Constructing Crashworthiness.

The Experimental Safety Vehicle (ESV) Program and the

Global Renegotiation of Automobile Safety in the 1970s

Dr. Stefan Esselborn

Technische Universität München (TUM) 1

Abstract:
The article proposes to take a fresh look at the global career of “crashworthiness” as a
dominant “paradigm” (Peter Norton) of automobile safety by focusing on the so-called
Experimental Safety Vehicle (ESV) program of the early 1970s. Initiated by the U.S. Department
of Transport (DOT) and internationalized through the newly created NATO Committee for the
Challenges of Modern Society (CCMS), the scheme ultimately came to involve the governments
of all major car-producing countries, as well as practically all relevant automobile corporations
in the capitalist “West”. The ESV program provided a significant boost to automobile safety
research and contributed to the professionalization, institutionalization and standardization
of the field. It also supplied a platform for a transnational re-negotiation of the distribution of
responsibility for automobile safety, in which differences between engineering cultures and
user perceptions in North American and Europe/Japan came to the fore. In this context, the
experimental prototypes functioned as “evidence objects”, which different actors could use to
generate and validate technical knowledge, but also to make economic and political
arguments. By serving as material anchorage points for a transnational techno-political
debate, the ESVs played an important part in shaping the way in which the challenge of
“crashworthiness” influenced automobile safety practices worldwide.
Zusammenfassung: Der Beitrag untersucht die globale Karriere der “Crashworthiness” bzw.
“passiven Sicherheit” als neuem Paradigma (Peter Norton) der automobilen Sicherheit aus der
Perspektive des sogenannten Experimental Safety Vehicle (ESV)-Programms der frühen 1970er
Jahre. Initiiert vom US-amerikanischen Department of Transport (DOT) und internationalisiert
über das neugeschaffene Committee for the Challenges of Modern Society (CCMS) der NATO
konnte das Projekt letztlich die Regierung aller Länder mit einer nennenswerten
Automobilindustrie, sowie praktisch alle wichtigen Automobilfirmen des kapitalistischen
„Westens“ zu den Teilnehmern zählen. Das ESV-Programm beschleunigte die Entwicklung
spezifischer technischer Lösungen und trug zur Professionalisierung, Institutionalisierung und
Standardisierung der Forschung zum Thema automobile Sicherheit bei. Es bot außerdem eine
Plattform für die globale Neuaushandlung der Verantwortung für Sicherheit im
Straßenverkehr, die Unterschiede zwischen den Ingenieurskulturen und Nutzererwartungen in
Nordamerika und Europa bzw. Japan freilegte. Die experimentellen Prototypen selbst
fungierten in diesem Zusammenhang als „Evidenzobjekte“, die von verschiedenen Akteure
sowohl zur Datengenerierung und zur Validierung technischen Wissens, als auch zur

1
This article was researched and written within the context of the Deutsche Forschungsgemeinschaft (DFG)
Research Group 2448 “Practicing Evidence – Evidencing Practice”. I am particularly indebted to Karin Zachmann,
who constantly provided feedback and suggestions throughout the process, and Sascha Dickel for his input and
help on the theory and (social) practice of prototypes.
1
Untermauerung ökonomischer und politischer Argumente genutzt wurden. Als materielle
Ankerpunkte einer globalen techno-politischen Debatte prägten die ESV die Art und Weise
entscheidend mit, in der die Idee der „Crashworthiness“ auf globale Praktiken der automobilen
Sicherheit durchschlug.
In June 2019, the 26th International Technical Conference and Exhibition on the Enhanced
Safety of Vehicles (ESV) reunited about 1,200 professionals from manufacturers and suppliers,
research institutions, government agencies, insurance companies and other interested parties
from around the world in Eindhoven, the Netherlands, to discuss the newest developments in
automobile safety. 2 Although the conference brought together some of the best-known
specialists in the field, arguably one of its biggest attractions was not a human, but a car: With
great fanfare, Mercedes-Benz presented its ESF 2019 – the acronym stands for
“Experimentelles Sicherheitsfahrzeug”, or Experimental Safety Vehicle. The fully functional
vehicle was meant to showcase some of the company’s recent developments in automobile
safety, particularly relating to what Mercedes referred to as the two “key technologies for the
mobility of the future” – electrification and autonomous driving.3 Unusually for a prototype,
however, the purpose of the ESF 2019 was not only to promote a specific vision of the future,
but also to draw attention to the past. As the accompanying promotional material
emphasized, the new vehicle was merely the latest addition to the “long-standing tradition”
of more than 30 ESF models, which the company had built in the context of various ESV
conferences since their beginning in the early 1970s. 4
From a historical perspective, Mercedes-Benz’s choice to advertise its dedication to
automobile safety in this way raises a number of questions. Why would a company notoriously
proud of its German engineering tradition, reaching back to the very invention of the
automobile, go to such lengths to remind its potential customers of a project started by the
U.S. federal bureaucracy in 1970? What influence did the original Experimental Safety Vehicle
(ESV) program have on the development of automobiles since the early 1970s, and how does
it fit in with the oft described “paradigm changes” in the field during this time? What can the
program’s history tell us about the way in which these conceptual changes and new regulatory
requirements translated into safety practices – not only in the United States, but worldwide?
And what role did the experimental prototypes themselves play in this?
As Mercedes implicitly seems to suggest with its use of the ESF 2019, there are some parallels
between the potentially profound disruptions to our familiar system of automobile mobility
widely discussed today and the situation in the early 1970s. In the shadow of the triple crisis
of the seemingly unstoppable rise of accident deaths, the increasing awareness of
environmental degradations caused by mass-motorization, and the economic disruptions of

2
Cf. the official conference program: https://www.esv2019.com/docs/Program_Book_ESV2019.pdf. While the
official title of the conference series was changed from “International Technical Conferences on Experimental
Safety Vehicles“ to “International Technical Conferences on the Enhanced Safety of Vehicles” in 1991, the
acronym remained unchanged.
3
Rodolfo Schöneburg, ESF 2019. Experimental Safety Vehicle Meets Automated Driving Mode, Eindhoven 2019.
4
ESF 2019: New safety ideas for a new mobility, available online at https://www.mercedes-
benz.com/en/mercedes-benz/innovation/esf-2019/
2
the first oil price shock, many observers predicted far-reaching changes to what automobiles
of the future would look like, and how they would be used. 5 Then as now, automation was
widely considered the ultimate technological solution to many of these problems. In marked
contrast to today, however, only the most daring technological visionaries of the 1970s would
have considered the total replacement of human drivers a viable option for the immediate
future.6 Instead, engineers and officials of the time concentrated their efforts on less
ambitious versions of partial automation, limiting themselves to the transfer of some selected
functions from human users to the machine. Some of the solutions developed in this context
are easily recognizable as direct predecessors of today’s assisted driving systems, such as anti-
lock braking or early radar distance-warning systems. Arguably the most consequential of
these transfers of responsibility, however, concerned what Peter Norton has described as the
replacement of the “paradigm of control” by the “paradigm of crashworthiness” as the
dominant approach to automobile safety. 7 Whereas until the late 1950s most traffic safety
measures had concentrated on the driver’s responsibility to avoid accidents, subsequently a
new approach started to gain ground. Its proponents claimed that it should be incumbent
upon the car (and hence indirectly its makers) to protect occupants under all circumstances.
Functionally, this also constituted a form of automation, as a vital function of the automobile,
the protection of it passengers, which had so far mainly been the responsibility of the human
operator, was (partially) entrusted to new technical systems – such as crumple zones or
passenger restraints. 8
In the historiography of automobility, the “invention” and establishment of crashworthiness
have often been presented as an American story – from the “invention” of impact
biomechanics in the 1940s via the activities of Ralph Nader and his associates in the mid-1960s
to the passing of the National Traffic and Motor Safety Act of 1966. 9 However, this only

5
On the interactions between car culture and the societal changes of the 1970s cf. recently Sina Fabian, Boom
in der Krise. Konsum, Tourismus, Autofahren in Westdeutschland und Großbritannien 1970-1990, Göttingen
2016; Ingo Köhler, Auto-Identitäten. Marketing, Konsum und Produktbilder des Automobils nach dem Boom,
Göttingen 2018; for a broader contextualization in (West German) processes of societal change Frank Bösch,
Boom zwischen Krise und Globalisierung. Konsum und kultureller Wandel in der Bundesrepublik der 1970er und
1980er Jahre, in: Geschichte und Gesellschaft 42, 2016 (2), p. 354–376.
6
While the idea was occasionally mentioned as a sort of utopian solution for the medium-to-long-term, the early
1970s seem to have marked a sort of lull in the production of visions for fully automated personal automobiles;
cf. Wetmore, this volume. Instead, automation enthusiasts during this time appear to have focused on collective
(or hybrid private-collective) transport systems, cf. Barbara Schmucki, Individualisierte kollektive
Verkehrssysteme und kollektive individuelle Verkehrssysteme. Die Vision von Neuen Technologien zur Lösung
der Verkehrsnot in den Städten in den 1970er Jahren, in: Hans-Liudger Dienel and Helmuth Trischler (eds.),
Geschichte der Zukunft des Verkehrs. Verkehrskonzepte von der frühen Neuzeit bis zum 21. Jahrhundert,
Frankfurt/Main 1997, p. 147–169.
7
Peter D. Norton, Four Paradigms. Traffic Safety in the Twentieth-Century United States, in: Technology and
Culture 56, 2015 (2), p. 319–334. The redistribution of responsibility (and not least legal liability) implied by this
paradigm change is most clearly described by Jameson M. Wetmore, Redefining Risks and Redistributing
Responsibilities. Building Networks to Increase Automobile Safety, in: Science, Technology, & Human Values 29,
2004 (3), p. 377–405.
8
Cf. the definition of automation quoted in the introduction to this issue.
9
For the “classic” version of this narrative see Joel W. Eastman, Styling vs. safety. The American automobile
industry and the development of automotive safety, 1900-1966, Lanham 1984; for an up-to-date account cf. Lee
Vinsel, Moving violations. Automobiles, experts, and regulations in the United States, Baltimore 2019, ch. 3–5.
3
provided the starting point for the difficult task of translating this new idea into regulatory
standards, establishing new evidence practices and infrastructures for testing, and ways and
means of enforcing them. 10 Even more importantly, crashworthiness was never exclusively an
American issue. While Peter Norton developed his paradigm model explicitly with a view to
“traffic safety in the twentieth-century United States”, he suggests that “American patterns
were echoed across the Atlantic and perhaps elsewhere” – albeit with potentially significant
regional or national divergences. 11 This not only leaves the question of the nature of the
connection between the U.S. example and its overseas “echoes” unanswered, but also
neglects the fact that the American developments themselves took place in a larger,
transnational context. 12 The automobile industry of the late 1960s and early 1970s was
already such a highly globalized endeavor that, even in the U.S. as the “motherland” of mass
motorization, a radical change of safety strategies necessarily implied some amount of trans-
and international coordination.
As this article will argue, a closer look at the ESV program can help to fill in both of these
historiographical gaps. Originally set in motion by the National Highway Traffic Safety
Administration (NHTSA) of the U.S. Department of Transportation (DOT) to prepare new
national safety standards, the program provided a highly visible public statement of the
agency’s idea of the concept of crashworthiness. Subsequently extended and
internationalized through the North Atlantic Treaty Organization (NATO), the scheme
ultimately came to involve not only the governments of the most important industrial powers
of the alliance, but also almost all major automobile corporations in the Western world at the
time. By enabling – and indeed forcing – these very different actors to directly engage with
the NHTSA’s ideas, the ESV program was essential for systematizing and institutionalizing the
inter- and transnational conversation about automobile safety, creating new forums and
institutional structures – some of which, such as the above-mentioned ESV conferences,
persist to this day. This did not mean that all those involved necessarily came to share the
NHTSA’s idea of crashworthiness. On the contrary, as will be shown below, the ESV program
often highlighted the differences between “American” and “European” (and Japanese)
approaches to automobile safety. In some cases, the fierce opposition of some of the
international participants also served to modify the U.S. regulators’ position on the issue.

