is turning a blind eye. In: Ward, D. (Ed.), Make RoadsSafe: A Decade of Action for Road Safety. Commission for Global Road Safety,London.

Wegman, F., Aarts, L., (Eds.), 2006. Advancing Sustainable Safety – National RoadSafety Outlook for 2005–2020. SWOV, Leidschendam.

Wegman, F., Commandeur, J., Eksler, S., Gitelman, V., Hakkert, S., Lynam, D., Oppe, S.,2008. SUNflower Next – Towards a Composite Road Safety Performance Index.SWOV, Leidschendam.

Wegman, F., Eksler, S., Hayes, S., Lynam, D., Morsink, P., Oppe, S., 2006.SUNflower+6: A Comparative Study of the Development of Road Safety inEuropean countries. SWOV, Leidschendam.

Western Australian Road Safety Council, 2008. Towards Zero: getting theretogether. In: Council’s Recommendations to Government to Reduce RoadTrauma in Western Australia 2008–2020. Road Safety Council, Perth.

WHO, 2004. World Report on Road Traffic Injury Prevention. WHO, Geneva.WHO, 2009. Global Status Report on Road Safety – Time for Action. WHO, Geneva.Wiegmann, D.A., von Thaden, T.L., 2007. A review of safety culture theory and its

potential application to traffic safety. In: AAA Foundation for Traffic Safety(2007) Improving Traffic Safety Culture in the United States – The JourneyForward. AAA Foundation for Traffic Safety, Washington, DC.

Williams, A.F., Haworth, N. 2007. Overcoming Barriers to Creating a Well-Functioning Safety Culture: A Comparison of Australia and the United Statesin AAA Foundation for Traffic Safety (2007) Improving Traffic Safety Culture inthe United States – The Journey Forward. AAA Foundation for Traffic Safety.Washington, D.C.