Part of the reason why the ESV program so far seems to have largely escaped the attention of
automotive historians – with the notable exception of the involvement of two German

10
Cf. on this question Jameson M. Wetmore, Delegating to the Automobile. Experimenting with Automotive
Restraints in the 1970s, in: Technology and Culture 56 2015 (2), p. 440–463; Lee Vinsel, Designing to the Test.
Performance Standards and Technological Change in the U.S. Automobile after 1966, in: Technology and culture
56, 2015 (4), p. 868–894.
11
Norton (footnote 7), p. 320.
12
Amongst the existing transnational histories of automobility, two in particular stand out: Gijs Mom, Atlantic
automobilism. Emergence and persistence of the car, 1895 - 1940, New York, Oxford 2015, concentrates on the
pre-World-War-II history of automobiles; Ann Johnson, Hitting the brakes. Engineering design and the production
of knowledge, Durham 2009, traces transnational “expert communities” forming around Anti-Lock-Braking (ABS)
technology.
4
automakers, VW and Mercedes-Benz 13 – might have to do with its performative aspects,
which could make it seem more like a fanciful PR stunt than “serious” engineering. This is
particularly true for the experimental safety vehicles themselves, which contemporary critics
routinely dismissed as over-expensive curiosities that attracted a lot of attention, but were of
minimal practical relevance to the improvement of automobile safety. In this article, I will
argue that the ESVs, on the contrary, had an essential part in the renegotiation of safety
standards – both as interventions in the public discourse, and as specific technical artefacts. I
am particularly interested in the functioning of the prototype cars as what I will call “evidence
objects”, comprising both an epistemic and a discursive dimension. On the one hand, ESVs
were used to generate and validate specific technical knowledge, for example on the technical
feasibility and consequences of certain safety specifications. On the other hand, many, if not
all of the prototypes were built with the explicit purpose of making the case for particular
(socio-technical) conceptions of automobile safety, which were influenced at least as much by
economic and political as by technical considerations.

As recent works in Science and Technology Studies (STS) have pointed out, this dual nature –
as presently existing technical artefacts and representations of an envisioned (socio-technical)
future – is typical for prototypes. While their claim to attention rests on their purported status
as anticipations of things to come, their present materiality accounts for much of their
argumentative force. 14 In contrast to narrative scenarios, which verbally describe specific
visions of the future, the fact that prototypes like the ESVs could be physically experienced
and materially tested made them much more difficult to argue against – unless one could
point to one’s own material prototype. 15 In addition, being able to point to an existing object
also fit very well with the logic of the mass media, making the ESVs a very effective strategic
choice to create publicity for visions of a different, safer automobile future. 16

This article is based on original research conducted in the German Federal Archives in Koblenz
(BAK), published sources created in the context of the ESV program itself, as well as the
analysis of a variety of periodicals such as professional journals, motor magazines targeted at
a wider public, and general news media in West Germany and the United States. Inevitably,
this leads to a tendency to reflect the West German and U.S. points of view more thoroughly
than other possible angles. However, this problem can at least be somewhat mitigated by the
fact that much of the material used is explicitly concerned with matters of international

13
Heike Weishaupt, Die Entwicklung der passiven Sicherheit im Automobilbau von den Anfängen bis 1980. Unter
besonderer Berücksichtigung der Daimler-Benz AG, Bielefeld 1999, esp. p. 314–356, presents a relatively
extensive account of the project from the point of view of Mercedes-Benz, using the company archives. Norbert
Stieniczka, Das "narrensichere" Auto. Die Entwicklung passiver Sicherheitstechnik in der Bundesrepublik
Deutschland, Darmstadt 2006, esp. p. 263-318,
14
Cf. Ingo Schulz-Schaeffer and Martin Meister, Laboratory settings as built anticipations. Prototype scenarios as
negotiation arenas between the present and imagined futures, in: Journal of Responsible Innovation 4, 2017 (2),
p. 197–216; Sascha Dickel, Prototyping Society. Zur vorauseilenden Technologisierung der Zukunft, Bielefeld
2019.
15
As Dickel (ibid., p. 51), points out, prototypes thus function analogous to the laboratory in scientific
knowledge production, which likewise promises to back up verbal arguments with material demonstrations.
16
Dickel (ibid., p. 56), calls this function of prototypes “inszenatorische Objekte” (mise-en-scène objects).
5
cooperation, and/or includes material produced by program participants from other nations.
As far as possible, I have tried to fill remaining gaps by making use of secondary literature. I
will proceed in three steps: Firstly, I will outline the origins of the ESV program in the public
discussions about the post-war accident crisis and the idea crashworthiness during the 1950s
and 1960s, with a special regard to the role of early safety prototypes in the debate. Secondly,
I will give a brief account of the original aims and structure of the U.S. ESV project, and its
subsequent development into an internationally coordinated, industry-wide research and
development effort. Thirdly, I will have look more closely at some of the different prototypes
built and the arguments developing around them in the framework of the program, before
summing up my arguments in the conclusion.

Object(ive) Evidence for Safety: Experimental Vehicles and the Crashworthiness Debate of the
1950s and 1960s
For the industrialized countries of the Western world, the first three decades following the
Second World War – Eric Hobsbawm’s “Golden Years” 17 – were also a golden age of the
automobile. Whereas in the interwar years, private motor cars had remained a privilege of
the middle-to-upper classes, especially outside of North America, the combination of rising
wages and falling production costs during the post-war “economic miracle” brought
ownership within the reach of “ordinary” citizens. 18 With its promises of mobility, personal
freedom, and social and economic advancement, in many countries, the car came to
symbolize the benefits of the new postwar order. 19 However, the democratization of the
automobile did not come without costs. Arguably the most visible and most hotly debated
side effect of mass-motorization was the seemingly unstoppable rise of the number of deaths
and serious injuries caused by automobiles. In the United States, the global pioneer of mass-
motorization, yearly traffic fatalities had already reached around 30,000 in 1946. By 1950, they
exceeded 35,000, reached 40,000 in 1963, and finally surpassed 55,000 in 1969.20 While
absolute numbers were lower in other countries, their roads were hardly less dangerous.
Calculated per registered vehicle or distance travelled, the fatality rates in almost all other
highly industrialized nations – be it Britain, Australia, Italy, West Germany, the Netherlands,

17
Eric J. Hobsbawm, Age of Extremes. The Short Twentieth Century, 1914–1991, London 1995, p. 257–286.
18
Following the periodization by Mom (footnote 10), the era of “mass-motorization” in Europe and North
America can be said to span the years between 1945 and 1975. On the general development of cars and car
culture in the 20th century cf. Kurt Möser, Geschichte des Autos, Frankfurt, New York 2002.
19
For the West German case cf. Dietmar Klenke, Die deutsche Katastrophe und das Automobil. Zur
'Heils'geschichte eines nationalen Kultobjekts in den Jahren des Wiederaufstiegs, in: Michael Salewski and Ilona
Stölken-Fitschen (eds.), Moderne Zeiten. Technik und Zeitgeist im 19. und 20. Jahrhundert, Stuttgart 1994, p.
157–174; for “car culture” in the U.S. e.g. Mark S. Foster, Nation on Wheels. The Automobile Culture in America
since 1945, Belmont, CA 2003; Tom McCarthy, Auto mania. Cars, consumers, and the environment, New Haven
2007. For an exploration of the (positive as well as negative) emotive power of automobiles in West Germany
and the U.S. cf. most recently Thomas Zeller, Loving the Automobile to Death? Injuries, Mortality, Fear, and
Automobility in West Germany and the United States, 1950-1980, in: TG Technikgeschichte 86, 2019 (3), p. 201–
226.
20
U.S. Department of Transportation, Highway Statistics Summary to 1995, table FI-200,
https://www.fhwa.dot.gov/ohim/summary95/index.html
6
France, Finland or Japan – were even more appalling. 21 In a one-year time span in 1965/1966,
over 111,000 people lost their lives in motor vehicle accidents in the then-15 NATO member
states, causing the World Health Organization to refer to traffic deaths as “an epidemic
spreading around the world from which no country is exempt”. 22

In the immediate post-war years, governmental responses to this problem in most countries
were still shaped by the paradigm of “control” or “crash avoidance”.23 This approach, which
had first become prevalent in the United States during the interwar period, thanks to the
efforts of a coalition of automobile clubs, car dealers and the auto industry, held at its core
that ensuring road safety meant preventing accidents.24 This was possible, proponents
argued, by keeping three factors under control: the driver, the road, and the car. As drivers
(and other human participants in traffic, such as pedestrians) were thought to be responsible
for the overwhelming majority of accidents, educating and disciplining them was considered
the most important task. 25 To improve the safety of roads, post-war governments invested
vast sums of public money into construction programs – even though utopian visions of
eliminating accidents through “fool-proof” or even completely automated highway systems
ultimately proved elusive. 26 The automobile itself initially tended to be the least scrutinized of
the three factors. Under the paradigm of control, even fast and powerful cars were not
considered dangerous, as long as they functioned reliably and provided some help to their
drivers to avoid collisions – by sporting features such as good handling, reliable brakes, or
adequate lighting, which were later described as “active safety”. Since most new automobiles
already fulfilled these requirements to a reasonable degree, the auto industry could claim that
it bore little responsibility for the horrifying death toll on the roads. As late as 1953, the journal
of the West German Automobile Club (ADAC) – while deploring the rising number of traffic
deaths – congratulated international automobile companies for successfully improving the
safety of their products “year after year”. 27

Over the course of the 1950s, however, the escalating accident crisis started to erode the
credibility of the crash avoidance approach. A new generation of safety advocates began to
argue that, under the conditions of mass-motorized traffic, the complete prevention of

21
According to NATO statistics, the yearly traffic death rate per 100 million vehicle miles travelled for 1966/7
ranged from 5.6 for the USA, 7.0 for Britain, 12.1 for Italy, 13.3 for West Germany and France, to 16.1 for the
Netherlands. In terms of yearly deaths per 100,000 registered vehicles, West Germany had by far the worst
record with 126.0, compared to 92.5 in Finland, 85.4 in Japan, 78.7 in Australia, 67.7 in France and 54.4 in the
USA; cf. NATO Letter 18 (1970), p. 19f.
22
Ibid.
23
Norton (footnote 7), p. 326f. Wetmore’s “crash avoidance approach” describes essentially the same idea, using
a slightly different focus; cf. Wetmore, Redefining Risks (footnote 7), p. 380-382.
24
Cf. Peter D. Norton, Fighting traffic. The dawn of the motor age in the American city, Cambridge, Mass. 2008,
Ch. 8; for the postwar period Steve Bernardin, “Taking the Problem to the People”. Traffic Safety from Public
Relations to Political Theory, 1937–1954, in: Technology and Culture 56, 2015, p. 420–439.
25
For the (West) German case, Stieniczka (footnote 11) has therefore described the official approach to road
safety between the 1920s and the 1960s as “Strategie des disziplinierten Verkehrsteilnehmers” (“strategy of the
disciplined participant in traffic”).
26
Cf. Wetmore, this volume.
27
Kuhn, Herbert: Ist das Kraftfahrzeug sicher? ADACM 6 (1953), p. 314f.
7
accidents was clearly impossible. However, drawing on the first results in the fledgling field of
crash accident research, obtained by a handful of dedicated pioneers in the U.S. since the
1930s, they contended that much more could be done to mitigate the consequences of
accidents.28 Automobiles, in particular, had to be redesigned in a way to keep their occupants
safe not only during “normal” operation, but also – and especially – in the case of an accident.
This would require features such as padded interiors, passenger restraints, or frame structures
suitable for absorbing kinetic energy. This new approach, which soon became subsumed
under “crashworthiness” or “passive safety” 29, radically shifted responsibility for safety from
drivers and government officials – in charge of law enforcement as well as infrastructure –
towards the car itself, and hence the automobile industry. As a result, car companies, which
had so far largely remained on the sidelines of the traffic safety debate, started to see
themselves publicly accused of being “killers”, selling products that turned into “death traps”
in the case of accidents, thanks to sharp-edged “meat cleaver” dashboards, rigid “chicken
skewer” steering columns, or doors that sprung open upon impact.30

Under increasing public pressure, some car manufacturers in the U.S. and in Europe started
to embrace the idea of passive safety and develop new engineering solutions. While in
America, Ford tentatively equipped its 1956 models with a “safety package”, in Europe, Volvo
introduced the three-point seat belt and Mercedes-Benz the “crumple zone” into their
production vehicles in 1959. 31 However, these examples generated public attention precisely
because they were seen as exceptional. In general, the industry resisted any drastic changes
to their designs and practices, claiming that this would be, firstly, technically too difficult,
secondly, economically disastrous, and thirdly, unpopular with customers, who presumably
did not care much for safety. “Safety doesn’t sell” became a kind of catchphrase in the
industry. 32 When pressed on specifics – as for example representatives of the U.S. “Big Three”,
General Motors, Ford, and Chrysler, were in a Congressional Hearing in 1956 – auto executives
resorted to the argument that there was still not enough evidence for the effectiveness of the
proposed measures. 33

28
Cf. Amy Beth Gangloff, Safety in Accidents. Hugh DeHaven and the Development of Crash Injury Studies, in:
Technology and culture 54, 2013 (1), p. 40–61; Eastman (footnote 9), p. 177-208; Vinsel, Moving Violation
(footnote 9), p. 88-116.
29
„Crashworthiness“ was the common term in the U.S.; “active” and “passive” safety seems to have been coined
by an Italian engineer in 1964 and was introduced by Mercedes-Benz to German-speaking engineers, while their
French colleagues preferred to speak of “primary” and “secondary” safety; Stienicka, p. 46f.
30
„Are Car Manufacturers Killers?” Magazine Digest, 1953, quoted by Eastman, p. 220; Ist das Auto eine
Todesfalle? Ein SPIEGEL-Gespräch mit dem Chefkonstrukteur der Opel-Werke, Dr.-Ing. E.h. Karl Stief, in:
DER SPIEGEL, 4.12.1957, p. 44–53.
31
Cf. Eastman (footnote 24), p. 224–232; Heike Bergmann, Angeschnallt und los. Die Gurtdebatte der 1970er
und 1980er Jahre, in: Technikgeschichte 76, 2009; Weishaupt (footnote 11), p. 174–180.
32
This oft-repeated conviction seems to have been based largely on the experience of the 1956 Ford campaign.
However, its results were far from uniformly negative, as Eastman (footnote 24), p. 223–232, shows. For the
West German case, Stieniczka (footnote 11), p. 186f, argues persuasively that, as early as the late 1950s, (passive)
safety had become a quite persuasive sales pitch, in spite of the industry’s claim to the contrary.
33
Eastman (footnote 24), p. 242.
8
To confront what they saw as deliberate foot-dragging on the part of the industry, and to
demonstrate to the public that building a “crash-proof” car was actually feasible and
practicable, safety advocates repurposed a well-known practice in automobile engineering:
constructing experimental prototypes. 34 The first widely noted experimental vehicle with an
explicit focus on passive safety was the creation of the Cornell Aeronautical Laboratory, a
university-affiliated aviation research center in upstate New York. The Laboratory had been
contracted in 1951 by the Liberty Mutual Insurance Company to undertake research into
passenger injuries in automobile crashes. Drawing on the results of this work, Cornell
researchers developed an experimental “safety car”, which was presented to the public in
1957.35 Extensively remodeling a current Ford production model to incorporate more than 60
safety systems – including some “exotic” solutions such as backwards-facing rear seats,
accordion-style doors and control levers instead of a steering wheel – they claimed to have
created a vehicle that would allow passengers to survive collisions of up to 50 mph without
serious injuries. 36

The Cornell-Liberty car started what can almost be called a fad for safety prototypes. Private
enthusiasts picked up on the idea and presented their own, even more idiosyncratic
contributions, such as the “Aurora” (1957) or the “Sir Vival” (1958). 37 In 1961, a second
Cornell-Liberty Safety Car followed, this time with fewer spectacular novelties and increased
attention towards practicability and affordability. 38 The reaction of the auto industry was
much less enthusiastic. Asked about his opinion on the second attempt by Cornell-Liberty, a
spokesman declared that the vehicle did a “fine job of dramatizing the need for auto safety”,
but would stand little chance of being accepted by consumers.39 Nevertheless, the prototypes
created a considerable amount of public attention: They made the national media in the
United States, and were also discussed by automobile publications in Europe. Soon, the
European car industry reacted with its own prototype. At the 1963 Turin car salon, Italian car

34
Prototypes demonstrating different kinds of engineering innovations can be found throughout automotive
history, starting with the very first automobiles. In addition, a small number of car models especially emphasizing
injury reduction (through padded dashboards etc.) were built in the U.S. in the interwar years, such as the 1926
“Safety Stutz”, the 1937 Dodge (with a design influenced by crashworthiness research pioneer Claire Straith), or
the 1939 Studebaker; cf. ibid., p. 178–183. However, all of these seem to have been production models (albeit
of the more experimental type) rather than “true” prototypes.
35
According to Vinsel, Moving Violations (footnote 9), p. 88, even the writer John Updike contributed a piece
on the project for the New Yorker magazine.
36
Ibid.; Eastman (footnote 9), p. 192–193; Weishaupt (footnote 11), p. 106–107. The vehicle itself has been
preserved and is now on display at The Henry Ford Museum in Dearborn, Michigan; cf. Matt Anderson, 1957
Cornell-Liberty Safety Car, https://www.thehenryford.org/explore/blog/1957-cornell-liberty-safety-car/.
37
The “Aurora” was created by Father Alfred Juliano, a Catholic priest and automotive design enthusiast from
Connecticut; the “Sir Vival” owed its existence to Walter C. Jerome, a high school teacher with an engineering
degree from Worcester, Massachusetts. Both cars looked highly unusual, and repeated attempts by their
creators to have them manufactured in higher numbers were unsuccessful; cf. Jerry Garrett, How Ugly? Put a
Bag on That Car, in: New York Times, 23.12.2007; Last Word in Safety Cars?, in: Mechanix Illustrated, 1959 (April
1959), p. 59–61.
38
Eastman (footnote 9), p. 193; cf. also Behrendt, Ernst: Survival II. Das Auto, in dem einem nichts passieren
kann, in: ADACM,
14 (1961), p. 660–666.
39
Quoted by Eastman (footnote 24), p. 193.
9
designers Pininfarina showed a demonstration vehicle called the “PF Sigma”, which also
incorporated a number of passive safety systems patented by Mercedes-Benz. Although the
car was not operable, its enthusiastic reception in the specialized press demonstrated the
rising popularity of crashworthiness ideas in Europe as well. 40 However, as long as the
discussion remained confined to media campaigns and did not trigger any legislative or
administrative action, this had only limited impact on industry practices. 41

The situation was very different in the United States, where safety advocates and crash injury
researchers started to form strategic alliances with members of the administrative and
legislative system in the early 1960s. One result of this collaboration took the shape of a new
safety prototype – commissioned this time not by a private entity, but by the State of New
York. 42 The “New York Safety Sedan”, developed between 1965 and 1967 by another aviation-
related contractor, the Republican Aviation division of the Fairchild Hiller Corporation (FHC),
constituted a direct precursor to the future ESV program. The large, “tanklike” car (as TIME
magazine describe it) not only shared a number of specific design goals with the later
project,43 but also already pioneered some of their most remarkable technical features, such
as energy-absorbing hydraulic bumpers and a periscope for better rear vision. 44

While the New York Safety Sedan was still in development, a series of events suddenly pushed
crashworthiness to the top of the national political agenda.45 In the summer of 1965, Senator
Abraham Ribicoff, chairman of a little-know Senate committee, began a series of public
hearings on automobile safety, which unexpectedly turned into a national public relations
disaster for the automobile industry. Shortly afterwards, one of Ribicoff’s aides, the young
lawyer and consumer activist Ralph Nader published “Unsafe at Any Speed”, a book-length
indictment of the car industry, which quickly became an international bestseller.46 On the
back on this sudden and decisive swing in public opinion – further aided by the revelation of
GM’s clandestine surveillance of Nader –, Congress passed the National Traffic and Motor
Vehicle Safety Act (NTMVSA), signed into law by President Lyndon B. Johnson in September
1966.47 It created the Department of Transportation (DOT), a federal highway safety program,
as well as the National Highway Safety Bureau (NHSB), which became the National Highway
Traffic Safety Administration (NHTSA) in 1970. The latter was given the task to write and
enforce Federal Motor Vehicle Safety Standards (FMVSS), with which all cars sold in the United

40
Weishaupt (footnote 11), p. 108f; Stieniczka (footnote 11), p. 193f.
41
As was for example the case in West Germany, were the responsible authorities proved very reluctant to
issue any regulations pertaining to crashworthiness; cf. Stieniczka (footnote 11), p. 198f.
42
On the central role of the “New York safety scene” in the early crashworthiness movement cf. Vinsel, Moving
Violations (footnote 9), p. 118–127.
43
Such as the survivability of 50 mph frontal and rear impacts, 40 mph side impacts and 70 mph rollovers.
44
Cf. New York State Safety Car Makes Its Debut, in: SAE Journal 75, 1967 (3), p. 36–47; Proposals & Prototypes,
in: TIME Magazine 90, 1967 (23), p. 108.
45
Cf. P. W. Gikas, Crashworthiness as a Cultural Ideal, in: David Lanier Lewis and Laurence Goldstein (eds.), The
Automobile and American culture, Ann Arbor 1983, p. 327–339; Eastman (footnote 9), p. 241–250.
46
Ralph Nader, Unsafe at any speed. The designed-in dangers of the American automobile, New York 1965. On
Nader’s role cf. especially Vinsel, Moving Violations (footnote 9), p. 117–143.
47
More precisely, the legislation in question consisted of three separate acts, U.S. Public Law 89–563, 89–564,
and 89–670.
10
States would have to comply. This included a mandate to “conduct research, testing
development and training necessary”, notably by “procuring (by negotiation or otherwise)
experimental and other motor vehicles or motor vehicle equipment for research and testing
purposes”. 48 In March 1968, the DOT awarded three contracts for preliminary studies on a
federal experimental safety vehicle program. By early 1970, the Department sent out a
request for proposals for the “design, development, fabrication and delivery of a prototype
Experimental Safety Vehicle”. 49

Setting (Inter-) National Standards: The ESV Program and Its Internationalization
Apart from its federal purview, the DOT’s ESV program initially represented a more or less
direct continuation of earlier attempts, especially the Cornell and New York Safety Sedan
projects. Its aim was not to deliver a realistic precursor to future production models, but rather
to provide the agency with leverage in its extremely ambitious drive to radically reduce – or
even completely eliminate – traffic deaths in the U.S. within a decade. 50 To this end, the
project had four main objectives: Firstly, it would prove the technical feasibility of a much
higher degree of passive safety and crash protection – preferably without making too many
concessions in terms of driving performance, practicability or aesthetics. Secondly, it was to
influence public opinion and “stimulate public awareness” of the facts that cars could be
substantially safer. This would, thirdly, increase pressure on the automobile industry to invest
more in safety research and development, and to speed up the introduction of passive safety
features into their production models. Lastly, by testing and examining the resulting
experimental vehicles, the project was to generate specific technical data that the agency
could use to set new, legally binding safety standards. 51

To achieve these aims, the NHTSA intended to follow the lead of the recently completed
Apollo program and contract private firms from the aerospace and defense sector, working
according to government-defined specifications. Free from the constraints of the highly
competitive automobile market, they would be able to employ a “total systems engineering”
approach, “designing a car from the ground up with safety as a primary goal”, instead of a
mere “add-on”. 52 This would provide the much-needed “quantum jump” in automobile safety,
which car manufacturers had been unwilling or unable to deliver. “[W]e are goading the
industry”, Douglas Toms, the NHTSA’s energetic new director, told the journal of the American

48
U.S. Public Law 89-563, Sec. 106 (a).
49
NHSB, 1969 Report on Activities Under the National Traffic and Motor Vehicles Safety Act, Washington 1970,
p. 142; RFP DOT-OS-00050, “Subject: Experimental Safety Vehicle (Family Sedan)”, Feb. 17, 1970, copy in BAK B
108/23073.
50
Cf. People in the News. Douglas Toms Details Safety Bureau's Goals, in: Automotive Engineering 78, 1970 (8),
p. 83; Brenner, Robert: Opening Remarks, in: NHTSA (ed.), Report on the First International Technical Conference
on Experimental Safety Vehicles. Paris, France. January 25-27, 1971, Washington, D.C. 1971, p. 6–11.
51
NHSB (footnote 42), p. 141f. These four objectives remained constant talking points throughout the project,
which were repeated by NHTSA representatives at almost every presentation and can still be found in the final
project summary; cf. Committee on the Challenges of Modern Society (CCMS), Experimental Safety Vehicles
Project. CCMS Report No. 23, 1974, p. 3.
52
NHTSA (ed.), Report on the Third International Technical Conference on Experimental Vehicle Safety.
Washington, D.C., May 30–June 2, 1972, Washington, D.C. 1972, p. 1/5.
11
Society of Automotive Engineers in a 1970 interview with characteristic frankness: “If they
wanted to, they could make a safer vehicle. We are aiming to show the industry not only how
it can be done but that with our modest resources we can have it built.” 53

In June 1970, the NHTSA awarded contracts totaling about $8 million to two project teams. 54
Both consisted largely of established members of the U.S. military-industrial complex and
participants of earlier safety vehicle projects. Fairchild Hiller Corporation (FHC), a constructor
of military aircraft and the leader of the first team, had worked on the New York Safety Sedan.
The second team, led by American Machine and Foundry (AMF), a diversified industrial
conglomerate that produced everything from bowling pins to nuclear reactors, comprised
amongst others the Cornell Aeronautical Laboratory. 55 Against the NHTSA’s original
intentions, they were joined by two of the automobile giants: General Motors (GM) signed a
third, additional contract for a nominal $1 fee, and Ford followed with one year’s delay in June
1971 under the same conditions.56 All four contractors were required to deliver fully
functional prototypes of a “typical family sedan” in the 4000 lb/ 1800 kg class for inspection
and testing, fulfilling a long list of stringent safety requirements with a particular focus on
crashworthiness. 57

According to the NHTSA, the “full size” five-seater family sedan had been chosen because of
its “overwhelming popularity” in the United States.58 However, the large majority of
automobiles produced outside of North America belonged to the compact and subcompact
classes, which were gaining in popularity even within the U.S., as rising import numbers
attested. NHTSA officials were well aware that fulfilling stringent passive safety criteria would
be substantially harder for smaller cars, since the margins of maneuver in terms of physical
space, weight and price were much slimmer. To find new solutions for this model range, if
possible without undue expense to the American taxpayer, it seemed highly desirable to
include small cars and foreign automakers into the project in some form.59 While the U.S. ESV

53
People in the News (footnote 43), p. 83; The Crash Program That is Changing Detroit. Special Report, in:
Business Week, 27.2.1971, p. 78–83.
54
In addition, NHTSA spent 1.6 million between 1970 and 1974 on testing and similar activities in the framework
of the program. By comparison, research expenses related to alcohol and drug use in traffic – another one of the
NHTSA’s three “priority programs” during the period – added up to about $ million during the same period; cf.
NHTSA, Traffic Safety '74. A Report of Activities Under the National Traffic and Motor Vehicle Information and Cost
Savings Act of 1972, Washington, D.C., Annex E.
55
Both companies had already been contracted for the ESV preliminary study. The third preliminary contractor,
the Digitek Corporation, participated as a subcontractor to FHC. Both groups also included partners from the
automobile industry to help with design and styling, Chrysler for FHC and Minicars Inc. for AMF; cf. Albert
Schlechter, The United States 4000 lb. Experimental Safety Vehicle. Performance Specification, in: Report on the
First International Technical Conference (footnote 43), p. 26.
56
DOT Press Release, 21.6.1971, BAK B 108/37504. The third major automotive corporation in the U.S., Chrysler,
later joined the FHC group as a subcontractor.
57
Cf. Experimental Safety Vehicles Emphasize Crashworthiness, in: Automotive Engineering 79, 1970 (1), p. 54–
55; “Enclosure 5: Statement of Work”, BAK B 108/23073.
58
Schlechter (footnote 47), p. 25.
59
“Presentation for the First Technical Meeting of the U.S. Pilot Study on Road Safety: Experimental Safety
Vehicles (ESV) Program” [March 1970], BAK B 108/23073.
12
project was still under preparation, an opportunity to do both of these things arose in a
somewhat unexpected institutional context.

In April 1969, President Richard Nixon surprised allies and opponents alike by publicly
declaring his intention to endow NATO with a “third dimension”. In addition to the alliance’s
military and political functions, a newly created Committee on the Challenges of Modern
Society (CCMS) would work to safeguard the “quality of life” of citizens in the member
states.60 Although the main idea behind CCMS was to strengthen NATO by linking it to the
newly emerging field of environmentalism, the Committee interpreted its brief to encompass
all problems resulting from the “unheeded effects of technological change”. 61 “Road safety”
became one of the eight original “pilot projects” of the CCMS accredited by the NATO council
in February 1970 – largely due to the efforts of Daniel Moynihan, Nixon’s special advisor on
interior policy and one of the “intellectual fathers” of the CCMS, who had been a key-member
of the crashworthiness movement since the 1950s. 62 While the CCMS Road Safety project
officially consisted of seven separate sub-projects, each “piloted” by a different member of
the alliance, the U.S.-led ESV-subproject formed the undisputed centerpiece. 63 At the first
Technical Meeting of the CCMS in Brussels at the end of March 1970, Douglas Toms presented
the idea of a sub-2000 lb/900 kg counterpart ESV program, to be undertaken by interested
member states and contracted out to automobile companies in their respective countries. In
exchange, the U.S. would offer technical help with the planning and implementation of such
programs, and agree to an extensive exchange of information concerning the results of its own
work. 64

On the other side of the Atlantic, initial reactions to this proposal ranged from polite
reluctance to active resistance. Representatives of the European automobile industry, who
met in Frankfurt in late September of 1970 to coordinate their response, found it hard to
imagine that their governments would not resent “attempts to dictate to them on behalf of
the U.S. the way in which they or their industries might handle an e.s.v. project”.65 They
thought the idea fundamentally unsuited to the structure of the European car industry, and
generally seemed at a loss to understand what the U.S. was hoping to gain with such a

60
On the CCMS cf. J. D. Hamblin, Environmentalism for the Atlantic Alliance: NATO's Experiment with the
‘Challenges of Modern Society’, in: Environmental History 15, 2010 (1), p. 54–75; Thorsten Schulz-Walden,
Anfänge globaler Umweltpolitik. Umweltsicherheit in der internationalen Politik (1969 - 1975), München 2013,
p. 79-152; Evanthis Hatzivassiliou, Nixon's Coup. Establishing the NATO Committee on the Challenges of Modern
Society, 1969–70, in: The International History Review 38, 2015 (1), p. 88–108.
61
In the words of Daniel Moynihan, quoted by Schulz-Walden (footnote 52), p. 94.
62
Cf. Schulz-Walden (footnote 52), 84–86; on Moynihan’s role in the “New York safety scene” cf. Vinsel (footnote
9), p. 118–124.
63
The sub-projects were: Identification and Correction of Road Hazards (France), Pedestrian Safety (Belgium),
Motor Vehicle Inspection (FRG), Accident Investigation (Netherlands), Alcohol and Highway Safety (Canada),
Emergency Medical Services (Italy), and Experimental Safety Vehicles (USA); cf. Committee on the Challenges of
Modern Society (CCMS), Road Safety Pilot Study. CCMS Report No. 21, Washington, D.C. 1974.
64
“Presentation for the First Technical Meeting of the U.S. Pilot Study on Road Safety: Experimental Safety
Vehicles (ESV) Program” [March 1970], BAK B 108/23073.
65
“2,000 lb experimental safety vehicle. Note of Meeting on Monday, 21st September 1970 in the offices of VDA,
Frankfurt“, 25.09.1970, BAK B 108/23073.
13
program in the first place. 66 Since the Europeans felt they could not afford to simply ignore
the American proposal, they tried to deflect it with a counter-initiative. The United Kingdom,
which already had a very active national automobile safety research agency, the Road
Research Laboratory (RRL), 67 took the lead in establishing a “European Intergovernmental
Technical Committee on the Development of Experimental Safety Vehicles”, later renamed
European Experimental Vehicles Committee (EEVC). 68 As an alternative to the U.S.-led project,
the Committee proposed to initiate a coordinated research program by a number of European
countries – originally the UK, France and Italy, later joined by West Germany, Sweden and the
Netherlands. 69 Under the guidance of the RRL, the Committee drew up a long list of “safety
components”, from which members could pick those they wished to pursue. 70 This not only
went directly against Washington’s insistence on separate bilateral agreements with each
European partner, but was also fundamentally incompatible with the NHTSA’s “total systems
engineering” approach – much to the frustration of American officials. 71

However, the “unified front” the Europeans tried to present did last long. At the mentioned
manufacturers’ meeting in Frankfurt, there had already been rumors that Volkswagen “might
be willing to ‘go it alone’”. 72 The German company had been approached directly by the
NHTSA as early as March 1970.73 Three weeks later, the company publicly announced its
decision to build an ESV in the 2,000 lb-range, if necessary at its own expense. 74 This forced
the hand of the reluctant West German government. On November 5, 1970, the U.S. DOT and
the West German Ministry of Transport (BMV) concluded a Memorandum of Understanding
concerning the initiation of an ESV program in the Federal Republic. Even though the German
side was still unwilling to commit itself to any promises of financial subsidies, 75 this provided
the necessary diplomatic breakthrough. Two weeks later, Japan signed a similar arrangement.

66
Ibid.
67
On the history of the RRL cf. George Charlesworth, A history of the Transport and Road Research Laboratory,
1933-1983, Aldershot 1987. The RRL seems to have been ardently opposed to the ESV idea from the beginning.
Hatzivassiliou (footnote 52), p. 102, quotes a representative of the Ministry of Transport writing to RRL director
D. J. Lyons in February of 1970: “I understand that you had it in mind to try to kill the project at the start, and I
certainly would not want to dissuade you from this.”
68
Cf. the minutes of the foundational meeting: “Note of conference on the development of ESVs held at the
Ministry of Transport, October 14–15, 1970”, BAK B 108/23073.
69
Although all EEVC members with the exception of Sweden were also part of the European Economic
Community (EEC), and an EEC representative was admitted in 1971, the EEVC was never an official organ of the
EEC/EU. It remains an independent intergovernmental organization to this day; cf. www.eevc.org.
70
Cf. Bonk: “Reisebericht über die erste Sitzung des European Inter-Governmental Technical Committee on the
Development of Experimental Safety Vehicles am 4. und 5. Februar 1971 in Rom“, 15.2.1971, BAK B 108/23074.
71
Cf. e.g. “Remarks of Robert Brenner at European Conference on Experimental Safety Vehicles, London, October
14-15, 1970”, BAK B 108/23073.
72
“2,000 lb experimental safety vehicle” (footnote 57).
73
“Vermerk Anruf Dr. Lotz (VW)”, 18.3.1970, BAK B 108/23073.
74
Günther Brenken, Der Weg zum Sicherheitsautomobil, in: Automobiltechnische Zeitschrift (ATZ) 73, 1971 (5),
p. 1–9.
75
As an internal memo explained, this was the reason why the agreement deliberately spoke of the development
of an ESV “in” (as opposed to “by”) West Germany; “Reisebericht Bonk zur ad-hoc-Sitzung des ETC in Paris am
25.&26.1.1971”, 28.1.1971, BAK 108/23074.
14
Subsequently, the U.S. concluded further bilateral agreements with the United Kingdom and
Italy in May 1971, with France in October 1971, and finally Sweden in March 1972. 76

While negotiations on intergovernmental agreements were still ongoing, the ESV project
began to gather momentum within the auto industry. Toyota, Nissan, Honda and Mercedes-
Benz announced their participation in early 1971, followed by Volvo and Fiat. In addition,
British Leyland, Ford Europe, Opel, Peugeot-Renault, Citroen, Saab, BMW, Porsche and Alfa
Romeo all contributed work on “safety subsystems”. 77 From a rather limited government-led
research project, undertaken by a U.S. regulatory agency in order to prepare the issuing of
national standards, the ESV program had grown into a multinational industry-wide initiative.
By late 1973, the CCMS estimated that the combined total private and governmental
expenditure on the scheme had already surpassed $150 million. 78

What motivated profit-oriented automobile corporations to participate in the time-

consuming and resource-intensive NHTSA scheme – many of them, remarkably, largely or
even completely at their own expense?79 Firstly, the United States was still the most important
market for automobiles in the world by a very large margin, economically crucial not only for
North American automakers, but also for a number of their European competitors. This gave
U.S. officials a substantial amount of leverage, which they applied liberally to pressure
companies into participation. According to a major North American newspaper, the U.S.
government “made it clear in the secret negotiations that foreign car makers must cooperate
in this ‘Experimental Safety Vehicle’ program if they expect to continue importing cars into
the United States”.80 Volkswagen’s decision in particular was at least partly a consequence of
the company’s unusually high degree of dependence on the U.S. market, where it sold up to
40% of its signature model, the Beetle, in the late 1960s.81

Secondly, as the ESV project started to create an increasing amount of public attention and
not only trade journals, but also national news media began to take notice of the initiative,
the possible public relations effects of (non-)participation became an important
consideration. 82 For some companies – such as VW or GM, both of which had been particularly
singled out for criticism by Ralph Nader – the project offered an opportunity to repair a

76
Even more protracted were the negotiations concerning the annexes to the agreements pertaining to the
envisioned information exchange; cf. the extensive diplomatic communication on this topic in BAK B 108/23073
to 108/23075. This was a very sensitive issue, since it not only risked divulging proprietary information belonging
to participating corporations, but also threatened to run afoul of U.S. anti-trust legislation.
77
Cf. CCMS (footnote 44), p. 5–6.
78
Ibid., p. 5.
79
Apart from the non-automobile U.S. participants AMF and FHC, Toyota and Nissan were the only companies
receiving a substantial amount of public funding for the construction of “full” ESVs in the framework of the
project. The United Kingdom, France, Italy and Sweden all made public funds available for research and
development work on subsystems.
80
“U.S. and NATO Seeking Auto Safety”, Washington Post, 10.8.1970.
81
On VW’s precarious situation in the late 1960s cf. Bernhard Rieger, The People's Car. A Global History of the
Volkswagen Beetle 2013, p. 233–255.
82
Major news items in the general press included e.g. a 10-page feature in Business Week (footnote 46), and a
title story by the important West German weekly Der Spiegel, Sicherheitsautos. Für Tage ohne Tote, in:
DER SPIEGEL, 16.8.1971, p. 86–104.
15
somewhat battered reputation on safety.83 Others – such as Mercedes-Benz or Volvo – who
had made technically advanced safety features a central selling point of their relatively
expensive cars, felt that they had to live up to their carefully nurtured brand image, if they did
not want to be overshadowed by their competitors. Not building an ESV would “at least
outwardly make the advantage we have in regard to safety appear smaller and smaller”,
Mercedes’ head of development Karl Wilfert argued in a letter to his board members.84

Thirdly, participation in the ESV project also constituted an attempt to regain some influence
over the safety debate, and even the future form and shape of automobiles more generally.
Given the importance of the U.S. as a global automotive trendsetter, Business Week reported,
the sudden “regulatory barrage” emanating from the NHTSA was “bound to reshape auto
design, production and sales more radically than anything that has happened since Henry Ford
started up his first assembly line”. 85 While U.S. auto executives complained loudly that they
had “practically no say” in drafting the new regulation, the problem was even more acute for
non-American companies. Not without reason, they feared that new standards based on
American-sized cars might put their own, much smaller models at a distinct disadvantage.
Some European politicians even started to wonder publicly whether the whole ESV project
was not a “veiled form of protectionism”, aiming to push European import cars out of the U.S.
market. 86 Under these conditions, joining the ESV program in some form could seem like the
only opportunity for non-American automakers and their respective governments to make
their case to the NHTSA. As early as 1972, a perceptive German motor journalist concluded
that “the aim of this almost pan-European safety demonstration” was not so much “image-
cultivation” with consumers, but rather “the attempt to stir up the discussion on these issues
[i.e. safety standards, S.E.] again by introducing new arguments” – in the hope of convincing
the Americans regulators to “return to reason”. 87 An internal memo by a West German official
expressed a very similar idea somewhat more dryly: “Only by consequent cooperation in the
development of experimental safety vehicles can evidence be provided that smaller vehicles
can also fulfill the requirements made [by the NHTSA] particularly in respect to occupant
protection, or that said requirements are objectively unobtainable for technical-physical
reasons”.88

83
GM’s Chevrolet Corvair was the most prominent target of Nader’s “Unsafe at Any Speed”; cf. Nader (footnote
39). Nader had also publicly denounced the VW Beetle – designed in the 1930s and hence only partially
compatible with more recent safety standards – as the “most dangerous car on American roads”; “Der
Volkswagen tanzt Bossa Nova”, in: DER SPIEGEL, 4.10.1971, p. 46–60.
84
Quoted by Weishaupt (footnote 11), p. 318; cf. Stieniczka (footnote 11), p. 263–265.
85
The Crash Program (footnote 46), p. 78. In Europe, this sentiment was echoed by the German weekly Der
Spiegel, which claimed that Douglas Toms “now wields more power to determine the appearance, construction
and performance of new generations of automobiles than even the mightiest auto bosses, such as GM president
Edward Cole or Fiat boss Giovanni Agnelli, ever had”; Sicherheitsautos (footnote 74), p. 87.
86
Inquiry 407/70 by MEP Glinne (Belgium), 21.12.1970, BAK B 198/23074.
87
Clauspeter Becker, Zurück zur Vernunft, in: Auto Motor und Sport, 1972 (11), p. 46. Unless otherwise indicated,
all translations by the author.
88
Ref StV 1: „Betr.: Zuwendungen des Bundes an die Volkswagenwerk AG für den Bau und die Erprobung von 40
Prototypen eines Experimentier-Sicherheitsfahrzeugs“, 29.2.1972, BAK B 108/37505.
16
Arguing with Evidence Objects: Experimental Prototypes and the Automobile Safety Debate
The ESV project therefore soon became not only a media event and a technical research and
development effort, but also an important forum for the transnational (re-) negotiation of the
predominant automobile safety paradigm. The international ESV conferences in particular,
attended by both government and industry representatives, were notable for the direct – and
in parts quite confrontational – exchange of opinions and arguments between the NHTSA,
European governments, and automobile companies. 89 In addition to technical discussions
about safety systems and the performance of the different ESVs, the topics on the agenda also
included the progress and general direction of the program, its significance for the future of
automobile safety, and its translation into future (national and international) standards and
regulations.

The first two ESV conferences, held in January and October 1971, were dedicated largely to
the discussion of U.S. specifications and their possible adaption to smaller vehicle classes. As
NHTSA officials emphasized, their requirements unequivocally put crashworthiness at the
center of all concerns. While a detailed list of “accident avoidance” criteria – from acceleration
and handling to visibility and lighting – was part of the requirements, their purpose was mainly
to ensure that the resulting prototypes would not fall behind existing production models in
this regard. Some of the requirements regarding passive safety, on the other hand, were “very
severe”, as even NHTSA officials themselves admitted. 90 Prototypes notably had to be able to
withstand frontal crashes at 50 mph into a rigid barrier and a solid pole, 50 mph rear impacts,
30 mph side impacts, and 70 mph rollovers, without allowing intrusions of more than three
inches into the passenger compartment. At the same time, maximum body deceleration had
to be kept below 60 G for all five passengers. 91 The already challenging assignment was made
even more difficult by the additional specification that only “passive restraints, which require
no action by the vehicle occupants” were to be used – thus excluding conventional seat belts,
which passengers had to actively fasten themselves. 92 This last requirement was a
consequence of the NHTSA’s 1969 decision to mandate the introduction of air bags or similar
“passive restraints” for all vehicles by 1973, in spite of ferocious resistance from the auto
industry. 93 Perhaps more than any other specification, it exemplified the agency’s intention of
replacing human responsibility for safety as much as possible with technological fixes, in order
to avoid social and political complications. Although the NHTSA freely admitted that belts
might be an effective solution from an engineering standpoint, they argued that it would be

89
In the build-up to the second conference in October 1971, European automakers expected discussions with
the NHTSA to become so heated that they vehemently opposed the planned admission of the press to all sections
of conference; StV1: “Vermerk über Gespräch mit VW am 21.9.”, 23.9.1971, BAK B 108/37504.
90
Second Specification Discussion. Crashworthiness, in: Report on the First International Technical Conference
(footnote 43), p. 123.
91
100g were allowed for peaks of less than 3 milliseconds; cf. Experimental Safety Vehicles (footnote 49), p. 54–
55; “Enclosure 5: Statement of Work”, BAK B 108/23073. Other specifications included bumpers that would allow
impacts of up to 10 mph without damage to the body of the vehicle.
92
Schlechter (footnote 47), p. 34.
93
Cf. Jameson M. Wetmore, Implementing Restraint. Automobile Safety and the U.S. Debate over Technological
and Social Fixes, in: Jim Conley and Arlene Tigar McLaren (eds.), Car troubles. Critical studies of automobility and
auto-mobility, Farnham 2009, p. 111–126; Wetmore (footnote 10).
17
impossible to force the American public to use them. 94 “It is a lot easier to deal with the world
auto industry, where you’re dealing with dozen-and-a-half manufacturers, than it is to deal
with millions of people”, Douglas Toms was quoted in Business Week. 95

This approach presented European and Japanese participants with substantial technical
difficulties, given the significantly smaller margins in terms of space, weight and cost available
for smaller vehicles. Perhaps even more importantly, it also went against their basic
understanding of automobile safety and their conception of responsibility on the road. The
West German automobile industry trade association (VDA), for instance, who had received
the task of “translating” the U.S. specifications for their respective national 2,000 lb ESV
programs, flat out refused to adopt the “passive restraints only” rule, which a Volkswagen
representative called a “ridiculous” requirement. 96 Trying to explain this decision, the leading
West German automotive engineering journal ATZ observed that “[t]he main differences
concern the responsibility of the driver and the occupants. In Europe, we are of the opinion
that the driver is responsible for his vehicle and his conduct in traffic, whereas in the USA it is
assumed that, even in case of a grossly negligent attitude on the part of the driver, complete
protection for him and his occupants must be assured.” 97 Concerning the respective attitudes
to passive safety, a German motor magazine agreed, “Europe and America are still separated
by much more than an ocean”. 98

The NHTSA hoped to provide material evidence for the feasibility of their approach with the
presentation of the four American ESVs. They were first introduced to a wider public at the
third ESV conference, held in the early summer of 1972 in the framework of “transpo ‘72”, a
special transport-related trade show in Washington, D.C. Although the show offered a
plethora of spectacular attractions – such as levitating trains, driverless mass transit systems
and the “world’s largest air show” – the prototypes seemed to have created quite a stir:
According to Douglas Toms, the pavilion displaying them was the “most heavily attended of
all the exhibits”. 99 While GM and Ford’s vehicles looked outwardly more conventional than
the AMF and FHC, which sported some obviously “experimental” features such as rooftop
periscopes, all four U.S.-ESVs shared a number of common characteristics. In order to meet
the ambitious deceleration and intrusion requirements, they were designed around
reinforced frames and heavy-duty bumper systems, making them unusually large and heavy

94
An additional factor making this approach even more attractive was the structure of the U.S. government:
While safety standards could be set centrally, traffic legislation and enforcement were largely devolved to local
authorities, presenting federal agencies such as the NHTSA with an “infinite morass” of legal problems; cf. Second
Specification Discussion (footnote 82), p. 122.
95
The Crash Program (footnote 46), p.78–83.
96
Second Specification Discussion (footnote 82), p. 127–128.
97
Wolfgang Rosenau and Ulrich Seiffert, Zukünftige Gesetzesvorschriften bezogen auf die ESV-Spezifikationen,
in: Automobiltechnische Zeitschrift (ATZ) 74, 1971 (4), p. 166.
98
Becker (footnote 79), p. 47.
99
Report on the Third International Technical Conference (footnote 45), p. 1/9.
18
even by American standards.100 In spite of using a number of uncommon materials and
techniques partly borrowed from aerospace engineering, such as high performance steels,
fiberglass, and aluminum honeycomb structures, none of the four managed to stay within the
maximal weight allowance, with the AMF weighing as much as 5,800 lb (2,600 kg). 101 In
addition to airbags, they relied on extensive interior padding in order to keep the occupants
in place during crashes without violating the passive-restraints-only specification – inducing
the British government representative to quip in his conference speech that they looked “like
a cross between a tank and a padded cell”. 102

By contrast, the non-American entries presented at the same occasion covered a much wider
range of shapes, sizes and approaches. 103 Volkswagen and Toyota came closest to the NHTSA’s
original idea of the international ESV program, working to revised versions of the American
specifications issued centrally by their respective national trade associations. 104 Both firms
invested substantial amounts of financial and personal resources into completely new vehicle
designs in the 2,000 lb/900 kg class.105 The results largely delivered the desired safety
performance, albeit with some compromises in design: While VW replaced the airbags with a
“passive belt” system, Toyota settled for a two-seater to have larger crumple zones.106
Mercedes and Volvo presented substantially larger cars, opting not to work according to
“European” specifications, but to try to fulfill the original NHTSA requirements for a self-
defined weight class in-between the “American” and “Euro-Japanese” models. At over 5
meters in length and over 2,100 kg (4,630 lb) in weight, the Mercedes in particular almost
matched the American models. 107 At the same time, the two companies strongly emphasized
the close link to the existing safety technology of their production models, which they argued

100
While AMF and FHC used large hydraulic bumper systems for high-speed impacts, GM and Ford opted to work
with more conventional deformable-metal structures for impact energy absorption, using smaller hydraulic
cylinders only to fulfill the “10 mph no damage” requirement.
101
Cf. NHTSA (ed.), Report on the Second Technical Conference on Experimental Safety Vehicles. Sindelfingen,
Germany. October 27-29, 1971, Washington 1971, p. 2/3–2/63; Experimental Safety Vehicles. U.S. Designs and
Innovations, in: Automotive Engineering 80, 1972 (9), p. 30–37.
102
Report on the Third International Technical Conference (footnote 45), p. 1/15.
103
For an overview cf. Experimental Safety Vehicles. Where Do We Go From Here?, in: Automotive Engineering
80, 1972 (8), p. 19–27; H. Hontschik, Forschung am Sicherheitsauto. 1. Bericht über die 3. Internationale
Technische Konferenz über Experimentier-Sicherheitsfahrzeuge, in: ATZ 74, 1972 (9), p. 365–372.
104
While generally following the American requirements, these adaptions were somewhat more generous in
particular points, such as the intrusion specifications or the possibility to use active restraints, cf. Brenken
(footnote 66), p. 1–9.; VDA: “Technische Anforderungen für Experimentier-Sicherheits-Personenkraftwagen”,
21.12.1970, BAK B 106/23074; Report on the Second Technical Conference (footnote 93), p. 2/87–2/88.
105
According to Stieniczka (footnote 11), p. 311, Volkswagen invested 100 million DM (approx. $39 million) in
the ESVW I; the costs of the Toyota ESV project were estimated at 2 billion yen ($7 million); cf. Noriyoshi Uno,
Application of Research Results to Standards. Future Activities Japan, in: NHTSA (ed.), Report on the Fifth
International Technical Conference on Experimental Safety Vehicles. London, England, Washington, D.C. 1974,
p. 995–996. The Nissan ESV was equally built to centrally issued specifications in a similar weight class (2,500
lb/1250 kg), but was based on a production model and specifically emphasized closeness to production models;
cf. Yoshio Serizawa, The Nissan Motor Company, in: Report on the Second Technical Conference (footnote 92),
p. 2/89-2/93.
106
Cf. Ernst Fiala, The Volkswagen ESV, in: Report on the Third International Technical Conference (footnote 45),
p. 2/200-2/215; Jiro Kawano, Outline of the First Prototype and Experimental Study, in: Ibid., p. 2/240–2/249.
107
For an overview over the different Mercedes ESVs cf. Weishaupt (footnote 11), p. 345–350.
19
were already almost able to reach the U.S. design goals, or could be made to do so with
relatively minimal effort. 108 Fiat and Honda covered the low end of the weight spectrum,
proposing to construct ESVs in the 1,500 lb/650 kg range. Given the severe problems expected
due to the very limited amount of weight and (deformable) space available, their contributions
did not promise to reach any clearly defined crashworthiness goals, but rather aimed to
explore what level of passive safety could reasonably be reached within the size, weight and
price constraints of this class. 109 British and French participants did not strictly speaking
produce ESVs at all, but stuck with the “Experimental Safety Subsystems” (ESSS) approach
developed for the EEVC, arguing that this would provide better chances for quick
incorporation into regular production. To prove the point, British Leyland showed two
production models equipped with a number of new safety systems for presentation purposes.
With features such as radar distance warning, self-levelling headlights or alcohol screening
testers, the focus lay mainly on active safety, which was presented as “reflect[ing] safety ideas
consistent with European thinking and future legislation on vehicle and passenger safety”.110

From the start, a considerable part of the work undertaken by Europeans and Japanese in the
context of the ESV program was in fact targeted mainly at disproving the practicability of the
NHTSA’s approach and destabilizing the potential evidence provided by the American ESVs.
While almost all participants initially pronounced themselves highly optimistic about the
technical feasibility of the requirements, they went to considerable lengths to cast doubt on
their economic viability. Fiat, Honda, Citroen and Opel all began their respective ESV activities
with detailed studies on the substantial increase in size, weight and cost even a partial
fulfillment of the specifications would mean for their small production models. 111 Probably
the most meticulous calculations were presented by Volkswagen, who attempted to quantify
the cost-benefit ratio of different forms of active and passive restraints, and even of different
numerical values for particular requirements. They came to the conclusion that some of the
impact tests required, such as the 50 mph rear impact, contributed unduly to cost increases
while providing relatively little safety benefit, and that all forms of seat belts were much more

108
Cf. Hans Scherenberg, The Development of the ESV As Seen By Daimler-Benz, in: NHTSA (ed.), Report on the
Second Technical Conference (footnote 93), p. 2/67-2/72; Rolf Mellde, The Volvo ESV, in: Report on the Third
International Technical Conference (footnote 45), p. 2/129-2/133. While Mercedes worked with a modified
production model, Volvo presented what it claimed was the result of its own independent ESV program started
in 1969.
109
Cf. G. Puleo, Consequences on the Design of an Economy Car, in: Report on the Second Technical Conference
(footnote 93), p. 2/111-2/113; Hideo Sugiura, Occupant Protection of 1,500 lb ESV, in: Report on the Third
International Technical Conference (footnote 45), p. 2/250-2/256. In addition to the 1,500 lb vehicle, Fiat also
conducted similar experiments on 2,000 lb and 2,500 lb models, making it the only participant to provide ESVs
in three different weight classes; cf. Oscar Montabone, The Fiat Technical Presentation, in: Report on the Third
International Technical Conference (footnote 45), p. 2/157-2/184.
110
Report on the Third International Technical Conference (footnote 45), p. S/10.
111
Cf. Puleo (footnote 101), 2/111-2/113.; Sugiura (footnote 101), 2/250-2/256.; Maurice Clavel, Why Citroen
Chose 1,500 lb. Vehicle for its Studies and Experiments, in: Report on the Second Technical Conference (footnote
93), p. 2/135-2/142; Karl Brumm, The Opel Conception of an ESV in the Low-Weight Class, in: ibid., p. 2/72-2/77.
20
cost-efficient than airbags. 112 Consequently, European industry representatives, such as
Mercedes board member Hans Scherenberg, started publicly demanding laws mandating the
use of seat belts – which were still not even part of the standard equipment in many of the
company’s models at this point. 113

Secondly, the NHTSA’s critics questioned whether the new road traffic system likely to result
from the agency’s approach was desirable in the first place. The crashworthiness requirement
of the American specifications, a Citroen engineer warned darkly, would “unequivocally doom,
to the point of eliminating, the small, low-range European car”. 114 In Europe, with its narrow
city streets and lower average wages, this would be socio-economically disastrous, potentially
undoing the gains of mass-motorization and making the automobile once more “the
prerogative of the few well-off classes”. 115 In accordance with growing general awareness of
the issue, the argument of fuel consumption and environmental degradation associated with
overly large cars became an increasingly important argument as well. “Exhaust emission, as a
major social problem, is equally as important as the safety problem now”, opined a Honda
representative in 1973. 116 Even worse, “tank-like” vehicles such as the American ESVs might
even lead to a decrease in safety on (non-American) roads: When colliding with small cars or
even pedestrians, their weight and bulk would make them particularly dangerous to the other
parties in the crash – a problem for which the French delegation introduced the subsequently
much-debated concept of “vehicle aggressiveness”. 117

Thirdly, Europeans and Japanese, but also American corporations argued that the
specifications and tests prescribed by the NHTSA were not representative of actual conditions
on the roads. General Motors, for instance, prominently described their ESV activities not as
an attempt to increase traffic safety, but merely as “a study in meeting the Department of
Transportation requirements”.118 The ESV program, in other words, constituted an extreme
example of the phenomenon Lee Vinsel has described as “designing to the test”. 119 On

112
Ernst Fiala, General View About Progress and Problems Concerning ESV, in: Report on the Second Technical
Conference (footnote 93), p. 2/65-2/67; Hermann Appel, Benefit/Cost Analysis for Evaluation of ESV Impact
Tests. Proposal for Reduced Rear End Impact Speeds, in: ibid., p. 3/5-3/9.
113
Hans Scherenberg, Progress of the ESV Development at Daimler-Benz, in: NHTSA (ed.), Report on the Fourth
International Technical Conference on Experimental Vehicle Safety. Kyoto, Japan, Washington, D.C. 1973, p. 83–
94, p. 84. On the German discussion on seat belts cf. Bergmann (footnote 27).
114
Clavel (footnote 103), p. 2/135
115
Ibid.
116
Kiyoshi Kawashima, The Present Development Status of the Honda ESV, in: NHTSA (ed.), Report on the Fourth
International Technical Conference (footnote 105), p. 73. Environmental aspects and the development of a “low
pollution power system” had been part the ESV program since its inception. The complex history of its actual
implementation goes beyond the limits of this contribution. For a first orientation, cf. Schulz-Walden (footnote
52), p. 117–119.
117
M. Chillon, The Importance of Vehicle Aggressiveness in the Case of a Transversal Impact, in: Report on the
First International Technical Conference (footnote 43), p. 81–84; Claude Berlioz, Comparison of the
Aggressiveness of Different Vehicles and the Safety They Afford, in: Report on the Third International Technical
Conference (footnote 45), p. 2/4-2/9; Phillippe Ventre, Homogenous Safety Amid Heterogenous Car Population?,
in: Ibid., p. 2/39-2/57.
118
William Larsen and John Rosenkrands, The General Motors Corporation. ESV Development Report, in: Report
on the Second Technical Conference (footnote 93), p. 2/27–2/48, p. 2/48.
119
Vinsel (footnote 10).
21
whether the outcome related at all to “normal” traffic, GM claimed to be agnostic: “Our car is
designed for very specific crash test situations. The relationships of our test data to highway
crashes is unknown.” 120 European participants in particular expended a lot of effort to prove
that the specified test conditions did not allow wider conclusions about actual safety
performances.121 There was little sense in forcing companies to construct prototypes that
could withstand frontal collisions into a fixed barrier at 80 km/h, they reasoned, when the
overwhelming majority of crashes happened between two cars and at an offset angle.122 Using
accident statistics, the German Association of Automobile Insurers (HUK) calculated that 90
percent of injuries occurred at “equivalent test speeds” of less than 60 km/h, thus making
higher speed requirements unnecessary.123 Another major source of uncertainty were the
unstandardized and still rather primitive dummies used in the tests, whose fidelity to the
human body was at best questionable – especially since tolerance levels of the human body
were still only very imperfectly known. 124 Before hastily imposing near-impossible
crashworthiness standards on manufacturers, car companies argued that more cost-benefit
analyses, more accident data and more biomedical research were sorely needed. 125

In response, the NHTSA acknowledged the experimental nature of large parts of their
requirements, promised to consider the presented evidence and signaled its willingness to
compromise on certain numerical values and specifications – without however abandoning
their fundamental dedication to ambitious crashworthiness targets, which they considered
the cornerstone of the project. 126 When it became clear during the third ESV conference that
the fundamental “reorientation” the European and Japanese participants had hoped to
extract from the Americans was not forthcoming, the discussion turned increasingly
confrontational. 127 At the same time, the first results obtained from crash tests made clear
that none of the non-U.S. vehicles would manage to fulfill the NHTSA requirements completely
– not even the large and heavy Mercedes. 128 For symbolic as much as technical reasons, the

120
Larsen and Rosenkrands (footnote 110), p. 2/48.
121
They were attacking, in other words, the Achilles heel of all testing, namely the “similarity relationship”
between the testing situation and the actual working of the technology; cf. Trevor Pinch, “Testing - One, Two,
Three… Testing!”: Toward a Sociology of Testing, in: Science, Technology, & Human Values 18, 1993 (1), p. 28f.
122
Claude Berlioz, Distribution and Gravity of Collisions as a Function of the Damaged Part of the Vehicle and the
Obstacle Hit, in: Report on the Second Technical Conference (footnote 93), p. 2/129-2/135; Claude Prost-Dame,
Critical Review of ESV Program Technical Trends and Future Resulting Legislation, in: Report on the Third
International Technical Conference (footnote 45), p. 2/63–2/64.
123
Max Danner, New Investigations of HUK Accident Research, in: Report on the Fourth International Technical
Conference (footnote 105), p. 147–153; Max Danner and K. Langwieder, The Frequency of Corresponding Vehicle
Damage in Crash Tests and Actual Accidents, in: Report on the Fifth International Technical Conference (footnote
97), p. 421–426.
124
Willi Reidelbach, The Shortcomings of Anthropomorphic Test Devices, in: Report on the Second Technical
Conference (footnote 92), p. 3/57-3/59; M. A. Macauley, The Dynamics of Dummies, in: Ibid., p. 3/59-3/61.
125
Cf. the proposals for a “new orientation of the ESV program” by Hans Scherenberg, ESV Development at the
Daimler-Benz A.G., in: Report on the Third International Technical Conference (footnote 45), p. 2/191f.
126
Cf. Albert Schlechter, Summation and Concluding Remarks, in: Report on the Second Technical Conference
(footnote 92), p. 5/3-5/4; Albert Schlechter, United States ESV Program Status, in: Report on the Third
International Technical Conference (footnote 45), p. 2/281–2/289.
127
This expectation was probably most clearly formulated by Mercedes; cf. Scherenberg (footnote 105), p. 83.
128
Cf. Scherenberg (footnote 117), p. 2/186–2/193.
22
50 mph/80 km/h frontal impact specification became a particular bone of contention. If the
Americans truly meant to stick to this requirement, Fiat boss Montanari was quoted in the
press, “we have nothing more to talk to each other about”. 129

Having failed to modify the U.S. specifications in the manner they had wanted, most of the
non-American participants subsequently simply gave up on their ambitions to fulfil them, and
reoriented themselves towards self-defined design goals they considered more “realistic”.
This shift clearly materialized itself in a “second generation” of ESVs, shown at the fourth and
fifth ESV conferences in 1973 and 1974. The three West German prototypes presented at the
latter occasion, for instance – the Mercedes ESF 22, the Volkswagen ESVW II and the Opel OSV
– were all built to relatively lenient 65 km/h frontal-collision requirements. Even more
importantly, all were modified versions of the respective companies’ newest production
models, the S-Class, the Golf/Rabbit, and the Kadett. 130 From the highly experimental and
disruptive “idea cars” of the early program period, ESVs had transformed into more
conventional industry prototypes, blending almost seamlessly into “normal” market-oriented
research and development efforts on new production models. 131

At the same time, however, the fifth ESV conference, which marked the official ending of the
original “first phase” of the ESV project, can also be understood to mark the amalgamation of
the “European” and “American” approaches to automobile safety. Taking place in London, the
center of the initial European “resistance” to the U.S. approach, the meeting was notable for
the formal integration of the EEVC into the ESV framework. 132 Accordingly, the conference
allocated the majority of its technical sections to topics such as “accident analysis and data
collection”, “human tolerance levels and use of dummies”, “effectiveness of safety measures”
and “interactions of driver and vehicle”, cost-benefit-analysis, and “application to production
cars.” 133 Echoing earlier criticism from the auto industry, NHTSA officials concluded in their
closing report that “the existing ESV specifications, taken as a whole, result in designs of
questionable near-term practicality” – mainly because all of the tested designs suffered from
“weight increases on the order of 20-30 percent”.134

129
“Noch keine Chance für das Sicherheitsauto”, Welt am Sonntag, 7.4.1973, in BAK B 108/37506. Cf. also
Montanari’s almost equally sharp critique during the conference discussion session: Report on the Third
International Technical Conference (footnote 45), p. 3/11.
130
W. Rixmann, Deutsche Experimentier-Sicherheitsfahrzeuge (ESV) 1974. 1. Bericht von der 5. Internationalen
Technischen Konferenz über Experimentier-Sicherheitsfahrzeuge, Juni 1974 in London, in: ATZ 76, 1974 (10), p.
333–339.
131
The term “idea car“ was originally coined by GM for their ESV; cf. J. W. Rosenkrands, Experimental Safety
Vehicle - Phase Two. Designed and Developed by General Motors, in: Report on the Third International Technical
Conference (footnote 45), p. 2/257; however, it soon caught on as a description of the whole program; cf. DOT
News: Remarks by U.S. Secretary of Transportation John A. Volpe at Crash Test of Experimental Safety Vehicle,
Phoenix, Arizona, April 18, 1972, BAK B108/37505.
132
The Future for Car Safety in Europe. A Report of the EEVC, in: NHTSA (ed.), Report on the Fifth International
Technical Conference (footnote 97), p. 24–54.
133
Cf. Report on the Fifth International Technical Conference (footnote 97).
134
Vincent Esposito, Status Report on Experimental Safety Vehicle Development Programs. United States, in:
Report on the Fifth International Technical Conference (footnote 97), p. 12.
23
In light of the first oil crisis and growing environmental concerns, this trade-off seemed
increasingly untenable. Already in early 1973, the NHTSA had therefore announced a follow-
up initiative, the so-called Research Safety Vehicle (RSV) project.135 Aimed at producing cars
in the “intermediate” 3,000 lb/1350 kg class, the program abandoned the almost exclusive
focus on crashworthiness and primarily sought “to address the problems of minimizing fuel
consumption, controlling urban congestion and pollution, and maximizing resource
conservation as well as safety.” 136 Asked in a high-profile interview by Der Spiegel in late 1973
whether his company would consider participating in the new program, Daimler Head of
Development Scherenberg demurred, confessing to a “certain weariness regarding the
question whether to build even more experimental cars”. Instead, he expressed his hope that
other factors such as human behavior, the roads and traffic rules would soon return to the
fore: “One cannot always hold the automobile responsible for lack of safety.” 137

Conclusion: ESVs and the Construction of New Automobile Safety Practices

Given the gradual abandonment of its original aims, especially after the turning point in 1972,
it is perhaps not surprising that many participants ultimately considered the ESV program at
best a dubious investment, if not a costly failure. While the enormous amount of financial and
personal resources spent had led to some “new experiences and knowledge”, according to
Mercedes-Benz, the main insight derived from the program was that the kind of passive safety
performances demanded were “technically imaginable, but do not lead to a result in step with
actual practice.” 138 Indeed, the project’s most immediate outcome had arguably been to
disprove the practicability of a “pure” crashworthiness approach. Rather than universally
spreading the new paradigm, the ESV program could therefore be said to have been involved
in limiting its full implementation, and perhaps even prefigured the partial return of the “old-
fashioned emphasis on driver responsibility” that Peter Norton has observed for the 1980s.139

However, such an interpretation would overlook the significant contribution of the ESV
program to the changes in the theory and practice of automobile safety in the 1970s,
especially outside the United States. Firstly, the scheme was clearly successful in its attempt
to “stimulat[e] public awareness” for the benefits of the new approach. Although
crashworthiness was not a new concept in the participating states by the late 1960s, public
exhibitions of prototypes, and in particular the extensive reporting on various aspects of the

135
Cf. Gene Manella, Remarks on 3,000 Pound ESV Specification, in: Report on the Fourth International Technical
Conference (footnote 105), p. 541.
136
Esposito (footnote 126), p. 13. On the finished RSVs, presented at the seventh ESV conference in 1979, cf.
Donald Struble, Status Report on Minicar's Research Safety Vehicle, in: NHTSA (ed.), Report on the Seventh
International Technical Conference on Experimental Safety Vehicles. Paris, France, Washington, D.C. 1979, p. 63–
75; G. J. Fabian and G. Frig, Status Report on Calspan/Chrysler Research Safety Vehicle, in: ibid., p. 104–131.
Volkswagen, the only non-American participant in the preliminary phase of the program, was not awarded a
contract for the second phase.
137
"Was ist ein Mensch wert?" Daimler-Benz Entwicklungschef Scherenberg über Sicherheitsautos, in:
DER SPIEGEL, 10.9.1973, p. 68–84.
138
Presse Information ESF-Konferenz in London, 1974, quoted Weishaupt (footnote 11), p. 352.
139
Norton (footnote 7), p. 329.
24
program in the (specialized and general) media greatly increased the visibility of the issue –
thus putting public pressure on the automobile industry. 140 Secondly, the ESV program also
had a direct impact on research and development. It provided significant boost to the
development of a number of new technical solutions and systems, such as the airbag, and the
expedited improvement of others, such as crumple zones or seat belts. Arguably even more
important was, thirdly, the project’s effect on the institutionalization and standardization of
safety research – for instance through the establishment of international organizations and
networks such as the ESV conferences or the EEVC, both of which are still highly relevant for
the field today. In addition, many of the participating automobile companies enlarged,
updated, or even newly created testing facilities and safety engineering departments,
providing the basis for making crashworthiness an important consideration in standard
research and development practice. All of this contributed to the substantially improved
safety record of the new generation of automobiles hitting the global market after 1974 –
even if some features demonstrated in experimental prototypes were not widely available in
production models before the late 1980s.

While it would be simplistic to consider the global changes in automobile safety practices as
nothing more than an extension of the American development abroad, the United States did
play an important role as a pacesetter in this development. To the disappointment of safety
advocates, U.S. regulations on crashworthiness might never have been quite as rigorous and
effective as those on air pollution, as Lee Vinsel has pointed out.141 In comparison to its
European counterparts, however, such as the notoriously pusillanimous West German BMV,
the NHTSA’s much more proactive and confrontational approach stands out as notably
impactful. From this perspective, the ESV program provides a remarkable example of how, in
the age of economic globalization, one national regulatory agency – armed with sufficient
market might, favorable public opinion, and a sense of mission – could force a highly powerful
and prestigious industry to change their ways. By stressing the importance of NHTSA as the
main driving force, I do not mean imply that the ESV program simply constituted an imposition
of a U.S. model of automobile safety on America’s reluctant allies. On the contrary, one of the
main motivations for international governments and non-state actors to join was precisely the
potential opportunity to influence the NHTSA’s position – and thus future national U.S.
standards. In the resulting transnational negotiation process, the differences between
engineering cultures and mobility systems in North American and Europe (respectively Japan)
quickly became apparent – and were at times arguably even over-emphasized in the attempt
to find a politically palatable argument to avoid regulations. The substantial divergences in the
approach to crashworthiness between the NHTSA and most of the other ESV participants –
notably including the U.S. auto industry – did not vanish over the course of the program.
However, the project forced many participants to engage much more closely with the idea
and practice of crashworthiness than they had done before. In return, the detailed
interventions and arguments brought forth were not without resonance on the NHTSA’s policy

140
For a summary of the project’s public relations achievements cf. CCMS Report No. 23 (footnote 49), p. 7.
141
Vinsel, Moving Violations (footnote 9).
25
– even if the agency’s abandonment of its maximalist position on crashworthiness cannot be
explained by European protests alone.

This transnational (re-) negotiation of automobile safety in the framework of the ESV program
was made possible by the Experimental Vehicles as material artefacts. Not only were their
technical specifications and design goals, their performance characteristics, and their physical
limitations the object of heated arguments. It was also through the ESVs that participants
argued about what should count as a “safe” car, who was responsible for safety, and how and
to what degree it should be enforced. The experimental prototypes were thus evidence
objects in a twofold sense: On the one hand, they served as laboratory objects to produce
data, try out technical assumptions, and generate the detailed knowledge necessary to set
standards and issue regulations. On the other hand, the experimental vehicles were
interventions in a normative argument between competing visions of the (automobile) future,
specially constructed to illustrate certain ideas and points of view and imbued with political
agendas, economic interests and cultural significance. In both respects, the materiality of the
ESVs was crucial to their efficacy, making new arguments and ideas much harder to dismiss
out of hand and forcing opponents to engage with them in detail. The ESVs therefore
functioned as a sort of boundary objects, bringing together a variety of entities and institutions
with different interests and (national) backgrounds in a transnational, techno-political
discourse.142 However, while ESV program certainly contributed to the internationalization
and professionalization of the discussion about the future of automobile safety, this also came
at a price: Private consumers and safety activists, who had pioneered of the use of
experimental prototypes for the promotion of crashworthiness in the 1950s, were virtually
absent from ESV conferences. While the future of automobile safety certainly seems to be too
big an issue to be decided by one nation alone, a viable way of including consumer interests
still needs to be found.

142
Susan Leigh Star and James R. Griesemer, Institutional Ecology, 'Translations' and Boundary Objects. Amateurs
and Professionals in Berkeley's Museum of Vertebrate Zoology, 1907-39, in: Social Studies of Science 19, 1989
(3), p. 387–420.
26